Search

Evaluation of left ventricular (LV) function in critical care patients is useful for guidance of therapy and early detection of LV dysfunction, but the tools currently available are too time-consuming. To resolve this issue, we previously proposed a method for the continuous and automatic quantification of global LV function in critical care patients based on the detection and tracking of anatomical landmarks on transesophageal heart ultrasound. In the present study, our aim was to improve the performance of mitral annulus detection in transesophageal echocardiography (TEE). We investigated several state-of-the-art networks for both the detection and tracking of the mitral annulus in TEE. We integrated the networks into a pipeline for automatic assessment of LV function through estimation of the mitral annular plane systolic excursion (MAPSE), called autoMAPSE. TEE recordings from a total of 245 patients were collected from St. Olav's University Hospital and used to train and test the respective networks. We evaluated the agreement between autoMAPSE estimates and manual references annotated by expert echocardiographers in 30 Echolab patients and 50 critical care patients. Furthermore, we proposed a prototype of autoMAPSE for clinical integration and tested it in critical care patients in the intensive care unit. Compared with manual references, we achieved a mean difference of 0.8 (95% limits of agreement: –2.9 to 4.7) mm in Echolab patients, with a feasibility of 85.7%. In critical care patients, we reached a mean difference of 0.6 (95% limits of agreement: –2.3 to 3.5) mm and a feasibility of 88.1%. The clinical prototype of autoMAPSE achieved real-time performance. Automatic quantification of LV function had high feasibility in clinical settings. The agreement with manual references was comparable to inter-observer variability of clinical experts. [ABSTRACT FROM AUTHOR]

The stereotactic treatment of single cranial targets using noncoplanar volumetric modulated arc therapy (VMAT) allows for effective dose delivery to the target, while sparing normal brain tissue. In this study, the dosimetric effect of adding dynamic jaw tracking and automatic collimator angle selection in the optimization of single target cranial VMAT plans was investigated. Twenty-two cranial targets, previously treated with VMAT without dynamic jaw tracking and automatic collimator angle optimization (CAO) were chosen for replanning. Target volumes ranged from 0.441cc to 25.863cc with doses between 18Gy and 30Gy delivered in 1 to 5 fractions. Original plans were reoptimized with automatic CAO, keeping all other objectives the same (CAO plans). Next, original plans were reoptimized with both dynamic jaw tracking and CAO (DJT plans). Original, CAO, and DJT target doses were compared using the Paddick gradient index (GI) and the Paddick inverse conformity index (ICI), while normal tissue dose was compared using the volume of the normal brain receiving 5Gy, 10Gy, and 12Gy. The normal tissue volume was normalized to target size to allow cross comparison between plans. A one-sided t-test was performed to determine whether the changes in the plan metrics were statistically significant. CAO plans had improved GIs compared to the originals (p = 0.03) with insignificant changes in other plan metrics (p > 0.20). The addition of dynamic jaw tracking in DJT plans greatly improved ICIs and normal brain metrics (p < 0.01) compared to the CAO plans with minor improvement in ICIs (p = 0.07). The combined effect of adding dynamic jaw tracking and collimator optimization led to improvements in all metrics of the DJT plans when compared to the original (p < 0.02). The addition of dynamic jaw tracking and CAO led to improvements in both target and normal tissue dose metrics for single-target noncoplanar cranial VMAT plans. [ABSTRACT FROM AUTHOR]

High spatial-resolution satellite images show the presence of numerous eddies in the deep Kuril Basin of the Okhotsk Sea, where in-situ measurements acquired within eddies are relatively rare. We conducted the first altimetry-based systematic census of mesoscale eddies in the Kuril Basin in 1993–2021 using the automatic eddy tracking algorithm AMEDA. The dominance of cyclonic eddies over anticyclonic eddies was observed, which contradicts the common opinion that anticyclonic eddies prevail over cyclonic ones in the Kuril Basin. The paper focuses mainly on the long-lived eddies with the lateral size in the range from several tens of kilometers to some hundreds of kilometers and with the lifetime exceeding 30 days. It was found that these eddies are inhomogeneously distributed over the study area with high values of occurrence frequency in some domains. This is explained by the topographic features and peculiarities of the circulation in the Basin where Soya Warm Current water, Okhotsk Sea water and subarctic Pacific water circulate and mix. The fractions of these water masses and their seasonal and interannual variations within the surface cores of the eddy were estimated using a particle-tracking technique. The kinematic characteristics of these eddies have been computed as well. The vast majority of the anticyclonic and cyclonic eddies have the nonlinearity parameter exceeding one implying that the eddies in the Kuril Basin are coherent features transporting water with its properties. Peculiarities in distribution of formation, occurrence and decay locations have been analyzed. Our results have been compared with shipboard and buoy's observations and numerical simulation of eddies in the Kuril Basin. • The first altimetry-based census of mesoscale Kuril Basin eddies (KBE) in Okhotsk Sea. • Kinematic and dynamic properties of KBEs are studied in 1993–2021 using a tracking algorithm. • Content of various water types inside KBEs is estimating using daily-computed Lagrangian maps. [ABSTRACT FROM AUTHOR]

Automatic tracking of viral and intracellular structures displayed as spots with varying sizes in fluorescence microscopy images is an important task to quantify cellular processes. We propose a novel probabilistic tracking approach for multiple particle tracking based on multi-detector and multi-scale data fusion as well as Bayesian smoothing. The approach integrates results from multiple detectors using a novel intensity-based covariance intersection method which takes into account information about the image intensities, positions, and uncertainties. The method ensures a consistent estimate of multiple fused particle detections and does not require an optimization step. Our probabilistic tracking approach performs data fusion of detections from classical and deep learning methods as well as exploits single-scale and multi-scale detections. In addition, we use Bayesian smoothing to fuse information of predictions from both past and future time points. We evaluated our approach using image data of the Particle Tracking Challenge and achieved state-of-the-art results or outperformed previous methods. Our method was also assessed on challenging live cell fluorescence microscopy image data of viral and cellular proteins expressed in hepatitis C virus-infected cells and chromatin structures in non-infected cells, acquired at different spatial–temporal resolutions. We found that the proposed approach outperforms existing methods. [Display omitted] • Probabilistic particle tracking by multi-detector fusion and Bayesian smoothing. • Detections from multiple detectors fused by intensity-based covariance intersection. • Multi-detector fusion exploits image intensities, positions, and uncertainties. • Information from future time points integrated by Bayesian smoothing. • Quantitative evaluation using microscopy images of viral and chromatin structures. [ABSTRACT FROM AUTHOR]

• A novel approach to intelligently analyze litchi fruit abscission of branches in the orchard. • A fixed-point and fixed-focus data acquisition system was developed. • The improved SuperGlue network was employed to match branches and extract the ROI-B of the target branch. • The fruit abscission on a total of 80 branches , over a continuous 56-day period was analysed by using the proposed method. Automatic tracking of litchi fruit abscission provides important information for the orchard management. Currently, the prevalent method for tracking litchi fruit abscission involves manual selection of branches followed by daily fruit counting, a process deemed both inefficient and subjective. Therefore, this paper proposed a novel approach that integrated an improved image matching and object detection algorithms to intelligently analyze litchi fruit abscission of branches in the orchard. The method consisted of three steps. Firstly, an automatic fixed-point focus system was developed to capture long-term image sequences of selected branch areas. From April 27th to June 21st, 2022, the system autonomously navigated to specified locations and automatically adjusted the focal length to capture daily images of selected branches, thereby forming image time series is crucial for analysis. Secondly, to reduce fruit counting errors due to interferences from other branches, an improved SuperGlue algorithm named SE2-SuperGlue was developed to match branches and extract the region of interest of the target branch (ROI-B) from the image time series. The SE2-SuperGlue algorithm introduced an equivariant steerable CNN that is equivariant to translations and image rotations, rendering it robust to rotations and ensuring accurate matching amid changes in branch angles. Thirdly, the ROI-B sequences were input into the YOLOv8n to detect litchi fruits, yielding tracking results of litchi fruit abscission on the image time series. Experimental results showed the good performance of the proposed method. The SE2-SuperGlue network achieved higher matching score (0.53) and a speed of 177 ms/frame compared to conventional matching algorithms, successfully extracting the ROI-B sequence. Additionally, YOLOv8n achieved an mAP@0.5 of 98.7 %, and the correlation coefficient R2 between prediction and ground truth for automatic tracking of daily fruit count over a 56-day image time series of selected branches was 0.98. Furthermore, the study analyzed a continuous 56-day image sequence depicting relative fruit abscission for 80 branches of three varieties (Guiwei, Nuomici, and Huaizhi), along with the peaks of fruit abscission. [ABSTRACT FROM AUTHOR]

Despite considerable progress in the field of automatic multi-target tracking, several problems such as data association remained challenging. On the other hand, cognitive studies have reported that humans can robustly track several objects simultaneously. Such circumstances happen regularly in daily life, and humans have evolved to handle the associated problems. Accordingly, using brain-inspired processing principles may contribute to significantly increase the performance of automatic systems able to follow the trajectories of multiple objects. In this paper, we propose a multiple-object tracking algorithm based on dynamic neural field theory which has been proven to provide neuro-plausible processing mechanisms for cognitive functions of the brain. We define several input neural fields responsible for representing previous location and orientation information as well as instantaneous linear and angular speed of the objects in successive video frames. Image processing techniques are applied to extract the critical object features including target location and orientation. Two prediction fields anticipate the objects' locations and orientations in the upcoming frame after receiving excitatory and inhibitory inputs from the input fields in a feed-forward architecture. This information is used in the data association and labeling process. We tested the proposed algorithm on a zebrafish larvae segmentation and tracking dataset and an ant-tracking dataset containing non-rigid objects with spiky movements and frequently occurring occlusions. The results showed a significant improvement in tracking metrics compared to state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

• A resonant frequency tracking method based on phase control method is proposed. • A shunted piezoelectric transducer which could tune itself by digitally programmable inductance is presented. • Fuzzy PI-based frequency tracking algorithm is introduced with the advantage of eliminating the percent overshoot. • An impedance matching network is designed to accommodate piezoelectric transducers with different parameters. Piezoelectric ultrasonic transducers, vital in medical devices and aerospace, often face challenges like resonant frequency shifts and impedance variations affecting their operational efficiency. This paper introduces a shunted piezoelectric transducer which could tune itself by digitally programmable inductance. A transformer and inductance-capacitance matching network ensures enhanced compatibility and impedance management. Proposing a fuzzy PI-based phase control method achieves resonant frequency tracking, synchronizing operational frequency with the transducer. In contrast to traditional methods, our approach enables faster and more precise fine-tuning, detecting and rectifying real-world deviations for optimal performance. A comprehensive experimental validation, based on fundamental knowledge analysis, confirms the feasibility and superiority of our proposed method, and the commonly encountered issues of resonance frequency deviation and impedance variation in high-power piezoelectric transducer applications can be effectively mitigated. [ABSTRACT FROM AUTHOR]

This paper investigates the multi-sensor distributed fusion with memory configuration for multi-target automatic tracking (MTAT) in clutter. Aimed at fully addressing the problems of both track origin ambiguity and the strong-correlation among system track and its paired local tracks in the framework, this paper derives an Exact Unified Multisensory Tracks Fusion with Memory (EU-MTFwM) method that fuses the system track with enumerated combinatorial tracks to achieve fusing all sensors' local tracks at once. It deploys track-to-track probabilistic association (T2TPA) and target-existence based track automatic management (TAM) to mitigate track origin ambiguity caused by clutter disturbance and the presence of multiple targets. Additionally, it implements the exact distributed kinematic state estimation (exact DKSE) to remove strong correlation stemming from common historical measurements and process noise. However, as the number of sensors to be fused increases, the proposed EU-MTFwM suffers from the combinatorial explosion. An Implementable Unified Multisensory Tracks Fusion with Memory (IU-MTFwM) method is then derived to sequentially fuse the system track with origin-ambiguous local tracks from multiple sensors one-by-one, providing a unified but computationally efficient alternative to the EU-MTFwM. Furthermore, a novel system track initialization technology is proposed to trigger the multisensory tracks fusion recursion. Numerical results show the proposed two unified methods achieve comparable tracking performance to the centralized fusion benchmark but with significantly reduced communication consumption, and demonstrate intensive improvement over the enhanced state-of-the-art that incorporates our proposed system track initialization into its fusion recursion. [ABSTRACT FROM AUTHOR]

For increasing the solar energy harvesting performance, any solar tracking system must track the sun at the right angle during periods of daytime. We proposed an enhanced solar tracking instrument to solve the issues of low sunlight incident angles and manual orientation/elevation calibrations. This novel tracker used two Global Positioning System (GPS) receivers with Real-Time Clock (RTC) forming a satellite compass and combined with an inclinometer to follow the sun trajectory correctly without manual aligning the heading of system to orientation of North-South. Such a dual-axis instrument was controlled by a microcontroller unit (MCU) integrated with servo-motors, satellite compass and inclinometer. The embedded Solar Position Algorithm (SPA) can determine the angles of the sun vector at the given date, time and geographical information from the satellite compass. The power generated by the proposed tracking instrument, compared with a fixed-tilted photovoltaic (PV) tracker, was proven in scenarios of fixing, rotating and moving. The proposed solar tracker with satellite compass after field measurement extracted more energy than the fixed-tilted solar tracker under fixed location with mostly clear day 35.91%, in different locations with clear day 38.72% and 38.40%; in heading variation and moving conditions were 60.00% and 113.70%. Therefore, this solar tracker is not only suitable for PV grid but also well suitable for mobile vehicles with keeping the panel aligned with the sun automatically, whether geographical position in northern or southern Hemisphere independently and without solar panel tilted orientation adjustment to achieve the robust, convenient and reliable requirements. • An enhanced open-loop tracking system with GPS for auto-positioning • Satellite compass consisted of GPS and RTC calibrates solar position in real time. • Tracking operation controlled by MCU with embedded SPA and satellite compass [ABSTRACT FROM AUTHOR]

Precise measurement of joint-level motion from stereo-radiography facilitates understanding of human movement. Conventional procedures for kinematic tracking require significant manual effort and are time intensive. The current work introduces a method for fully automatic tracking of native knee kinematics from stereo-radiography sequences. The framework consists of three computational steps. First, biplanar radiograph frames are annotated with segmentation maps and key points using a convolutional neural network. Next, initial bone pose estimates are acquired by solving a polynomial optimization problem constructed from annotated key points and anatomic landmarks from digitized models. A semidefinite relaxation is formulated to realize the global minimum of the non-convex problem. Pose estimates are then refined by registering computed tomography-based digitally reconstructed radiographs to masked radiographs. A novel rendering method is also introduced which enables generating digitally reconstructed radiographs from computed tomography scans with inconsistent slice widths. The automatic tracking framework was evaluated with stereo-radiography trials manually tracked with model-image registration, and with frames which capture a synthetic leg phantom. The tracking method produced pose estimates which were consistently similar to manually tracked values; and demonstrated pose errors below 1.0 degree or millimeter for all femur and tibia degrees of freedom in phantom trials. Results indicate the described framework may benefit orthopaedics and biomechanics applications through acceleration of kinematic tracking. [ABSTRACT FROM AUTHOR]

The neural underpinnings of processing concrete and abstract semantics remain poorly understood. Previous fMRI studies have shown that multimodal and amodal neural networks respond differentially to different semantic types; importantly, abstract semantics activates more left-lateralized networks, as opposed to more bilateral activity for concrete words. Due to the lack of temporal resolution, these fMRI results do not allow to easily separate language- and task-specific brain responses and to disentangle early processing stages from later post-comprehension phenomena. To tackle this, we used magnetoencephalography (MEG), a time-resolved neuroimaging technique, in combination with a task-free oddball mismatch negativity (MMN) paradigm, an established approach to tracking early automatic activation of word-specific memory traces in the brain. We recorded the magnetic MMN responses in 30 healthy adults to auditorily presented abstract and concrete action verbs to assess lateralization of word-specific lexico-semantic processing in a set of neocortical areas. We found that MMN responses to these stimuli showed different lateralization patterns of activity in the upper limb motor area (BA4) and parts of Broca's area (BA45/BA47) within ∼100–350 ms after the word disambiguation point. Importantly, the greater leftward response lateralization for abstract semantics was due to the lesser involvement of the right-hemispheric homologues, not increased left-hemispheric activity. These findings suggest differential region-specific involvement of bilateral sensorimotor systems already in the early automatic stages of processing abstract and concrete action semantics. [Display omitted] • Neuromagnetic MMN responses to abstract and concrete spoken verbs show different lateralization. • The difference is due to a greater right-hemispheric activity for concrete than abstract stimuli. • The effect is regionally specific, it manifests within bilateral M1 and IFG areas, but not in temporal lobes. [ABSTRACT FROM AUTHOR]

[Display omitted] • Frontal midline θ (Fmθ) increases when tracking error exceed a threshold in controls. • Above threshold Fmθ scales with tracking error in controls. • Fmθ failed to increase and scale with error following cerebellar damage. • Cerebellum generates prediction error entraining θ rhythm in frontal midline cortex. • Cerebellum activates the executive network when above-threshold errors occur. Frontal midline θ (Fmθ) activity occurs in medial prefrontal cortices (mPFC), when expected and actual outcomes conflict. Cerebellar forward models could inform the mPFC about this mismatch. To verify this hypothesis we correlated the mPFC activation during a visuomotor tracking task (VM) with performance accuracy, in control and cerebellum-lesioned participants. Additionally, purely visual (V), motor (M) and a motor plus visual tasks (V + M) were performed. An Independent Component, with a mid-frontal topography scalp map and equivalent dipole location in the dorsal anterior cingulate cortex accounted for Fmθ. In control participants Fmθ power increased during VM, when the error level crossed a threshold, but not during V + M, M and V. This increase scaled with tracking error. Fmθ power failed to increase during VM in cerebellar participants, even at highest tracking errors. Thus, in control participants, activation of mPFC is induced when a continuous monitoring effort for online error detection is required. The presence of a threshold error for enhancing Fmθ, suggests the switch from an automatic to an executive tracking control, which recruits the mPFC. Given that the cerebellum stores forward models, the absence of Fmθ increases during tracking errors in cerebellar participants indicates that cerebellum is necessary for supplying the mPFC with prediction error-related information. This occurs when automatic control falters, and a deliberate correction mechanism needs to be triggered. Further studies are needed to verify if this alerting function also occurs in the context of the other cognitive and non-cognitive functions in which the cerebellum is involved. [ABSTRACT FROM AUTHOR]

• Early Candidate Elimination Module reduces computational load. • The Adaptive Appearance Update enhances the reliability of model updates. • The model exhibits good generalization capabilities. • This study provides technical support for motion analysis of dairy goat. Intelligent tracking and monitoring are fundamental to achieving precise posture recognition, health assessment, and accurate breeding in dairy goats. In view of the motion characteristics of dairy goats, which often exhibit significant posture changes due to their lively and active movements, we present an object tracking method for a single dairy goat based on MixFormer optimized through adaptive token removal and efficient appearance update. Firstly, we collected motion data from two different breeding farms and trained the model solely on one farm and tested it in the other to verify its generalization performance. Then, through an early Candidate Elimination (CE) module, we pre-filter redundant background and similar object information in dairy goats, thereby reducing computational load and enhancing tracking speed. Finally, we introduce the adaptive appearance update strategy named Average Peak-to-Correlation Energy (APCE) to maximize the utilization of target-related historical information, thus improving target localization accuracy. The experimental results indicate that the improved MixFormer achieves 75.79% and 74.28% in terms of AUC (Area Under Curve) and Precision, respectively. These results represent a 2.75% and 7.38% enhancement over the MixFormer, outperforming KYS, KeepTrack, STARK and TOMP. The improved MixFormer runs at 30.5 fps, meeting real-time tracking requirements. This study demonstrates that the proposed approach offers an effective solution for automatic tracking and monitoring of dairy goats in complex breeding conditions. [ABSTRACT FROM AUTHOR]

Drifting icebergs represent a significant hazard for polar navigation and are able to impact the ocean environment around them. Freshwater flux and the associated cooling from melting icebergs can locally decrease salinity and temperature and thus affect ocean circulation, biological activity, sea ice, and –on larger spatial scales– the whole climate system. However, despite their potential impact, the large-scale operational monitoring of drifting icebergs in sea ice-covered regions is as of today typically restricted to giant icebergs, larger than 18.5 km in length. This is due to difficulties in accurately identifying and following the motion of much smaller features in the polar ocean from space. So far, tracking of smaller icebergs from satellite imagery thus has been limited to open-ocean regions not covered by sea ice. In this study, a novel automated iceberg tracking method, based on a machine learning-approach for automatic iceberg detection, is presented. To demonstrate the applicability of the method, a case study was performed for the Weddell Sea region, Antarctica, using 1213 Advanced Synthetic Aperture Radar (ASAR) satellite images acquired between 2002 and 2011. Overall, a subset of 414 icebergs (3134 re-detections in total) with surface areas between 3.4 km2 and 3612 km2 were investigated with respect to their prevalent drift patterns, size variability, and average disintegration. The majority of the tracked icebergs drifted between 1.3 km and 2679.2 km westward around the Antarctic continent, following the Antarctic Coastal Current (ACoC) and the Weddell Gyre, at an average drift speed of 3.6 ± 7.4 km day−1. The method also allowed us to estimate an average daily disintegration (i.e. iceberg area decrease) rate of ~0.13% (~37% year−1) for all icebergs. Using the sum of all detected individual surface area reductions, we estimate a total iceberg mass decrease of ~683 Gt year−1, which can be freshwater input and/or new 'child' icebergs calved from larger icebergs. The extension to an automated long-term tracking method for icebergs is challenging as the iceberg shape can vary significantly due to abrupt disintegration or calving of bergy bits. However, our machine learning approach extended by automatic shape-based tracking capabilities proved to be a reliable alternative for automatic detection and tracking of icebergs, even under the ambiguous SAR background signatures often found in the Southern Ocean. In particular, the method works in the challenging near-coastal environment where the presence of sea ice and coastal ocean dynamics such as surface waves usually pose major obstacles for other approaches. [ABSTRACT FROM AUTHOR]

Optical-based navigation for space is a field growing in popularity due to the appeal of efficient techniques such as Visual Simultaneous Localisation and Mapping (VSLAM), which rely on automatic feature tracking with low-cost hardware. However, low-level image processing algorithms have traditionally been measured and tested for ground-based exploration scenarios. This paper aims to fill the gap in the literature by analysing state-of-the-art local feature detectors and descriptors with a taylor-made synthetic dataset emulating a Non-Cooperative Rendezvous (NCRV) with a complex spacecraft, featuring variations in illumination, rotation, and scale. Furthermore, the performance of the algorithms on the Long Wavelength Infrared (LWIR) is investigated as a possible solution to the challenges inherent to on-orbit imaging in the visible, such as diffuse light scattering and eclipse conditions. The Harris, GFTT, DoG, Fast-Hessian, FAST, CenSurE detectors and the SIFT, SURF, LIOP, ORB, BRISK, FREAK descriptors are benchmarked for images of Envisat. It was found that a combination of Fast-Hessian with BRISK was the most robust, while still capable of running on a low resolution and acquisition rate setup. For large baselines, the rate of false-positives increases, limiting their use in model-based strategies. [ABSTRACT FROM AUTHOR]

Combining the current development status of intelligent robots, obstacle avoidance and automatic tracking are the focus of robot travel problems.Based on the development status of domestic intelligent control systems, current technology, etc, this paper uses arduino as the core control system, combined with infrared tracking module. Four modules, such as ultrasonic obstacle avoidance module, motor drive module and power module, have designed a good control scheme, thus realizing the intelligent tracking and obstacle avoidance function of the wheeled robot. [ABSTRACT FROM AUTHOR]

The main obstacle to applying aluminum oxide doped with carbon and magnesium (Al 2 O 3 :C,Mg)-based fluorescent nuclear track detectors (FNTDs) in radionuclide therapy with alpha emitters is that it has not been known how to achieve an accurate and in situ dose assessment for in vitro cellular experiments. In this study, we developed an automatic 3D track reanalysis algorithm (ATRA) to analyze the alpha energy spectra in FNTDs. The algorithm provides an excellent track identification rate and an excellent track linking rate (0.961) without significant variations in various track densities. The algorithm also achieves a reliable track detection rate exceeding 0.927 for track densities at the level of in situ measurement, and it successfully measures the alpha energy spectra for obtaining the absorbed dose. We also confirmed that the track overlap clearly interferes with accurate track reconstruction for track densities beyond 1.5 × 106 tracks cm−2. We provided a correction equation that can be used to obtain the actual alpha tracks for high track densities; the equation may also correct estimations of absorbed doses due to alpha particles. • Automatic 3D track reanalysis algorithm (ATRA) was developed. • The track detection rates of ATRA exceeded 0.927 at in situ measurement level. • The effect of track overlap on ATRA was quantified with various track densities. • Correction equation was provided for correcting absorbed dose due to alpha particle. [ABSTRACT FROM AUTHOR]

• An anchor-free deep learning model was proposed to count seedlings and flowers. • The tracking-based counting method didn't use complex motion estimation algorithms. • The counting results of our method highly correlated with ground truth. • Our method would significantly benefit plant breeding and replanting decision-making. Accurate counting of plants and their organs in natural environments is essential for breeders and growers. For breeders, counting plants during the seedling stage aids in selecting genotypes with superior emergence rates, while for growers, it informs decisions about potential replanting. Meanwhile, counting specific plant organs, such as flowers, forecasts yields for different genotypes, offering insights into production levels. The overall goal of this study was to investigate a deep convolutional neural network-based tracking method, CenterTrack, for cotton seedling and flower counting from video frames. The network is extended from a customized CenterNet, which is an anchor-free object detector. CenterTrack predicts the detections of the current frame and displacements of detections between the previous frame and the current frame, which are used to associate the same object in consecutive frames. The modified CenterNet detector achieved high accuracy on both seedling and flower datasets with an overall AP50 of 0.962. The video tracking hyperparameters were optimized for each dataset using orthogonal tests. Experimental results showed that seedling and flower counts with optimized hyperparameters highly correlated with those of manual counts ( R 2 = 0.98 and R 2 = 0.95) and the mean relative errors of 75 cotton seedling testing videos and 50 flower testing videos were 5.5 % and 10.8 %, respectively. An average counting speed of 20.4 frames per second was achieved with an input resolution of 1920 × 1080 pixels for both seedling and flower videos. The anchor-free deep convolution neural network-based tracking method provides automatic tracking and counting in video frames, which will significantly benefit plant breeding and crop management. [ABSTRACT FROM AUTHOR]

Fault interpretation is a key step for seismic structural interpretation and reservoir characterization. In conventional approach, faults are identified as reflection discontinuity or abruption in post-stack seismic data. In order to highlight fault, several fault attributes by measuring waveform similarity or diversity have been proposed. However, noise and stratigraphic features can also be highlighted in a fault attribute image. We propose a forward and backward diffusion method to enhance the fault image so that the noisy features which unrelated to faults are suppressed and the fault features are sharpened. We first estimate the planarity (linearity in 2D) and fault orientation (strike and dip) of every sample in fault image by using a PCA-based algorithm. Combining the planarity and fault orientation, we then construct fault probability map based on the assumption that faults can be seemed as relatively steep planes in a small spatial window. Thanks to the fault probability map, we can adaptively switch the diffusion process from a forward mode to a backward mode. Forward diffusion force is used to suppress the non-fault features within the low fault probability area and backward diffusion force is used to enhance the fault features within the high fault probability area. A real 3D example is included to demonstrate that this method can effectively suppress the noise and sharpen the faults. Furthermore, automatic fault tracking on an enhanced fault image can get more accurate result than on an original fault image. • A Fault attribute image highlights not only fault features but also noise and stratigraphic features. • We implemented a fault image enhancement scheme based on forward and backward diffusion algorithm. • The proposed method can switch the forward and backward diffusion forces during the diffusion process. • Applying an automatic fault tracking method on enhanced fault image can obtain more accurate result. [ABSTRACT FROM AUTHOR]

Agriculture is the primary occupation in India and is called India's backbone. But of late, a lot of problems are being faced in agriculture by the farmers. One of the major problems being water scarcity. As per surveys, almost 20 percent of the agricultural land is wasted due to water scarcity and becomes a barren land. Thus, this research gives an idea of smart irrigation system. This irrigation system uses three sensors namely temperature sensor, humidity sensor and soil moisture sensor, and fuzzy logic is used to operate the solenoid valve. The data from the sensors is then sent to the cloud by using adafruit.io and the farmer can view the moisture level, humidity level and temperature recorded by these sensors. All the operations are governed by an Arduino and the power supply for the Arduino is given by a solar panel which uses LDR (Light Dependent Resistor) and makes it into an automatic tracking system. The crops are grown in two tubs and comparison of growth of plant with automated irrigation and normal irrigation is carried out. [ABSTRACT FROM AUTHOR]

• A MEMS high-temperature sensitive diaphragm is combined with the fiber microstructure to make the sensing probe. • The F–P acceleration sensor can successfully avoid the fading phenomenon of FBG sensor at high temperature of above 400°C. • A signal demodulation system is proposed based on fast automatically tracking the quiescent operation point. Vibration measurement at high temperature is of great significance to monitoring the operation of major projects and mechanical equipment. As the temperature resistance of electrical vibration sensors is usually below 600°C, it cannot meet the requirements of some high-temperature measuring conditions. This paper studies a high-temperature Fabry–Perot (F–P) acceleration sensing system with general single-mode fiber. A high-temperature sensitive diaphragm is machined based on micro-electromechanical system (MEMS) technology and it is combined with the fiber microstructure to enable the sensing probe to be packaged under high temperatures. According to the temperature characteristic of the length of the F–P cavity, a signal demodulation system was proposed based on the rapid automatic tracking of the quiescent operation point. The complete optical fiber F–P acceleration sensing system conducted real-time vibration detection of high temperature casting in the foundry factory. The experimental result indicates that the optical fiber high-temperature acceleration sensor showed a better vibration response in the working frequency of 100–1000 Hz and the relative error of the frequency measurement is less than 1.56% under normal temperatures and a high temperature of 800°C. [ABSTRACT FROM AUTHOR]

• Monitoring an entire flocks using common solutions such that GPS can be expensive. • Instead using drone image allows monitoring the flock with only one image. • We developed an automatic detection method allowing 73% of the goats to be detected. • We also developed a method to find goat activity with a sensitivity of 78.3%. Monitoring the position of animals in the outdoors can provide useful information in ecology and in agriculture. A common method is to use active sensors, such as GPS, to record their positions at constant intervals of time. But using active sensors can rapidly become expensive when several animals have to be monitored at the same time. Another method is to use a passive sensor to monitor the entire flock of animals. In this article, we propose a method to process images taken by a commercial drone in order to automate the tracking of animal activities. We developed a method that automatically detects goats from the images and tracks their activity using a combination of thresholding and supervised classification methods. We tested our method on 571 drone images taken over 11 days and found a sensitivity of 74% for animal detection and 78.3% for activity detection. [ABSTRACT FROM AUTHOR]

The ocean-side area of the Kuril Islands is a region with high bioproductivity and rich fishing and a convergence of transformed subtropical and subarctic waters with significant impact on weather and climate. Waters of different origin mix in the region including Kuroshio Extention, Japan Sea, Oyashio, Okhotsk Sea and Alaskan Stream waters. Despite the important role of Kuril eddies (KE) in transporting heat, salt, nutrients and their biological activity, there are still major gaps in our knowledge of the properties and dynamics of these energetic and mesoscale features in the northwestern Pacific Ocean. Using altimetry data during 1993–2021, we conducted the first systematic census of KE with the help of the automatic eddy tracking algorithm AMEDA. We separated KE into four groups according to their places of formation in relation to the Kuril-Kamchatka Trench and found a strong inhomogeneous spatial distribution of the Kuril cyclones (KC) and anticyclones (KAC). Different kinematic properties of KE have been studied and analyzed. The Trench has been found to play a role of an eddy guide for long-lived eddies with a comparatively small number of KE being able to cross the Trench. The highest occurrence frequency of KAC was observed along the Trench, especially in the area east off the Bussol Strait, whereas KC occurred more frequently along the ocean side of the Trench. Among the observed long-lived KE, there were more than 90% of eddies with the nonlinearity parameter exceeding 1 implying that the majority of KE are coherent vortices transporting water with its properties. The relative concentrations of waters of different origin inside KAC and KC cores were computed for each group applying the new Lagrangian algorithm. It allowed us to detect transportation of subtropical water in the KE cores to north and subarctic water to south over the distance of the order of 1000 km and more. [ABSTRACT FROM AUTHOR]

• Development of a machine learning approach for artichoke plant detection intended for real-time UAS spraying applications. • Creation of an automatic multi-temporal tracking procedure for crop monitoring and UAS path planning development. • Optimization of agrochemicals distribution operation from UAS. • Detection performances comparison between the Feature Pyramid and YOLOv5 SSD networks for plant detection. Input optimization is a distinguishing characteristic of Precision Agriculture approaches, helping reduce the environmental impact and costs and increase vegetable production quality. Thanks to the high automation evolution of Unmanned Aerial Systems (UAS), a new approach derived from their combination with Deep Learning techniques is leading to significant improvements in agricultural management practices. The study aims at artichoke plants detection and georeferencing as a first step for an on-the-fly, real time, UAS spraying system, and use the gathered information to monitor crop development through a multi-temporal approach. A commercial UAS, equipped with an RGB sensor, acquired images of the artichoke field located in Sardinia (Italy) during the 2021–2022 season in different crop growth stages. The FPN (Feature Pyramid Network), trained and compared with the YOLOv5 (You Only Look Once) network, showed a high detection level with an average F1 score of around 90%, and satisfactory off-line performances on the Nvidia Jetson Nano board. YOLOv5 achieved the best overall result. The FPN recorded a lower recall, which is desirable to achieve a minimum number of detection errors and limit the leakage of agrochemicals on false-positive targets. The multi-temporal approach influenced detection performances, with an inverse response of precision and recall metrics. The growing index trend showed a distinct value in October, peaking at the beginning of December as expected. The proposed approach contributes to designing future automatic and reliable site-specific UAS agrochemicals application and the classification of management zones. [ABSTRACT FROM AUTHOR]

Abstract Correlation filter based trackers have achieved appealing performance and high efficiency in recent years. However, for long-term tracking where target objects undergo dramatic appearance variation due to heavy occlusion or out-of-view, conventional correlation filter based tracking algorithms would be distracted by irrelevant objects. Once the trained tracker loses its way, it is impossible to recover the information for the following frames as the model has drifted. In this paper, we decompose the long-term tracking task into tracking and detection. Tracker learns separate correlation filters for explicit translation and scale estimation. Specifically, in order to improve tracking accuracy, the convolutional features for translation filter are extracted, and the scale filter is learned using the target appearance sampled at different scales. Detector trains an online long-term filter and applies it to the entire frame to generate detection proposals. By exploiting these detection proposals, it helps the tracker to recover from problems such as temporary or persistent occlusions. In this way, the proposed approach could handle the model drifting problem effectively for long-term tracking with more accurate estimation of object scale and location. Extensive experimental results on large-scale benchmark sequences have shown the robustness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract The convolutional network-based tracking (CNT) algorithm provides a training network with warped target regions in the first frame instead of large auxiliary datasets, which solves the problem of convolutional neural network (CNN)-based tracking requiring very long training time and a large number of auxiliary training samples. However, the two-layer CNT uses only gray feature that causes sensitivity to appearance variations. Besides, some samples with useless information should be removed to avoid drifting problems. For these reasons, a multi-layer convolutional network-based visual tracking algorithm via important region selection (IRST) is proposed in this paper. The proposed important region selection model is built via high entropy selection and background discrimination, which enables the training samples to be informative in order to provide enough stable information and also be discriminative so as to resist distractors. The feature maps are also obtained by weighting the template filters with cluster weights. Instead of simple gray features, IRST adds the Gabor layer to explore the texture feature of the target that is effective on coping with illumination and rotation variations. Extensive experiments show that the proposed algorithm achieves superior performances in many challenging visual tracking tasks. [ABSTRACT FROM AUTHOR]

Highlights • Tensor based discriminative tracking framework is presented in this paper. • A multi-linear classifier with structured output is proposed for tensor input. • Parameter tensor reconstruction in online updating provides robustness against noise. • Tensor block coordinate descent optimization is introduced in online learning. • The proposed tracker shows superior performance on benchmark videos. Abstract Discriminative tracking algorithms have witnessed continued progress for distinguishing the target from background in unconstrained environments. The learning and detection task in existing visual tracking methods often convert a multidimensional data array into a vector-based observation. By altering the 2-D spatial structure of the image, transformation variants and global noises influence the discriminative ability of target representation, often result in degradation of performance. Different from vector representations, this paper presents a tensor-based large margin discriminative framework for visual tracking that utilizes the supervised tensor learning. In our method, an online structured support tensor classifier is designed which produces the multi-linear decision function, incorporating the nonlinearity of tensor-based feature over the target. In order to provide better spatial cues of target representation against noises and facilitate online tracking, we further introduce truncated tucker decomposition in structured multi-linear learning. The proposed algorithm poses an effective parameter tensor reconstruction in the classifier updating procedure and has a robust discriminative ability against several video background variants. Furthermore, a tensor block coordinate descent optimization is presented to achieve a closed form solution specific to the proposed truncated structured Tucker machine (TSTM). Experiment results on a recent comprehensive tracking benchmark demonstrate a promising performance of the proposed method subjectively and objectively compared with several state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

This paper presents a combined Nonlinear Model Predictive Control (NMPC) for position and velocity tracking of underactuated surface vessels, and collision avoidance of static and dynamic objects into a single control scheme with sideslip angle compensation and environmental disturbances counteraction. A three-degree-of-freedom (3-DOF) dynamic model is used with only two control variables: namely, surge force and yaw moment. External environmental forces are considered as constant or slowly varying disturbances with respect to the inertial frame, and hence nonlinear for the body frame of the vessel. Nonlinear disturbance observer (NDO) is used to estimate these disturbances in order to be fed into the prediction model and enhance the robustness of the controller. A nonlinear optimization problem is formulated to minimize the deviation of the vessel states from a time varying reference generated over a finite horizon by a virtual vessel. Sideslip angle is considered in the cost function formulation to account for tracking error caused by the transverse external force in the absence of sway control force. Collision avoidance is embedded into the trajectory tracking control problem as a time-varying nonlinear constraint of position states to account for static and dynamic obstacles. MATLAB simulations are used to assess the validity of the proposed technique. [ABSTRACT FROM AUTHOR]

Tracking of signals in Nuclear Magnetic Resonance (NMR) spectra is a basic technique used in drug discovery, systems and structural biology. Current experimental setups allow to measure hundreds of spectra, which require analysis, ideally in an automated and reproducible manner. In this study, we present a novel approach to the automate tracking of signals in NMR spectroscopy. We model perturbation of the signal position with Gaussian process regression, which allows for scalability and robustness. The proposed method outperforms the other existing tracking routines, and is applicable for complex data featuring common hindrances such as: missing signals, noise and crossing trajectories. [ABSTRACT FROM AUTHOR]

An autotracking algorithm based on a hybrid grey prediction model is presented for autonomously navigating the laparoscopic visual window of a robot-assisted surgical system. This method can be applied to any view angle of the three-dimensional (3D) laparoscope with a 200 ms predictive motion. Firstly, a preset parameter-based tracking algorithm is proposed based on the kinematic relationships between instrument arms and laparoscope arm. Subsequently, a hybrid grey prediction model is constructed through the combination of the optimized GM(1,1) and grey Verhulst models with the use of an adaptive weight-tuning method and a filtered amendment method. Furthermore, the algorithm results in a constant distribution area ratio, whereby the instrument marks can be guaranteed to lie within the field of the visual window, such that the concurrent motion of the visual window and the instrument marks can be realized. The visual window can sustain automatic tracking of the movement of the marks. The user does not have to switch to the target controlling mode by adjusting the master-slave mapping. Finally, the proposed algorithm is verified through simulations with real motion trajectories from a Phantom Omni master manipulator. The results validate the correctness, feasibility, and robustness of this approach. [ABSTRACT FROM AUTHOR]

In this paper, we address the problem of tracking an object in a video sequence given its location in the first frame and no other information. Many existing discriminative tracking algorithms usually train a classifier in an on-line manner to separate the object of interest from the background. Slight inaccuracies in the tracking may result in incorrectly labelled training set, which can degrade the tracker. Although a number of approaches such as semi-supervised learning and multiple instance learning have been developed to address this problem, some critical issues still remain unsolved. This work aims to mitigate the shortcomings by exploiting a reliable generative model to support the discriminative learning process. A prior model based on a set of structured Dirichlet-multinomial distributions is proposed to preserve the target's structure information. This prior is then formulated as a regularization term in a training objective function, which casts the tracking task as a prior regularized semi-supervised learning problem. A multi-objective optimization method is developed to search for the solution, taking advantage of a decision maker inside to control the conflicts between different modules. The experiments show that this proposed method outperforms standard algorithms on challenging datasets. It is also demonstrated that the algorithm significantly mitigates the error accumulation effect. [ABSTRACT FROM AUTHOR]

The structural, optical and photoelectrical properties of the fabricated Al/PbS/p-Si/Al photodiode based on PbS nanocrystallites were investigated. The PbS nanocrystallites were characterized by X-ray diffraction (XRD), UV–VIS–NIR, Infrared and Raman spectroscopy. The XRD diffraction peaks show that the prepared PbS nanostructure is in high crystalline state. Various electrical parameters of the prepared photodiode were analyzed from the electrical characteristics based on I-V and C-V-G. The photodiode has a high rectification ratio of 5.85×10 4 at dark and ±4 V. Moreover, The photocurrent results indicate a strong photovoltaic behavior. The frequency dependence of capacitance and conductance characteristics was attributed to depletion region behavior of the photodiode. The diode was used to control solar panel power automatic tracking controller in dual axis. The fabricated photodiode works as a photosensor to control Solar tracking systems. [ABSTRACT FROM AUTHOR]

In this study, we report an automatic system for collection of tilt series for electron tomography based on the ultra-HVEM in Osaka University. By remotely controlling the microscope and reading the observation image, the system can track the field of view and do focus in each tilt angle. The automatic tracking is carried out with an image matching technique based on normalized correlation coefficient. Auto focus is realized by the optimization of image sharpness. A toolkit that can expand the field of view with technique of image stitching is also developed. The system can automatically collect the tilt series with much smaller time consumption. [ABSTRACT FROM AUTHOR]

The article proposes a new robot programming-by-demonstration framework, which integrates a visual servoing tracking control to robustly follow a trajectory generated from observed demonstrations. The constraints originating from the use of a visual servoing controller are incorporated into the trajectory learning phase, to guarantee feasibility of the generated plan for task execution. The observational learning is solved as a constrained optimization problem, with an objective to generalize from a set of trajectories of salient features in the image space of a vision camera. The proposed approach is evaluated experimentally for learning trajectories acquired from kinesthetic demonstrations. [ABSTRACT FROM AUTHOR]

• We used an automatic detection method to locate and track a warm mesoscale eddy from satellite data. • We deployed hydrophone and temperature-sensor arrays in the center and outside the eddy. • We recorded acoustic signals from explosive sources and chirped signals from a transceiver. • We combined the measured hydrological data to create a spatially and temporally dependent empirical model of the eddy. • We compared the simulations with and without an eddy to investigate the effects of the eddy on sound propagation. • We utilized the measured data to analyze the effects of the eddy on transmission losses to the two hydrographic arrays. • We employed recordings of the chirped signals to study the effects of the eddy on sound-energy fluctuations. The spatial and temporal redistribution of sound energy caused by a warm eddy was measured via a multiday experiment conducted in the South China Sea, utilizing two vertical line arrays, a mooring source, and some wide-band explosive sources. In the experiment area, a warm eddy was identified and tracked using an automatic detection method combined with the empirically reconstructed data. During the experimental period, the warm eddy diminished in size and kinetic energy until it vanished. The influence of the warm eddy on sound propagation during its lifespan is comprehensively analyzed and explained. The experimental acoustic results and simulations based on dynamic hydrographic data demonstrate that energy is redistributed both spatially and temporally due to the presence of the warm eddy. In comparison to the environment without an eddy, the first convergence zone disappears and the high-energy region appears in the typical position of the first shadow zone. The oscillating disturbance in energy is mainly affected by the fluctuation of sound speed caused by the warm eddy. [ABSTRACT FROM AUTHOR]

Submarine cables and pipelines laid on the seafloor are critical infrastructures for energy transportation and communication transmission and must be inspected in a timely manner to determine maintenance needs. Autonomous inspection of cables (including pipelines) and their surroundings using underwater vehicles is difficult because of sensor myopia and deviations in cable routing. This study proposes an automatic cable-tracking method (ACTM) based on side-scan sonar (SSS) that endows autonomous underwater vehicles (AUVs) with autonomous decision-making capabilities to achieve fast localization and stable tracking of submarine cables with large uncertainties. Compared with the traditional predefined waypoint method, the proposed method decomposes the tasks into a series of independent behaviors according to operational objectives; thus, the AUV can perform online dynamic task replanning and realize an autonomous coverage search of the target region. When a cable is detected, the cable-tracking task is constructed as a path-following control problem on the horizontal plane. To reduce uncertainties caused by environmental noise and sensor measurement errors, a uniform motion model based on target tracking theory was used to model the virtual mass abstracted from the cable, and the states of the cable were estimated using the Kalman filter. Subsequently, an adaptive line-of-sight guidance algorithm for the SSS is established to dynamically adjust the look-ahead distance by introducing states such as cable curvature and angular change rate, which somewhat alleviates the tracking loss caused by the SSS short-sightedness. The proposed ACTM effectiveness and robustness were validated through numerical simulations and field tests. The results showed that the proposed method can achieve automatic search and adaptive tracking of cables, further reducing the role of humans in the task. • Automatic submarine cable-tracking method based on side-scan sonar. • Behavior-based planner endows AUVs with autonomous decision-making capabilities. • Introducing target tracking theory to model the point abstracted from the cable. • Adaptive line-of-sight method used to alleviate the influence of the myopic sensors. • Full-scale cable tracking tests were utilized for method verification. [ABSTRACT FROM AUTHOR]

• The demand design investigates a large number of power ultrasonic applications that seeks out a valuable clue, namely the mechanism consistency of the acting medium and the similarity of ultrasonic signals, it proves the necessity and feasibility of designing a universal modular generator model. • Plentiful state-of-the-art power semiconductor devices, digital chips, power conversion circuit modules, impedance matching circuits and control methods are summarized, which lay the foundation of the PUG system. A detailed understanding of the advantages and limitations of each part is essential to help researchers determine the most suitable modular design for the specific application. • A new PUG proposed in this paper differing from other traditional ultrasonic generator possesses two feedback channels. The power ultrasonic generator (PUG) is the core device of power ultrasonic technology (PUT), and its performance determines the application of this technology in biomedicine, semiconductor, aerospace, and other fields. With the high demand for sensitive and accurate dynamic response in power ultrasonic applications, the design of PUG has become a hot topic in academic and industry. However, the previous reviews cannot be used as a universal technical manual for industrial applications. There are many technical difficulties in establishing a mature production system, which hinder the large-scale application of PUG for piezoelectric transducers. To enhance the performance of the dynamic matching and power control of PUG, the studies in various PUT applications have been reviewed in this article. Initially, the demand design covering the piezoelectric transducer application and parameter requirements for ultrasonic and electrical signals is overall summarized, and these parameter requirements have been recommended as the technical indicators of developing the new PUG. Then the factors affecting the power conversion circuit design are analyzed systematically to realize the foundational performance improvement of PUG. Furthermore, advantages and limitations of key control technologies have been summarized to provide some different ideas on how to realize automatic resonance tracking and adaptive power adjustment, and to optimize the power control and dynamic matching control. Finally, several research directions of PUG in the future have been prospected. [ABSTRACT FROM AUTHOR]

• Proposed an efficient and real-time algorithm YOLO-BYTE for multi-object tracking. • Introduced feature extraction module that integrates convolution and self-attention. • Reduced complexity through an optimized Spatial Pyramid Pooling module. • Improved state parameter prediction with the Enhanced ByteTrack algorithm. Dairy cows tracking is an essential means to obtain their behavioral information, real-time position, activity data, and health status. A multi-object tracking method (YOLO-BYTE) is proposed to address the problem of missed detection and false detection caused by complex environments in cow individual detection and tracking. The method improves upon the YOLO v7 Backbone network feature extraction module by adding a Self-Attention and Convolution mixed module (ACmix) to account for the uneven spatial distribution and target scale variation of the cows. Additionally, in order to reduce the number of model parameters, an improved lightweight Spatial Pyramid Pooling Cross Stage Partial Connections (SPPCSPC-L) module is adopted to reduce model complexity. At the same time, the state parameters in the Kalman filter are improved by directly predicting the width and height information of the tracking boxes, so as to improve the ByteTrack algorithm to make tracking boxes matching the cows more precisely and accurately. Experimental conducted on the dairy cow object detection and multi-object tracking dataset show that the proposed YOLO-BYTE model achieves a Precision (P) of 97.3% in the dairy cow target detection dataset, with an improved Recall (R) and Average Precision (AP) by 1.1% compared to the original algorithm, and an 18% reduction in model parameters. Moreover, the proposed method demonstrated significant improvements in High Order Tracking Accuracy (HOTA), Multi-Object Tracking Accuracy (MOTA), and Identification F1 (IDF1) by 4.4%, 6.1%, and 3.8%, respectively, compared to the original model, with a decrease of 37.5% in Identity Switch (IDS). The tracker runs in a real-time manner with an average analysis speed of 47 fps. Hence, it is demonstrated that the proposed approach is capable of effective multi-object tracking of dairy cows in natural scenes and provides technical support for non-contact dairy cow automatic monitoring. [ABSTRACT FROM AUTHOR]

Agents with intelligent perceptual capabilities are considered state-of-the-art in advanced intelligent systems. In addition, a multiagent system is considered an enabling technology for applications that rely on distributed and parallel processing, including data, information and knowledge in complex computing environments. With the aim of creating advanced intelligent systems with visual perception, this paper presents an agent-based framework to individual tracking in unconstrained environments. The framework has three types of agents that interact using the Contract Net Protocol. The face detector and tracker agents perform fully automatic single-sample face recognition using the Viola-Jones and the Scale Invariant Feature Transform/Speeded Up Robust Features algorithms. The experimental results show that the framework adequately recognizes and tracks individuals in unconstrained environments, indicating the path the individuals have taken and the time they spent in the field of view of the surveillance agents. Some of the open source framework advantages are the distribution in heterogeneous infrastructures, the expansion with new agents using different face recognition algorithms (e.g., Eigenfaces), and the individual tracking logs that can be used in different ways, e.g., improve security in surveillance areas such as automated teller machines, self-paying kiosks, movie box offices, and malls. [ABSTRACT FROM AUTHOR]

The paper presents a methodology for assigning passengers to individual trains using: (i) fare transaction records from Automatic Fare Collection (AFC) systems and (ii) Automatic Vehicle Location (AVL) data from train tracking systems. The proposed Passenger-to-Train Assignment Model (PTAM) is probabilistic and links each fare transaction to a set of feasible train itineraries. The method estimates the probability of the passenger boarding each feasible train, and the probability distribution of the number of trains a passenger is unable to board due to capacity constraints. The access/egress time distributions are important inputs to the model. The paper also suggests a maximum likelihood approach to estimate these distributions from AFC and AVL data. The methodology is applied in a case study with data from a major, congested, subway system during peak hours. Based on actual AFC and train tracking data, synthetic data was generated to validate the model. The results, both in terms of the trains passengers are assigned to and train loads, are similar to the “true” observations from the synthetic data. The probability of a passenger being left behind (due to capacity constraints) in the actual system is also estimated by time of day and compared with survey data collected by the agency at the same station. The left behind probabilities can be accurately estimated from the assignment results. Furthermore, it is shown that the PTAM output can also be used to estimate crowding metrics at transfer stations. [ABSTRACT FROM AUTHOR]

For route planning and tracking, it is sometimes necessary to know if the user is walking or using some other mode of transport. In most cases, the GPS data can be acquired from the user device. It is possible to estimate user’s transportation mode based on a GPS trace at a sampling rate of once per minute. There has been little prior work on the selection of a set of features from a large number of proposed features, especially for sparse GPS data. This article considers characteristics of distribution, auto- and cross-correlations, and spectral features of speed and acceleration as possible features, and presents an approach to selecting the most significant, non-correlating features from among those. Both speed and acceleration are inferred from changes in location and time between data points. Using GPS traces of buses in the city of Tampere, and of walking, biking and driving from the OpenStreetMap and Microsoft GeoLife projects, spectral bins were found to be among the most significant non-correlating features for differentiating between walking, bicycle, bus and driving, and were used to train classifiers with a fair accuracy. Auto- and cross-correlations, kurtoses and skewnesses were found to be of no use in the classification task. Useful features were found to have a fairly large (>0.4) correlation with each other. [ABSTRACT FROM AUTHOR]

Truck flow patterns are known to vary by season and time-of-day, and to have important implications for freight modeling, highway infrastructure design and operation, and energy and environmental impacts. However, such variations cannot be captured by current truck data sources such as surveys or point detectors. To facilitate development of detailed truck flow pattern data, this paper describes a new truck tracking algorithm that was developed to estimate path flows of trucks by adopting a linear data fusion method utilizing weigh-in-motion (WIM) and inductive loop point detectors. A Selective Weighted Bayesian Model (SWBM) was developed to match individual vehicles between two detector locations using truck physical attributes and inductive waveform signatures. Key feature variables were identified and weighted via Bayesian modeling to improve vehicle matching performance. Data for model development were collected from two WIM sites spanning 26 miles in California where only 11 percent of trucks observed at the downstream site traversed the whole corridor. The tracking model showed 81 percent of correct matching rate to the trucks declared as through trucks from the algorithm. This high accuracy showed that the tracking model is capable of not only correctly matching through vehicles but also successfully filtering out non-through vehicles on this relatively long distance corridor. In addition, the results showed that a Bayesian approach with full integration of two complementary detector data types could successfully track trucks over long distances by minimizing the impacts of measurement variations or errors from the detection systems employed in the tracking process. In a separate case study, the algorithm was implemented over an even longer 65-mile freeway section and demonstrated that the proposed algorithm is capable of providing valuable insights into truck travel patterns and industrial affiliation to yield a comprehensive truck activity data source. [ABSTRACT FROM AUTHOR]

Background Real-time recording and manipulation of neural activity in freely behaving animals can greatly advance our understanding of how neural circuits regulate behavior. Ca 2+ imaging and optogenetic manipulation with optical probes are key technologies for this purpose. However, integrating the two optical approaches with behavioral analysis has been technically challenging. New method Here, we developed a new imaging system, ICaST ( I ntegrated platform for Ca 2+ imaging, S timulation, and T racking), which combines an automatic worm tracking system and a fast-scanning laser confocal microscope, to image neurons of interest in freely behaving C. elegans. We optimized different excitation wavelengths for the concurrent use of channelrhodopsin-2 and G-CaMP, a green fluorescent protein (GFP)-based, genetically encoded Ca 2+ indicator. Results Using ICaST in conjunction with an improved G-CaMP7, we successfully achieved long-term tracking and Ca 2+ imaging of the AVA backward command interneurons while tracking the head of a moving animal. We also performed all-optical manipulation and simultaneous recording of Ca 2+ dynamics from GABAergic motor neurons in conjunction with behavior monitoring. Comparison with existing method(s) Our system differs from conventional systems in that it does not require fluorescent markers for tracking and can track any part of the worm’s body via bright-field imaging at high magnification. Consequently, this approach enables the long-term imaging of activity from neurons or nerve processes of interest with high spatiotemporal resolution. Conclusion Our imaging system is a powerful tool for studying the neural circuit mechanisms of C. elegans behavior and has potential for use in other small animals. [ABSTRACT FROM AUTHOR]

This paper presents three state estimators grounded in the variable-stepsize Gauss- and Lobatto-type Nested Implicit Runge–Kutta (NIRK) formulas of orders 4 and 6 and designed for treating continuous-time stochastic systems arisen in radar tracking. Our filters are built within the Extended Kalman Filtering (EKF) framework and based on accurate numerical integrations of the corresponding Moment Differential Equations (MDEs). Automatic local and global error regulation mechanisms implemented in these methods allow the committed discretization error to be under control and made negligible in automatic mode. The latter raises the state estimation accuracy of the constructed filters, significantly. This also leads to the advanced notion of Accurate Continuous–Discrete Extended Kalman Filtering (ACD-EKF) developed by Kulikov and Kulikova in 2013–2016. Our novel methods are constructed within the same approach, but possess the improved accuracy and efficiency in comparison to their predecessors due to both more effective error control mechanisms implemented for integrating MDEs and more accurate iterations used for treating arisen nonlinear equations in the revised filters. Numerical experiments with the updated state estimators and their comparison to the cited earlier-designed ACD-EKFs are fulfilled in severe conditions of tackling a seven-dimensional radar tracking problem, where an aircraft executes a coordinated turn, in Matlab. This examination suggests that the novel state estimation algorithms outperform their predecessors and possess a promising potential for solving target tracking tasks in real-world applications. [ABSTRACT FROM AUTHOR]

This paper proposes a new approach to archive the robust tracking of moving objects under moving camera environment where the similar moving objects cross each other. Tracking with moving camera sometimes fails to track the object with similar color objects or similar background. Proposed approach is a particle filter based approach. It introduces the likelihood calculated by probabilistic background model which is constructed using dense optical flow and fast density estimation. Proposed approach introduces SVM (Support Vector Machine) 1 to judge the scene where it is difficult to construct the probabilistic background model with non-uniform optical flow. This SVM uses the degree histogram of optical flow. Usefulness of proposed approach is evaluated in the experiments using actual video and the performance is compared with recent tracking approaches by quantitative evaluations. [ABSTRACT FROM AUTHOR]

This study exhibits a self-tuning speed control scheme for the surface-mounted permanent magnet synchronous motor (SPMSM) against the parameter variations through the multivariable approach. The proposed method has two novelties. The first one is to combine the adaptive controller with the self-tuning algorithm so as to make the decay ratio of the tracking errors higher in the transient period. The second one is to provide a systematical way to find a robust stabilizing control gain corresponding the speed and current tracking errors by solving an optimization problem. The efficacy of the proposed method was experimentally investigated using a 3- kW SPMSM. [ABSTRACT FROM AUTHOR]

Background and purpose To report the clinical usefulness of coiled fiducial markers as an internal surrogate in gated stereotactic body radiotherapy (SBRT) for hepatocellular carcinoma (HCC) using a real-time tumor-tracking radiotherapy (RTRT) system. Materials and methods Seventeen HCC patients with Child-Pugh (CP) scores of A or B received gated SBRT (45–50 Gy in 5–10 fractions) using an RTRT system and Visicoil markers. Local control (LC), progression-free (PFS), and overall survival (OS) rates were assessed using the Kaplan-Meier method. Toxicities were assessed using the Common Terminology Criteria for Adverse Events, Version 4.0. Results Of the 17 patients, 14 had a CP score A. The mean planning target volume was 54.6 cc. Only 1 patient developed pneumothorax after marker implantation. Visicoil tracking during SBRT was possible in all cases. With a median follow-up of 16 months, 1-year LC, PFS, and OS rates were 100%, 53%, and 82%, respectively. Grade ≥ 2 late toxicity was observed in 2 patients (grade 2 duodenal ulcer and grade 3 temporary transaminase elevation). Conclusions Using an RTRT system and Visicoil markers, gated SBRT was well tolerated in patients with HCC. This can be considered a safe treatment strategy with potential for delivering favorable outcomes. [ABSTRACT FROM AUTHOR]

Robot-based tracking of spontaneous respiration-triggered motion of human tissue becomes increasingly important in robot-guided imaging, interventional radiology, surgery and radiotherapy. This work provides a first general comparative assessment of the tracking performances that can potentially be achieved by common control schemes under realistic conditions. Cartesian exemplary spontaneous respiratory motions of 185 min duration were recorded from a healthy male subject and transformed into reference trajectories for tracking experiments. Prediction error was modeled by random perturbation of these trajectories with increasing amplitude over the prediction horizon. The controlled system, an industrial robot, was represented by an identified ARX model. A repetitive (RC), adaptive (ALC), feedforward (FFC) and model-based predictive controller (MPC) were implemented. ALC adaptively identifies robot latency and compensates this latency by according time-shift of reference coordinates. Control performance was assessed based on the control error as well as this errors’ estimated worst case dosimetric consequence in a radiotherapy application. Deviations from periodicity in reference trajectories were statistically quantified and reproduced for simulation-based assessment of RC. ALC, FFC and MPC were assessed experimentally. Realistic deviations from periodicity made performance of RC unprofitable, even with ideal function of controller and period-robustness techniques. The same applies potentially to related methods such as iterative learning control. ALC, FFC and MPC performed comparably for ideal prediction. Despite realistic prediction error, FFC and MPC achieved submillimeter control error, significantly outperforming ALC. Thus, FFC or MPC should be employed, RC and ALC should not be employed in robot-based tracking of spontaneous respiration-triggered motion of human tissue. [ABSTRACT FROM AUTHOR]

An automatic track following system has been successfully developed to follow tracks in nuclear emulsion sheets exposed with beam up to the limit to be observed for the first time. The track followed rate of the system is 99.5% with the assistance of the new techniques. The working speed for a track is less than 1 min through one thick emulsion sheet, whereas it is 15 times faster than that of semiautomatic system with human. The system working for 24 h is applied for the E07 experiment at J-PARC and makes it possible to detect ~10 2 nuclei with double strangeness (S=−2 nuclei) within one year. Regarding analyses to identify nuclear species of S=−2 nuclei, the system shows quite decent job for significant steps such as following tracks emitted to spherical directions from S=−2 nuclei, measurement of lengths of followed tracks, and so on. [ABSTRACT FROM AUTHOR]

Imitating the behaviors of an arbitrary visual tracking algorithm enables many higher level tasks such as tracker identification and efficient tracker-fusion. It is also useful for discovering the features essential in a black-box tracker or learning from several trackers to form a super-tracker. In this study, we propose a non-linear feature fusion framework, “ MIMIC ” that imitates many popular trackers by mixing a pool of heterogeneous features. The MIMIC framework consists of two subtasks, feature selection and feature weight tuning. These subtasks, however, tended to suffer from an overfitting problem when the number of videos available for training is limited. To address this issue, we incorporated Dropout algorithm into the training, which grants the trained MIMIC tracker a high degree of generalization. Extensive experiments testified the effectiveness of the proposed framework so that its applications would be promoted into different related tasks in visual tracking. [ABSTRACT FROM AUTHOR]