Search

Traditional displacement sensors and computer vision technologies exhibit significant limitations when measuring multiple targets, including restrictions related to the measurement environment, manual tagging and the wiring issues associated with multiple sensors. This paper proposes a method for multi-point vibration displacement measurement that integrates bicubic interpolation, sparse optical flow and oriented FAST and rotated BRIEF (ORB) detection algorithms, addressing the aforementioned limitations. The proposed method eliminates the need for manual tags and the deployment of multiple cameras or sensors. By incorporating bicubic interpolation, the accuracy of the measurement is enhanced. Feature points are selected from the initial frame of a video and the sparse optical flow method is utilised to measure structural vibration displacement. Finally, the true displacement is obtained using the proportional factor method. This method allows for the displacement measurement of multiple feature points within the same frame, achieving synchronous multi-point displacement measurement. Two validation experiments are presented: the first involves mounting an elastic structure on a vibration table to compare the displacement response data obtained using the proposed method with that measured by traditional accelerometers; the second experiment focuses on testing a spring from a grinding machine, emphasising the simultaneous measurement of multi-point vibration displacements of the spring structure. The results from both experiments indicate that the proposed method is capable of synchronously measuring the vibration displacements of multiple targets, with accuracy that meets the requirements for practical applications. [ABSTRACT FROM AUTHOR]

The ORB (oriented FAST and rotated BRIEF) feature detection algorithm is shown to have significant variances in the number of extracted feature points and the accuracy of matching in blurry scenes and environments with drastic lighting changes. Additionally, feature points tend to cluster at the corners of image objects. In response to these issues, an improved ORB feature detection algorithm was proposed. Firstly, the multi-scale retinex (MSR) algorithm was employed to enhance the features of the image. Subsequently, the image was divided into grids, and the threshold for feature point detection was adjusted based on the grayscale distribution of each grid. Finally, a dynamic regional non-maximal suppression method was adopted to filter the optimal feature points. The experimental results indicate that, compared to the original ORB algorithm, the distribution of feature points extracted by the improved algorithm is more uniform across images. The repeatability rate of feature points remains stable at over 75% when the brightness varies within a range of 80%, and the accuracy of matching is on average increased by 22% . [ABSTRACT FROM AUTHOR]

A postmortem iris recognition system is a specialized biometric identification system used to identify individuals based on their iris patterns after death. Such a system requires a set of specific requirements to ensure its effectiveness and accuracy. So, with this objective, this paper proposes a postmortem iris recognition system approach, which aims to accurately identify individuals based on their iris patterns even after death. The approach utilizes a specialized dataset of postmortem cases, including careful extraction of the iris and accounting for postmortem changes. In addition, the proposed system uses handcrafted features and deep learning techniques to enhance identification accuracy. The system has been tested on a dataset of postmortem iris images, and results show high accuracy in identifying individuals. This approach has potential applications in forensic investigations, disaster victim identification, and identification of remains in mass graves. In addition, the proposed system offers a valuable tool for identifying individuals even in challenging postmortem cases, contributing to biometrics and forensic science. [ABSTRACT FROM AUTHOR]

The present paper aims to conduct an experiment that compares different methods of detecting objects in images. Programs were developed to evaluate the efficiency of SURF, BRISK, MSER, and ORB object detection methods. Four static gray images with sufficiently different histograms were used. The experiment also highlighted the need for image preprocessing to improve feature extraction and detection. Thus, a programmed method for adjusting pixel groups was developed. This method proved useful when one of the listed algorithms failed to detect the object in the original image, but succeeded after adjustment. The effectiveness of detection methods and the evaluation of their performance depend on the application, image preparation, algorithms used, and their implementation. Results of the detection methods were presented numerically (similarities, gradients, distances, etc.) and graphically.

Industrial photogrammetry is a reliable method to achieve submillimeter accuracy when mapping 2D or 3D objects. In the field of fatigue testing of steel welded details, it can be used to find a new method of crack initiation assessment. Fatigue testing is an important method for determining and predicting the durability of structural details in service. The research presented in this paper is based on a computer vision algorithm developed using the open-source code OpenCV library and the Oriented FAST and Rotated BRIEF (ORB) method to provide a solution for the assessment of crack initiation. Within this research, a method for determining the crack initiation period using polynomial functions of a certain degree is developed. The developed algorithm fully automatically determines the test specimen displacement for all imagery and assesses the crack initialization period by polynomial interpolation with a percentage threshold. The algorithm shows us the best results based on a 26th-degree polynomial with a deviation from the critical value of 5%. The validation of the algorithm was carried out using completely independently recorded data from the hydraulic press used for fatigue tests. The results of all test specimens show that the percentage accuracy of determination crack initiation period is between −0.04% for test specimens S355-TA-AW-02 and S355-TA-HFMI-03 and −0.82% for test specimen S355-TA-HFMI-03, with the mean of all results being 0.39%. [ABSTRACT FROM AUTHOR]

The present paper aims to conduct an experiment that compares different methods of detecting objects in images. Programs were developed to evaluate the efficiency of SURF, BRISK, MSER, and ORB object detection methods. Four static gray images with sufficiently different histograms were used. The experiment also highlighted the need for image preprocessing to improve feature extraction and detection. Thus, a programmed method for adjusting pixel groups was developed. This method proved useful when one of the listed algorithms failed to detect the object in the original image, but succeeded after adjustment. The effectiveness of detection methods and the evaluation of their performance depend on the application, image preparation, algorithms used, and their implementation. Results of the detection methods were presented numerically (similarities, gradients, distances, etc.) and graphically. [ABSTRACT FROM AUTHOR]

Computer vision is a catch-all term for a variety of applications. This makes it a good research environment for new ideas and concepts. Feature extraction is considered an essential step in such applications. Several research studies introduced the Scale-Invariant Feature Transform (SIFT) as a benchmark method to extract visual features from objects inside images. This ensures the need for a deep study of SIFT in a variety of settings. Hence, this paper presents an assessment of SIFT from different perspectives that are not explicitly expressed in the literature. In addition, it presents an illustration of the majority of Oriented FAST and Rotated BRIEF (ORB) feature extraction characteristics to facilitate the choice procedure between SIFT and ORB. Several experimental cases are included, each of which evaluates the performance of such methods from distinct and different aspects. At first, the paper presents an assessment of these methods to identify objects inside geometrically–affine-transformed images. This is done by comparing how well their gathered feature descriptors from images perform against one another. Second, this paper presents an evaluation of the invariance of these methods to the changes in illumination. Furthermore, the computational and asymptotic complexity of such methods is investigated to examine its impact on the complexity of any feature-based system. Finally, the efficiency of these methods is verified by assessing their ability to support real-time applications, through the evaluation of their time and space complexities over all investigated test scenarios. [ABSTRACT FROM AUTHOR]

Pedestrian detection is a vital aspect of Advanced Driver Assistance Systems (ADAS), crucial for ensuring driving safety and minimizing collision risks. While detecting pedestrians is important, it must be paired with precise distance estimation to create a robust safety solution. Stereovision cameras are well-regarded for their effectiveness and affordability in measuring depth through disparity between two images. Despite this, research on pedestrian distance estimation using only stereovision remains sparse, with many studies relying on computationally heavy dense depth maps. This paper proposes an innovative method for computing object-level disparity specifically for pedestrian detection using stereo cameras. The approach integrates Canny edge detection with ORB (Oriented FAST and Rotated BRIEF) feature matching to efficiently identify and track keypoints within pedestrian bounding boxes. This method not only improves the accuracy of distance estimation but also reduces computational demands, making it suitable for real-time applications. The approach was thoroughly tested on a Raspberry Pi 4, a resource-constrained device, and achieved promising results, demonstrating its potential for practical use in ADAS.

In this work, an optimal control scheme based on the Maximum Power Point tracking for the Wind Energy Conversion System using Permanent Magnet Synchronous Generator (PMSG) is proposed and also modeled. The system studied in this paper consists of wind energy system powering a battery using a buck-boost converter as an interface. By modifying the buck-boost duty cycle, we vary the reflected voltage at PMSG, and accordingly its speed. It is controlled with Perturb and Observe (P&O) Maximum Power Point Tracking (MPPT) approach. The overall system is simulated with MATLAB/SIMULINK. [ABSTRACT FROM AUTHOR]

In the fruit fly Drosophila melanogaster, two cells in a cyst of 16 interconnected cells have the potential to become the oocyte, but only one of these will assume an oocyte fate as the cysts transition through regions 2a and 2b of the germarium. The mechanism of specification depends on a polarized microtubule network, a dynein dependent Egl:BicD mRNA cargo complex, a special membranous structure called the fusome and its associated proteins, and the translational regulator orb. In this work, we have investigated the role of orb and the fusome in oocyte specification. We show here that specification is a stepwise process. Initially, orb mRNAs accumulate in the two pro-oocytes in close association with the fusome. This association is accompanied by the activation of the orb autoregulatory loop, generating high levels of Orb. Subsequently, orb mRNAs become enriched in only one of the pro-oocytes, the presumptive oocyte, and this is followed, with a delay, by Orb localization to the oocyte. We find that fusome association of orb mRNAs is essential for oocyte specification in the germarium, is mediated by the orb 3′ UTR, and requires Orb protein. We also show that the microtubule minus end binding protein Patronin functions downstream of orb in oocyte specification. Finally, in contrast to a previously proposed model for oocyte selection, we find that the choice of which pro-oocyte becomes the oocyte does not seem to be predetermined by the amount of fusome material in these two cells, but instead depends upon a competition for orb gene products. [ABSTRACT FROM AUTHOR]

Due to its ability to capture natural light without emitting additional signals, a stereo camera is a promising option for low-power visual odometry (VO). However, achieving real-time processing of VO using a stereo camera on embedded CPUs and GPUs is challenging due to the high computational demands and irregular computational patterns. These irregular patterns arise from the dynamic and non-uniform nature of feature detection, stereo matching, and map management, which vary significantly with scene complexity and camera movement. Various hardware accelerators have been proposed to address this challenge, but they typically tackle individual aspects of stereo VO, leading to difficulties in combining them efficiently. This paper introduces a comprehensive stereo VO accelerator implemented on an AMD Kria KV260 FPGA board, aimed at improving processing efficiency and reducing energy consumption. Our accelerator features innovative techniques including on-FPGA map management and pipelined stereo matching with the use of descriptors instead of image patches. These advancements result in significant reduction in off-chip data transfer, memory savings, and faster processing times. Additionally, an adaptive bucketing-based feature selection method is employed to enhance system accuracy without significant increases in hardware resource usage. Evaluation using the KITTI dataset shows that our accelerator achieves speedups of up to $3.08\times $ and $2.75\times $ over CPU-only and CPU+GPU implementations of LVT, an optimized algorithm, on an NVIDIA Jetson Nano developer kit B01 4GB, with corresponding energy efficiency improvements of $3.55\times $ and $1.95\times $ . The results demonstrate the accelerator’s effectiveness for real-time applications, highlighting the benefits of FPGA-based solutions in complex visual processing tasks.

Copyright of Telecommunication Engineering is the property of Telecommunication Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

3′ untranslated regions (3′ UTRs) of mRNAs have many functions, including mRNA processing and transport, translational regulation, and mRNA degradation and stability. These different functions require cis‐elements in 3′ UTRs that can be either sequence motifs or RNA structures. Here we review the role of secondary structures in the functioning of 3′ UTRs and discuss some of the trans‐acting factors that interact with these secondary structures in eukaryotic organisms. We propose potential participation of 3′‐UTR secondary structures in cytoplasmic polyadenylation in the model organism Drosophila melanogaster. Because the secondary structures of 3′ UTRs are essential for post‐transcriptional regulation of gene expression, their disruption leads to a wide range of disorders, including cancer and cardiovascular diseases. Trans‐acting factors, such as STAU1 and nucleolin, which interact with 3′‐UTR secondary structures of target transcripts, influence the pathogenesis of neurodegenerative diseases and tumor metastasis, suggesting that they are possible therapeutic targets. [ABSTRACT FROM AUTHOR]

Objectives:This research presents an approach to detect occluded face images. In order to achieve this, the research presents a novel technique that involves feature extraction and occluded face recognition. Methods: Feature extraction is performed by the enhanced ORB algorithm, which is proposed by the modification of the Oriented Fast and Rotated Brief (ORB) algorithm, by adding a phase for contrast adjustment, together with CNN features. For occluded face recognition, a Generative Adversarial Network (GAN) optimized by the proposed SR-SSA is designed. SR-SSA is proposed by the integration of Search and Rescue Optimization (SAR) in the Sparrow Search Algorithm (SSA). Results: The experimental results demonstrate that the SR-SSA-based GAN algorithm outperforms existing methods in terms of accuracy of 0.956, FAR of 0.045 and FRR of 0.021. [ABSTRACT FROM AUTHOR]

Owing to the impact of vibration on the carrier of a vehicle-mounted camera, video is shaking, resulting in decreased or failed recognition accuracy based on visual-target detection. To solve this problem, a video stabilization algorithm based on grid motion statistics and an adaptive Kalman filter is proposed. Two important processes in video stabilization are motion estimation and motion smoothing. In the motion estimation stage, we adopt an erroneous matching removal algorithm that integrates grid motion statistics (GMS) to enhance the accuracy of motion estimation while reducing the matching time, further meeting the real-time and precision requirements of vehicle-mounted video stabilization. In the motion smoothing stage, we adaptively update the measurement noise covariance R in the adaptive Kalman filter based on the camera shake level, further improving the accuracy of motion smoothing under the condition of ensuring filter convergence. Finally, we compensate for the motion based on the relationship between the pre- and postsmooth motion trajectories, generating a stable video sequence. Experimental results demonstrate that the proposed algorithm exhibits good stability and effectiveness in vehicle-mounted video stabilization. [ABSTRACT FROM AUTHOR]

Search by image has taken the place of the prior wordy approach, considering spatial image comparison in the feature space. Several query image forms are available. Sketch-to-real image matching is very challenging problem, which is commonly based on edge feature detectors to extract features from sketches. However, they are inconvenient, and it is crucial to select different methods. Thus, methods assessment is conducted for similarity matching between sketched and real images for different drawing contents and clarity levels. The assessment approach includes comparing feature keypoints gathered using the Scale-Invariant Feature Transform (SIFT) and Oriented FAST and Rotated BRIEF (ORB) methods. In addition, the comparison includes matching of objects inside High-Definition (HD) images and 3D images based on their generated features. The textual-Hcontent image search is another search type that is frequently used in brand logo cases. Hence, it is presented for text-logo image search based on visual features. Besides, the motives and downsides of the considered methods are declared, including the space and time complexities within test cases. [ABSTRACT FROM AUTHOR]

Feature extraction and matching are the basic procedures of the so-called Visual Odometer (VO), Simultaneous Localization and Mapping (SLAM) and many other image processing algorithms. Oriented features from accelerated segment test (FAST) and rotated binary robust independent elementary features (ORB) algorithm are widely used since they are computationally faster. In this paper, we proposed a method to generate a value for a feature, the value is called signature. In the matching step, we only compute Hamming distances of descriptors with the same signatures. Hence, the matching time is shortened. Compared with the original ORB algorithm, features to be matched dropped 69.63% on TUM datasets and 85.7% on VGG datasets by adopting our strategy. In addition, the precision is above 85% on both VGG and TUM datasets. We design a customized hardware architecture for ORB feature extraction and matching based on the proposed method. The hardware structure is implemented on Xilinx ZCU102 evaluation board. The clock frequency is set to 150 MHz. Our Field Programmable Gate Arrays (FPGA) system achieves 193 fps on 1 2 8 0 × 7 2 0 images with 1984 features on average and 314 fps on 6 4 0 × 4 8 0 images with 700 features on average, which is more efficient compared to the state-of-the-art works. [ABSTRACT FROM AUTHOR]

Industrial photogrammetry is a reliable method to achieve submillimeter accuracy when mapping 2D or 3D objects. In the field of fatigue testing of steel welded details, it can be used to find a new method of crack initiation assessment. Fatigue testing is an important method for determining and predicting the durability of structural details in service. The research presented in this paper is based on a computer vision algorithm developed using the open-source code OpenCV library and the Oriented FAST and Rotated BRIEF (ORB) method to provide a solution for the assessment of crack initiation. Within this research, a method for determining the crack initiation period using polynomial functions of a certain degree is developed. The developed algorithm fully automatically determines the test specimen displacement for all imagery and assesses the crack initialization period by polynomial interpolation with a percentage threshold. The algorithm shows us the best results based on a 26th-degree polynomial with a deviation from the critical value of 5%. The validation of the algorithm was carried out using completely independently recorded data from the hydraulic press used for fatigue tests. The results of all test specimens show that the percentage accuracy of determination crack initiation period is between −0.04% for test specimens S355-TA-AW-02 and S355-TA-HFMI-03 and −0.82% for test specimen S355-TA-HFMI-03, with the mean of all results being 0.39%.

Image registration plays a vital role in the mosaic process of multiple UAV (Unmanned Aerial Vehicle) images acquired from different spatial positions of the same scene. Aimed at the problem that many fast registration methods cannot provide both high speed and accuracy simultaneously for UAV visible light images, this work proposes a novel registration framework based on a popular baseline registration algorithm, ORB—the Oriented FAST (Features from Accelerated Segment Test) and Rotated BRIEF (Binary Robust Independent Elemental Features) algorithm. First, the ORB algorithm is utilized to extract image feature points fast. On this basis, two bidirectional matching strategies are presented to match obtained feature points. Then, the PROSRC (Progressive Sample Consensus) algorithm is applied to remove false matches. Finally, the experiments are carried out on UAV image pairs about different scenes including urban, road, building, farmland, and forest. Compared with the original version and other state-of-the-art registration methods, the bi-matching ORB algorithm exhibits higher accuracy and faster speed without any training or prior knowledge. Meanwhile, its complexity is quite low for on-board realization. [ABSTRACT FROM AUTHOR]

Ocular diseases are a prevalent disease among the aging population across the world. The retinal damage and vision loss can be substantially decreased through early-stage diagnosis with computer-aided ocular disease diagnosis. With the use of color fundus photography for obtaining digital retinal fundus images, there is a growth in the online accessibility of digital fundus images. For diagnosing ocular diseases, an attempt was made to model graphs from images for feature learning. Three feature detection algorithms, namely scale-invariant feature transform, binary robust invariant scalable keypoints and oriented fast and rotated BRIEF (ORB) techniques are computed individually. As graphs are represented in the non-Euclidean domain, the graph neural network is used to learn the node embedding to model the network for ocular disease diagnosis. Three distance measures: the Euclidean, Manhattan and Chebyshev distances, are computed for analyzing the discriminative power of the model. The proposed RDD-Net model is trained and evaluated on the ODIR-2019 dataset with eleven different performance indicators. The results show that mapping images to non-Euclidean geometric space have obtained a successful diagnosis of ocular diseases from digital fundus images. The ORB descriptor outperforms the other two feature descriptors as well as the existing algorithms for ocular disease diagnosis. Results of the Chebyshev distance measure show superior performance when compared to the other two distance measures based on computation time and performance evaluation metrics. The proposed RDD-Net achieves an F1-score of 0.9970 and a sensitivity of 0.9969 with the ORB descriptor and shows state-of-the-art performance. [ABSTRACT FROM AUTHOR]

The ORB (Oriented fast and rotated brief, ORB) algorithm is limited by its inability to extract feature points or extract only a small number of feature points when a fixed threshold is used in complex lighting conditions. To address this issue, this paper proposes a method that combines image enhancement with truncated adaptive threshold to improve the ORB feature extraction algorithm. Firstly, the original image is converted to grayscale. Secondly, the image is enhanced by applying Gaussian filtering for noise reduction, truncated adaptive gamma brightness adjustment, and unsharp masking operation. Then, the enhanced image is segmented into subregions of a specified size, and an improved truncated OTUS method is employed to calculate the adaptive threshold for each subregion. Finally, ORB feature points are extracted by utilizing the adaptive threshold. Experimental results show that the improved ORB algorithm can significantly increase the number of feature points in complex lighting conditions, with good accuracy, real-time performance, and robustness.

Visual odometer is an important part of SLAM system. Visual odometer based on feature points is the current mainstream. Featured based method completes the positioning by extracting feature points and matching descriptors. However, under low texture conditions, there are few features to extract, and the cost of matching descriptors is also high. to solve this problem, we propose a visual odometer based on Lucas-Kanade(LK) optical flow method and ORB feature. The method improves the performance of the ORB algorithm and acquires more feature points in a low-texture environment. A coarse-to-fine LK method is proposed for robustly and efficiently matching feature points between two frames. Based on this, a semi-direct visual odometry is proposed to improve the performance of visual odometry in low-texture environments. The main contribution of this method is that we introduce the direct method in the feature point-based approach, which significantly improves the localization accuracy in low-texture environments. The results show that the method can accomplish localization in low-texture environments without closed-loop. Experimental evaluation on two benchmark datasets shows that the method has higher accuracy and robustness in motion estimation compared to other state-of-the-art methods. We have done strict tests on EuRoC and TUM data sets, and the results show that our visual odometer has higher accuracy and robustness in low-texture environment than the state-of-art feature-based scheme.

To improve the accuracy of feature extraction and description of various scales in traditional Oriented FAST and Rotated BRIEF (ORB) feature matching algorithm, this paper proposes an ORB feature matching algorithm based on multi-scale feature description fusion and feature point mapping error correction. Firstly, when establishing the feature scale pyramid, the method of using the same patch-size for description on each feature layer is adopted instead of using different patch-sizes on a unified feature layer in the original algorithm, which enhances the robustness of the descriptor and improves the matching accuracy. Secondly, FAST-SCORE maps are established on different scales, and the coordinates of high-level feature points mapped to the bottom layer are corrected to further improve the positioning accuracy of feature points. The algorithm is verified in remote sensing images, autonomous driving, and industrial automation fields. Experimental results show that when resisting theoretical interference, the average matching accuracy of the proposed algorithm is 67.9%, which is about 2.0 times that of the ORB algorithm, and the average stability is 14.0, which is about 1.5 times that of the ORB algorithm. After correcting the feature point mapping, the matching accuracy can be further improved by 19.2%, indicating that the improved algorithm has excellent robustness when resisting interference. In the experiments on the KITTI and custom datasets, the matching accuracy of the proposed algorithm reached 88.70% and 96.88%, respectively, which is an improvement of 10.15% and 1.2% compared to the ORB algorithm. At the same time, the matching time was reduced by 17.34% and 24.30%, ensuring the accuracy and real-time performance of the algorithm in practical scenarios.

An electron spin echo in a nitroxide-containing polymer cathode film for organic radical batteries is observed for various states of charge at cryogenic temperatures. The EPR-detected state of charge (ESOC), as inferred from the number of paramagnetic centers in the film, is compared to the results of Coulomb counting based on galvanostatic charging. Spin concentration, longitudinal relaxation times T1 and phase memory times Tm strongly correlate with the ESOC. In the discharged film, the spin concentration reaches 5±3×1020 cm−3, causing a phase memory time Tm≪ 100 ns (shorter than the resonator ring-down time) that hinders the detection of the spin echo. In the charged film, the decreased spin concentration results in a longer Tm between 100 ns and 300 ns that enables spin-echo detection, yet limits the length of the microwave pulse sequence. The short, broad-band pulses cause instantaneous diffusion in the unoxidized domains across the oxidized film, affecting the relative peak intensities in the pulsed EPR spectrum. By simulating the spectral distortion caused by instantaneous diffusion, we obtain information on the local spin concentration, which complements the information on the ‘bulk’ spin concentration determined by electrochemistry and continuous-wave EPR spectroscopy.

Product recognition on grocery shelf images is a compelling task of object detection because of the similarity between products, the presence of the different scale of product sizes, and the high number of classes, in addition to constantly renewed packaging and added new products' difficulty in data collection. The use of conventional methods alone is not enough to solve a number of retail problems such as planogram compliance, stock tracking on shelves, and customer support. The purpose of this study is to achieve significant results using the suggested multi-stage end-to-end process, including product detection, product classification, and refinement. The comparison of different methods is provided by a traditional computer vision approach, Aggregate Channel Features (ACF) and Single-Shot Detectors (SSD) are used in the product detection stage, and Speed-up Robust Features (SURF), Binary Robust Invariant Scalable Key points (BRISK), Oriented Features from Accelerated Segment Test (FAST), Rotated Binary Robust Independent Elementary Features (BRIEF) (ORB), and hybrids of these methods are used in the product classification stage. The experimental results used the entire Grocery Products dataset and its different subsets with a different number of products and images. The best performance was achieved with the use of SSD in the product detection stage and the hybrid use of SURF, BRISK, and ORB in the product classification stage, respectively. Additionally, the proposed approach performed comparably or better than existing models. [ABSTRACT FROM AUTHOR]

Labeling landmarks for the mobile plan of the automatic electric wheelchair is essential, because it can assist disabled people. In particular, labeled landmark images will help the wheelchairs to locate landmarks and move more accurately and safely. Here, we propose an automatic detection of natural landmarks in RGBD images for navigation of mobile platforms in an indoor environment. This method can reduce the time for manually collecting and creating a dataset of landmarks. The wheelchair, equipped with a camera system, is allowed to move along corridors to detect and label natural landmarks automatically. These landmarks contain the camera and wheelchair positions with the 3D coordinates when storing the labeled landmark. The feature density method is comprised of Oriented FAST and Rotated BRIEF (ORB) feature extractors. Moreover, the central coordinates of the marked points in the obtained RGB images will be mapped to the images with the depth axis for determining the position of the RGB-D camera system in the spatial domain. An encoder and kinematics equations are applied to determine the position during movement. As expected, the system shows good results, such as a high IoU value of over 0.8 at a distance of less than 2 m and a fast time of 41.66 ms for object detection. This means that our technique is very effective for the automatic movement of the wheelchair. [ABSTRACT FROM AUTHOR]

Image processing is a key research area in computer vision that recognizes images and assigns labels to the extracted features. This new paradigm has recently received significant attention in biometric features that aid in the identification of breed and individual cattle based on extracted muzzle point features. The primary goal of this article is to present a comparison of various feature descriptors and classifiers for cattle identification using muzzle points. This paper also involves identification based on various features extracted from an image. There are a few feature extraction techniques available, such as Holistic features extraction, ORB, SIFT, SURF, Shi-Tomasi, Harris corner detection, and so on. The authors of this article considered three feature descriptor algorithms, SIFT (Scale Invariant Feature Transform), SURF (Speeded up Robust Feature), and ORB (Oriented Fast and Rotated BRIEF), to carry out experimental work on cattle images for breed identification system. A differentiation among three descriptors is presented in this article by determining them individually and in a combination of these methodologies. Classifiers such as Decision Tree, k-NN, and Random Forest are used to categorize images based on extracted features. The experiments are carried out on a dataset of four breeds: Holstein Friesian (470 images), Jersey (200 images), Rathi (100 images), and Sahiwal (100 images) (160 images). In the partitioning strategy, 80% of the data is considered as training dataset, and the remaining 20% is considered a testing set. The accuracy of 97.23% is achieved by hybrid feature techniques composed of SIFT, SURF, and ORB for the cattle identification system. This paper describes the feature extraction and classifiers used in the cattle identification system, data collection, methodology, and design of the proposed system, evaluates results, and represents the system's relevance and future prospects. [ABSTRACT FROM AUTHOR]

In computer-vision feature extraction algorithms, compressing the image into a sparse set of trackable keypoints, empowers navigation-critical systems such as Simultaneous Localization And Mapping (SLAM) in autonomous robots, and also other applications such as augmented reality and 3D reconstruction. Most of those applications are performed in battery-powered gadgets featuring in common a very stringent power-budget. Near-to-sensor computing of feature extraction algorithms allows for several design optimizations. First, the overall on-chip memory requirements can be lessened, and second, the internal data movement can be minimized. This work explores the usage of an Application Specific Instruction Set Processor (ASIP) dedicated to perform feature extraction in a real-time and energy-efficient manner. The ASIP features a Very Long Instruction Word (VLIW) architecture comprising one RV32I RISC-V and three vector slots. The on-chip memory sub-system implements parallel multi-bank memories with near-memory data shuffling to enable single-cycle multi-pattern vector access. Oriented FAST and Rotated BRIEF (ORB) are thoroughly explored to validate the proposed architecture, achieving a throughput of 140 Frames-Per-Second (FPS) for VGA images for one scale, while reducing the number of memory accesses by 2 orders of magnitude as compared to other embedded general-purpose architectures. [ABSTRACT FROM AUTHOR]

Until today, various approaches have been proposed in order to create Augmented Reality (AR) environment where virtual-real integration takes place. One of these approaches is vision-based model, and it is divided into two branches as Marker-Based Augmented Reality (MBAR) and Markerless Augmented Reality (MAR). In the use of MBAR model, a reference image is introduced to the system before, and when this image enters the camera view, an AR environment is created. However, in MAR model, no image is introduced to the system before. Instead, it uses natural characteristics present in the image, such as edges, corners, and geometrical shapes to create an AR environment. In order to use MAR model, it is necessary to use algorithms which require high processing power and memory capacity. Within the scope of this study, MAR model was chosen as reference and an evaluation on combinations of descriptive extractors (such as ORB, SIFT, and SURF) and matchers (such as Bruteforce, Bruteforce-Hamming, and Flannbase) was presented. In this context, it was aimed that we could obtain knowledge about i) the number of key points and detection time with the use of different descriptor extractors, and ii) the number of matching key points and the amount of positional deviation of a virtual object placed on a real world scene with the use of different matchers. In line with this goal, analyses were made, using different image scales and brightness levels on both PC and mobile platforms. Results showed that, for both platforms, combinations using ORB method could work faster with less deviation than the combinations using other methods in all conditions. In addition, RANSAC algorithm was also used to reduce the total mean deviation ratio, and it was seen that the rate could be reduced from 70% to 4.5%.

In pattern matching, estimating the spatial distance between generated feature descriptors is known as "similarity matching". It is a fundamental procedure in many application disciplines. The primary axis for rapid and precise similarity matching is the extraction method, which can effectively represent images with low-dimensional descriptors. In this paper, we evaluate the Scale-Invariant Feature Transform (SIFT) due to its relevance and use in numerous applications. However, it has several limitations. Thus, the usage of Oriented FAST and Rotated BRIEF (ORB) is recommended. In the form of three scenarios, this paper introduces similarity matching performance assessment for each method from several perspectives not explicitly expressed in the literature. The first scenario is introduced to assess the extraction of the same features from multiple images. The method evaluation to identify the K most important keypoints is included in the second scenario. The sorting of the used images hinders the feature extraction method as well as the similarity matching. The third scenario involves evaluating the matching performance of visual features extracted with SIFT and ORB from real and sketched images. Furthermore, this paper introduces a step forward to fill the gap between sketch and real domains by converting real images to pencil-sketch images. Finally, the computational complexity of the overall test cases is computed and assessed. According to the results of real-to-pencil-sketch image matching, ORB outperforms SIFT in terms of the computed performance metrics as well as the correct matching by a factor of around two. Besides, SIFT has a temporal complexity that is nearly double that of ORB, which directly affects the capacity of any application relying on similarity matching. [ABSTRACT FROM AUTHOR]

The Oriented FAST and Rotated BRIEF (ORB) algorithms have been improved, and a new method for calculating descriptors based on symmetrical multicentric weighted binary encoding is proposed, which enhances the robustness of feature points. This method employs a string of binary descriptors to encode the features and uses the multiple descriptor centre strategy to sample descriptors at the feature point and on the symmetrical circumference around it. Furthermore, a weighted summation is introduced in the descriptor calculation process to address the noise in the image during the sampling process. Specifically, the pixel values around the sampled point and the sampled point itself are combined using a certain weight to produce the final pixel value of the sampled point. The reliability of the descriptor is enhanced by introducing the pixel information around the sample point while solving the noise problem. Our method makes full use of the pixel information in the various parts of the descriptor sampling region to improve the distinguishability of the descriptors. We compare it with the ORB algorithm and experimentally show that the feature extraction method achieves better matching results with almost constant computation time. [ABSTRACT FROM AUTHOR]

Simultaneous Localization And Mapping (SLAM) is crucial for autonomous navigation. ORB-SLAM is a state-of-the-art Visual SLAM system based on cameras used for self-driving cars. In this paper, we propose a high-performance, energy-efficient, and functionally accurate hardware accelerator for ORB-SLAM, focusing on its most time-consuming stage: Oriented FAST and Rotated BRIEF (ORB) feature extraction. The Rotated BRIEF (rBRIEF) descriptor generation is the main bottleneck in ORB computation, as it exhibits highly irregular access patterns to local on-chip memories causing a high-performance penalty due to bank conflicts. We introduce a technique to find an optimal static pattern to perform parallel accesses to banks based on a genetic algorithm. Furthermore, we propose the combination of an rBRIEF pixel duplication cache, selective ports replication, and pipelining to reduce latency without compromising cost. The accelerator achieves a reduction in energy consumption of 14597× and 9609×, with respect to high-end CPU and GPU platforms, respectively. • Feature extraction takes 60% of the localization execution time in current solutions. • Streaming architecture in which the bottleneck is the computation of rBRIEF. • A genetic algorithm can statically reorder the rBRIEF pattern to reduce the latency. • Optimizations include: pixel cache, selective replication of ports and pipelining. • Search of the most energy efficient factors that fulfill the self-drivings requirements. [ABSTRACT FROM AUTHOR]

Video stitching is a technique that stitches multiple overlapped videos acquired from different cameras, which is widely applied in many applications, including video surveillance, autonomous driving, and virtual reality. Feature-based stitching methods are popular in this area because of their invariance property and efficiency. However, the video stitching pipeline is relatively complicated and the amount of data calculation is large, which impedes its real-time applications. In this paper, we propose a real-time video stitching framework based on vision Digital Signal Processing (DSP). Real-time processing is achieved by the algorithm-level and system-level optimizations. In the algorithm of ORB feature extraction, methods, including look-up table, linear approximation, and single instruction multiple data (SIMD), are adopted to optimize its computation on DSP. In the algorithm of feature matching, the Hamming distance calculation is eliminated for some feature point pairs when their coordinates and angles are not satisfied with certain conditions. Reverse feature matching is proposed to improve registration accuracy. In system-level optimization, the directed acyclic graph (DAG)-based scheduling is proposed to improve the calculation efficiency on dual DSPs, and a ping-pong buffer is utilized to speed up the data transmission between DSP and external memory. Experimental results show that the proposed method can achieve a ten times speedup than that of the CPU, and it can achieve 1536 × 1024@37fps real-time processing on vision DSP. [ABSTRACT FROM AUTHOR]

Image classification is getting more attention in the area of computer vision. During the past few years, a lot of research has been done on image classification using classical machine learning and deep learning techniques. Presently, deep learning-based techniques have given stupendous results. The performance of a classification system depends on the quality of features extracted from an image. The better is the quality of extracted features, the more the accuracy will be. Although, numerous deep learning-based methods have shown enormous performance in image classification, still due to various challenges deep learning methods are not able to extract all the important information from the image. This results in a reduction in overall classification accuracy. The goal of the present research is to improve the image classification performance by combining the deep features extracted using popular deep convolutional neural network, VGG19, and various handcrafted feature extraction methods, i.e., SIFT, SURF, ORB, and Shi-Tomasi corner detector algorithm. Further, the extracted features from these methods are classified using various machine learning classification methods, i.e., Gaussian Naïve Bayes, Decision Tree, Random Forest, and eXtreme Gradient Boosting (XGBClassifier) classifier. The experiment is carried out on a benchmark dataset Caltech-101. The experimental results indicate that Random Forest using the combined features give 93.73% accuracy and outperforms other classifiers and methods proposed by other authors. The paper concludes that a single feature extractor whether shallow or deep is not enough to achieve satisfactory results. So, a combined approach using deep learning features and traditional handcrafted features is better for image classification. [ABSTRACT FROM AUTHOR]

Cervical Spondylosis (CS) is a chronic spinal condition in which the spine gradually stiffens and can finally become completely inflexible. It is arduous to diagnose in early stages and leads to delay in medication. The risk level of Cervical Spondylosis can be reduced if it is detected in primary care. Based on this objective, a system is designed and developed to diagnose and predict the severity of cervical spondylosis in early stages. Different machine learning techniques are evaluated for this and results indicate that machine learning techniques can provide a low cost and accurate mechanism for early stage spondylosis detection.

In recent years, some learning-based methods are proposed to detect and locate humans in real-time via convolutional neural networks (CNN). However, high-performance graphics processing units (GPUs) are required in those methods. To resolve this problem, a preprocessing procedure based on video segmentation is proposed to speed up face detection. Meanwhile, an accelerating toolkit is employed in this study to perform face detection in real-time on a standard central processing unit (CPU). Experimental results indicate that the proposed method can achieve an F1-Score of 93.2% and 4.5 times of real-time speed with one CPU on 155883 test frames from the RAI dataset, YouTube, and YOUKU. Notably, when the video sequence is with fewer frames of human faces, the highest speed is nearly 18 times faster than that without video segmentation. [ABSTRACT FROM AUTHOR]

This paper presents a novel two-view Structure-from-Motion (SfM) algorithm with the application of multiple Feature Detector Operators (FDO). The key of this study is the implementation of multiple FDOs into a two-view SfM algorithm. The two-view SfM algorithm workflow can be divided into three general steps: feature detection and matching, pose estimation and point cloud (PCL) generation. The experimental results, the quantitative analyses and a comparison with existing algorithms demonstrate that the implementation of multiple FDOs can effectively improve the performance of a two-view SfM algorithm. Firstly, in the Oxford test dataset, the RMSE reaches on average 0.11 m (UBC), 0.36 m (bikes), 0.52 m (trees) and 0.37 m (Leuven). This proves that illumination changes, blurring and JPEG compression can be handled satisfactorily. Secondly, in the EPFL dataset, the number of features lost in the processes is 21% with a total PCL of 27,673 pt, and this is only minimally higher than ORB (20.91%) with a PCL of 10,266 pt. Finally, the verification process with a real-world unmanned aerial vehicle (UAV) shows that the point cloud is denser around the edges, the corners and the target, and the process speed is much faster than existing algorithms. Overall, the framework proposed in this study has been proven a viable alternative to a classical procedure, in terms of performance, efficiency and simplicity. [ABSTRACT FROM AUTHOR]