Search

Radar sensors and micro-Doppler signatures have been widely used to recognize human motions. Apart from the motion classification task, human motion detection has attracted much attention as an emerging topic. A majority of existing motion detectors are designed for a specific motion, such as falling. In some scenarios, however, a broader range of human actions is of interest, hence a general motion detector is desired. In this paper, we propose a radar-based motion detection model named dual generative adversarial network (DGN). The proposed model tackles the detection task as a one-class classification problem and is applicable to detecting various motions. Unlike prior fall detection algorithms, which depend on manually collected alien data, the DGN employs a dual generation scheme to automatically produce valid alien samples in both the pixel level and the semantic level. The model is verified on two measured radar datasets containing individual motions and interactive motions, respectively. The experimental results show that our method outperforms other existing models on the human motion detection task.

Abnormal event detection is an important task in research and industrial applications, which has received considerable attention in recent years. Existing methods usually rely on standard frame-based cameras to record the data and process them with computer vision technologies. In contrast, this paper presents a novel neuromorphic vision based abnormal event detection system. Compared to the frame-based camera, neuromorphic vision sensors, such as Dynamic Vision Sensor (DVS), do not acquire full images at a fixed frame rate but rather have independent pixels that output intensity changes (called events ) asynchronously at the time they occur. Thus, it avoids the design of the encryption scheme. Since events are triggered by moving edges on the scene, DVS is a natural motion detector for the abnormal objects and automatically filters out any temporally-redundant information. Based on this unique output, we first propose a highly efficient method based on the event density to select activated event cuboids and locate the foreground. We design a novel event -based multiscale spatio-temporal descriptor to extract features from the activated event cuboids for the abnormal event detection. Additionally, we build the NeuroAED dataset, the first public dataset dedicated to abnormal event detection with neuromorphic vision sensor. The NeuroAED dataset consists of four sub-datasets: Walking, Campus, Square, and Stair dataset. Experiments are conducted based on these datasets and demonstrate the high efficiency and accuracy of our method.

Integrating local motion signals detected by the primary motion detector is crucial for representing a rigid, two-dimensional motion. The nature of motion integration has been studied using stimuli consisting of two superimposed sinusoidal gratings of different orientations, called plaid motion, and it has been shown that humans perceive integrated motion in the direction where the component constraint lines are intersected. We previously found that pigeons and humans perceive different movement directions from plaid motion; pigeons responded to the vector average direction of the gratings. Although this suggests that the underlying processes of motion integration differ between the two species, the viewing distance in the pigeon experiment, which used a touch panel procedure, was much smaller than in typical human experiments. The current study investigated the potential effect of viewing distance on the perception of plaid motion in pigeons. We trained six pigeons to detect whether motion directions were tilted leftward or rightward while a visual display was presented 0 or 40 cm from an operant chamber. The pigeons responded to plaid stimuli for both viewing distance conditions as if they perceived motion in the vector average direction of two-component gratings. The result indicates that the species difference in plaid perception is not an artefact of viewing distance and suggests that pigeons use a different strategy than humans for integrating visual motion.

Every object is always moving, and movement of an object can occur at any time. The value of the motion which produced from an object can be very small to very large, and the impact of the movement can be at risk until very risky. For this reason, the movement of objects at risk must be observed whenever changes and observations for data retrieval can be done remotely. The purpose of this research to design an Internet of Things (IoT) devices that can observe and detect changes in the motion of an object. Â The device is designed to be small, around 44 x 48 millimeters with very low power consumption. The design phase begins with recording motion data using the MPU6050 accelerometer sensor as a motion detector, arduino nano as a control device, WiFi ESP8266 as a communication medium for sending data from a receiver apllication motion data with UDP protocol. The test results show that this device is very sensitive to detect changes in the motion and angle of X, Y and Z of an object.

Pyroelectric infrared (PIR) sensor is widely used as motion detector, due to its sensitivity to the infrared radiation changes (IRC) induced by human motion. In particular, it is also possible to determine the human activity patterns based on the IRC signal by improving the feature-specific sensitivity of the PIR sensors. In this paper, the feature-specific sensitivity is enhanced by visibility modulation at the geometric sensing layer. At the cooperative sensing layer, a novel stereo configuration is used in developing the infrared sensing system, which aims to handle the difficulties caused by occlusion and limited coverage area. The experimental results demonstrate the effectiveness of the developed system in daily activity recognition.

We developed and tested the architecture of a bio-inspired Spiking Neural Network for motion estimation. The computation performed by the retina is emulated by the neuromorphic event-based image sensor DAVIS346 which constitutes the input of our network. We obtained neurons highly tuned to spatial frequency and orientation of the stimulus through a combination of feed-forward excitatory connections modeled as an elongated Gaussian kernel and recurrent inhibitory connections from two clusters of neurons within the same cortical layers. Sums over adjacent nodes weighted by time-variable synapses are used to attain Gabor-like spatio-temporal V1 receptive fields with selectivity to the stimulus’ motion. In order to gain the invariance to the stimulus phase, the two polarities of the events provided by the neuromorphic sensor were exploited, which allowed us to build two pairs of quadrature filters from which we obtain Motion Energy detectors as described in [2]. Finally, a decoding stage allows us to compute optic flow from the Motion Detector layers. We tested the approach proposed with both synthetic and natural stimuli.

Aiming at continuous unobtrusive respiration monitoring, motion robustness is paramount. However, some types of motion can completely hide the respiration information and the detection of these events is required to avoid incorrect rate estimations. Therefore, this work proposes a motion detector optimized to specifically detect severe motion of infants combined with a respiration rate detection strategy based on automatic pixels selection, which proved to be robust to motion of the infants involving head and limbs. A dataset including both thermal and RGB (Red Green Blue) videos was used amounting to a total of 43 h acquired on 17 infants. The method was successfully applied to both RGB and thermal videos and compared to the chest impedance signal. The Mean Absolute Error (MAE) in segments where some motion is present was 1.16 and 1.97 breaths/min higher than the MAE in the ideal moments where the infants were still for testing and validation set, respectively. Overall, the average MAE on the testing and validation set are 3.31 breaths/min and 5.36 breaths/min, using 64.00% and 69.65% of the included video segments (segments containing events such as interventions were excluded based on a manual annotation), respectively. Moreover, we highlight challenges that need to be overcome for continuous camera-based respiration monitoring. The method can be applied to different camera modalities, does not require skin visibility, and is robust to some motion of the infants.

This paper suggests a motion detection software system that enables us to see the movement around an object or a visual area. Motion detection software is widely preferred as one of the most important security features in recent days. It is used to enhance existing security devices such as motion sensor lighting on indoor and outdoor security cameras. Security cameras are one of the most common devices that use motion detection technology [1]. Now, for better protection, motion detector cameras are preferred. The paper suggested here looks at one of the python motion detection software and detects the movement of objects in front of the camera and determines the time they stay close to the camera.

Face recognition and biometric analytics is a novel domain of research and enormous algorithms going in this dimension. The identification of smile is a two-stage operation. You sense a face first and then wait for a smile and a motion detector breaks the shot in thousands of zones, Criteria for analysis such as white balance and facial level. If a human smile, by specifying various parameters, the camera detects a facial defect, including closing your eyes, clear teeth, bending your lips and raising your lips. In order to improve the sensitivity of the smile automatic function you should change the camera parameters. Authentic smile identification is more successful when participants, particularly their eyes, (with bangs, etc.) do not cover their faces. Hats, masks or sunglasses may also obstruct the work. You should have a big and open-mouthed grin on your subjects. When your teeth are clear, your camera can still detect smiling better. The presented work is focusing on the face smile analytics with the Perceptual User Interfaces for cumulative performance in the biometric analytics.

Object detection and motion parameters estimation are crucial tasks for self-driving vehicle safe navigation in a complex urban environment. In this work we propose a novel real-time approach of temporal context aggregation for motion detection and motion parameters estimation based on 3D point cloud sequence. We introduce an ego-motion compensation layer to achieve real-time inference with performance comparable to a naive odometric transform of the original point cloud sequence. Not only is the proposed architecture capable of estimating the motion of common road participants like vehicles or pedestrians but also generalizes to other object categories which are not present in training data. We also conduct an in-deep analysis of different temporal context aggregation strategies such as recurrent cells and 3D convolutions. Finally, we provide comparison results of our state-of-the-art model with existing solutions on KITTI Scene Flow dataset., Comment: Accepted to ICRA 2020

A visitor's interest in the exhibits at a museum can be understood by analyzing his or her behavior. Conventionally, behavior analysis is performed manually, but it is troublesome to do this in practice. Therefore, our aim is to reduce this trouble by the automation and semi-automation of behavior analysis. We focus on motion detection by machine learning as the first step in this process. To detect motion using machine learning, it is necessary to collect training motions in advance and create a detector. In this study, human models and motions were generated using three-dimensional computer graphics. Then, we synthesized them and collected training motions using a virtual environment. We used the collected training motions to create a detector and detected motions collected from people in a real environment. The results of the experiments indicate that this training data collection method is effective.

This work proposes the development of a sociometric sensor capable of detecting proximity, movement and verbal interaction between people. The sensor will allow measuring the level of existing interrelations between the collaborators of an organization, in order to determine and to measure the organizational climate. The way to evaluate the current organizational climate is subjective, through surveys that are not accurate and that make the results are not adequate. The sensors developed until now focus on proximity or movement, but not on the 3 proposed variables: movement, proximity and speech. The proposed portable device consists of a proximity detector using RSSI (Received Signal Strength Indicator) based on the NodeMCU WIFI ESP 8266 module. It also has a motion detector that uses an accelerometer and finally a voice detection algorithm, acquired via microphone, aimed at determining verbal interaction between people. The validation of the equipment was done by measuring the 3 variables in a sample of 50 people. The results show a percentage of proximity detection of 90% up to a distance of 3 meters between individuals, a percentage of success of 92.5% in the detection of posture and physical activity and a percentage of success of 87.5% in the detection of verbal interaction between people.

Summary The detection of visual motion is a fundamental function of the visual system. How motion speed and direction are computed together at the cellular level, however, remains largely unknown. Here, we suggest a circuit mechanism by which excitatory inputs to direction-selective ganglion cells in the mouse retina become sensitive to the motion speed and direction of image motion. Electrophysiological, imaging, and connectomic analyses provide evidence that the dendrites of ON direction-selective cells receive spatially offset and asymmetrically filtered glutamatergic inputs along motion-preference axis from asymmetrically wired bipolar and amacrine cell types with distinct release dynamics. A computational model shows that, with this spatiotemporal structure, the input amplitude becomes sensitive to speed and direction by a preferred direction enhancement mechanism. Our results highlight the role of an excitatory mechanism in retinal motion computation by which feature selectivity emerges from non-selective inputs., Graphical Abstract, Highlights • Space-time wiring between ON DS cells and bipolar cell types was identified • Presynaptic inhibition and NaV channels set the glutamate release dynamics • Preferred direction enhancement supports the tuning to direction and slow speed • Global dendritic summation is involved in computing image motion, Matsumoto et al. find the space-time wiring between the dendrites of ON direction-selective cells and bipolar cell types in the mouse retina. Temporally diverse glutamatergic inputs, which are asymmetrically organized across the dendrites, are globally summated by a preferred direction enhancement mechanism for computing motion speed and direction.

This letter presents a low-power motion gesture recognition system-on-chip (SoC) for smart devices. The SoC incorporates a low-power image sensor and a memory-efficient outermost-edge-based gesture sensing DSP. The DSP utilizes a self-adaptive motion detector that automatically updates a motion-pixel threshold for accurately sensing hand movements. A convolution-based noise-tolerant feature extraction (FE) technique is also developed for preventing detection errors caused by random noises in the images from the low-power sensor. The FE architecture is highly accelerated utilizing parallelisms and pipelining for achieving low-latency real-time gesture recognition. Measurements from a test chip fabricated in 65-nm CMOS show that the SoC consumes $213.7~{\mu }\text{W}$ with only 3- ${\mu }\text{W}$ dynamic power at 30 f/s. The SoC occupies only 0.54 mm2, making it well applicable for wearable devices and sensor nodes. The image sensor is fully operational down to 0.6 V while the DSP can be scaled down to 0.46 V. The average recognition accuracy of the system is 85% while the latency is 1.056 ms.

It is important to track people's face efficiently and accurately in many Intelligent Transportation Systems ITSs and Safety Driving Assistant Systems SDASs. This paper presents a high-performance and low-cost real-time face tracking system, which runs on general onboard computer with very low CPU consumption. The proposed face tracking system is composed of four modules: the motion detector, face detector, face tracker, and face validator. The motion detector extracts motion areas by using a spatial-temporal bi-differential method with a very low computational cost. The face detector integrates motions into a cascade face detection framework to reject most of non-face scanning-windows to ensure efficient face localization. The face tracker fuses motion feature with color feature to alleviate the drifting problem during tracking. The face validator builds face appearance models online and identifies each specific tracked face to avoid confusion. Experimental results on three challenging video sequences show that the proposed face tracking system outperforms the state-of-the-art face tracker and consumes only 5-13% CPU resources of a low-spec onboard computer while processing in real time. Copyright © 2016 John Wiley & Sons, Ltd.

In this study, we developed 44 working motion detector templates and a behavioral recognition system with the motion templates to recognize worker behaviors, such as turning, squatting, and moving arms, performed by a worker along furrows in the cultivated field of peas. The 44 motion detector templates were designed to identify working motions such as sitting down, moving arms, turning counterclockwise, turning clockwise, up-and-down, side-to-side. The behavioral recognition system with the motion templates was designed to recognize the 12 behavior types and the three behavior category types of turning, squatting, and moving arms behaviors. In the experiments, the developed motion detector templates could consistently well respond to the working motions in the 12 types of behaviors performed by two workers. The behavioral recognition system with the motion templates could recognize the three behavior category types could be robustly determined with the total recognition rate of 80.5%. The results of the experiments demonstrated the applicability and robustness of the developed working motion templates and the behavioral recognition system with the motion templates to recognize the three behavior category types.

Building a reliable and efficient collision avoidance system for unmanned aerial vehicles (UAVs) is still a challenging problem. This research takes inspiration from locusts, which can fly in dense swarms for hundreds of miles without collision. In the locust’s brain, a visual pathway of LGMD-DCMD (lobula giant movement detector and descending contra-lateral motion detector) has been identified as collision perception system guiding fast collision avoidance for locusts, which is ideal for designing artificial vision systems. However, there is very few works investigating its potential in real-world UAV applications. In this paper, we present an LGMD based competitive collision avoidance method for UAV indoor navigation. Compared to previous works, we divided the UAV’s field of view into four subfields each handled by an LGMD neuron. Therefore, four individual competitive LGMDs (C-LGMD) compete for guiding the directional collision avoidance of UAV. With more degrees of freedom compared to ground robots and vehicles, the UAV can escape from collision along four cardinal directions (e.g. the object approaching from the left-side triggers a rightward shifting of the UAV). Our proposed method has been validated by both simulations and real-time quadcopter arena experiments.

Rapid increase in human population and their encroachment into forest areas have posed several threats to human life and their livelihood. This has subsequently led to a demand for surveillance system that detect the motion of objects in particular areas where threat is possible. This application also detects the pattern of the object i.e. the time at which he comes and goes to their place. For detecting, the presence of particular wildlife in a particular area and their sleep pattern and various things about their identity can be known by the use of this application. The application detects the motion of the object that come in the visual region of the camera and plot the time for which the object remain in its visual sight.

In this paper, we propose a nonintrusive video-based system for vehicle speed measurement in urban roadways. Our system uses an optimized motion detector and a novel text detector to efficiently locate vehicle license plates in image regions containing motion. Distinctive features are then selected on the license plate regions, tracked across multiple frames, and rectified for perspective distortion. Vehicle speed is measured by comparing the trajectories of the tracked features to known real-world measures. The proposed system was tested on a data set containing approximately 5 h of videos recorded in different weather conditions by a single low-cost camera, with associated ground truth speeds obtained by an inductive loop detector. Our data set is freely available for research purposes. The measured speeds have an average error of −0.5 km/h, staying inside the [−3, +2] km/h limit determined by regulatory authorities in several countries in over 96.0% of the cases. To the authors' knowledge, there are no other video-based systems able to achieve results comparable to those produced by an inductive loop detector. We also show that our license plate detector outperforms two other published state-of-the-art text detectors, as well as a well-known license plate detector, achieving a precision of 0.93 and a recall of 0.87.

Pyroelectric infrared (PIR) sensor is widely used as motion detector, due to its sensitivity to the Infrared Radiation Changes (IRC) induced by human motion. In particular, by improving the space resolution and coverage of the PIR sensor, it's possible to determine the human activity patterns using IRC signal. In this paper, we present a novel method of fall detection with PIR sensor network. To improve the space resolution, we use mask arrays as reference structures in the developed IRC sensing modules. A stereo infrared sensing scheme is also designed to enlarge the sensing field for fully capturing the IRC signals. Based on the collected IRC signals, we have performed classification analysis with well known machine learning algorithms. The experimental results demonstrate the effectiveness of the proposed PIR sensor system in fall detection.

Summary Visual motion detection in insects is mediated by three-input detectors that compare inputs of different spatiotemporal properties. A new modeling study shows that only a small subset of possible arrangements of the input elements provides high direction-selectivity.

The use of technology in teaching and learning is very common today. In physics learning, practicum has an important role, namely deepening the understanding of concepts and applying concepts. However, it is interesting to know the importance of using technology in practicum. This study aims to know to know the student’s Physics education students’ perception on the use of motion detector in linear motion practicum The present study was descriptive research using a qualitative approach. This research was carried out in mechanics experiment class with 16 students as research subjects. This study used students’ reflection and interview in collecting perception’s of the students. The research found that practicum with motion detector has significant role in results obtained. On the other hand, practicum with motion detector has unsignificant role in the student’s carefullness.

This research aims to design, test, and analyze the object tracking application system using a motion detector method that is able to take the output image of several cameras by combining the output images of the two cameras into a wider field of view (FOV). This research uses 2 cameras, i3-6006U laptop camera and Logitech C270 HD webcam using OpenCV and python programming language. The stitching results from the initialization of the two cameras are able to detect the motion of objects performed on panoramatic videos in realtime by showing time in detail starting from days, dates, hours, minutes and seconds. This object tracking application system has been tested in Adi Sucipto airport exactly in the area of arrival and departure of passengers. The system can detected the motion in the stitching images by exploring it in red rectanguler in the real time video.

Responses of direction-selective and orientation-selective motion detectors were recorded extracellularly from the axon terminals of ganglion cells in the superficial layers of the tectum opticum of immobilized goldfish, Carassius gibelio (Bloch, 1782). Color stripes or edges moving on some color background (presented on the CRT monitor with known emission spectra of its phosphors) served as stimuli. It was shown that stimuli of any color can be more or less matched with the background by varying their intensities what is indicative of color blindness of the motion detectors. Sets of stimuli which matched the background proved to represent planes in the three-dimensional color space of the goldfish. A relative contribution of different types of cones to the spectral sensitivity was estimated according to orientation of the plane of color matches. The spectral sensitivity of any motion detector was shown to be determined mainly by long-wave cones with a weak negative (opponent) contributions of middle-wave and/or short-wave ones. This resulted in reduced sensitivity in the blue-green end of the spectrum, what may be considered as an adaptation to the aquatic environment where, because of the substantial light scattering of a blue-green light, acute vision is possible only in a red region of the spectrum.

As a result of the increasing popularity of remotely triggered (trail) cameras among hunters, amateur naturalists, and field biologists, a large variety of equipment and technologies have become available. Among various models of cameras, such features as the size of detection zone, trigger speed, and sensitivity of infrared-motion detector can influence performance, including animal detection rates. Variation in detection (or photographic) rates among cameras may be problematic in studies where several models of cameras are used, especially when performance differences are not incorporated in the study design. To evaluate this potential source of sampling bias, we conducted head-to-head field trials with 2 cameras (Cuddeback™ Capture IR and Reconyx™ HC 600 Hyperfire) that had substantially different attributes that can affect sampling efficiency. Among initial comparisons, we found that Reconyx cameras (large detection zone and high-sensitivity motion detector) recorded approximately twice as many independent photographs as did the paired Cuddeback cameras. In subsequent trials, we reduced the sensitivity of the motion detector on the Reconyx cameras and positioned a wooden stake with bait in the center of the immediate detection zone of both cameras. These adjustments reduced performance differences between the 2 cameras. We recommend that biologists involved with research, inventory, or monitoring programs that rely on more than one model of trail camera consider a testing phase where they can evaluate and possibly adjust differences among cameras. © 2014 The Wildlife Society.

Mobile robots are now widely used in various security and surveillance applications. The contribution describes a design of the robot with the function of the security system in the home. This robot can be controlled by mobile phone using DTMF or hobby RC system. In case of unavailability of operator, robot can operate in autonomous mode too. In this mode, robot automatically reacts to stimuli from the environment e.g. If motion detector detects movement, robot can automatically start recording video from camera. As propulsion system, two modified RC servos were used. Robot has many sensors e.g. PIR movement sensor, gyroscope, infrared distance sensors, ultrasonic distance sensor. Some of these sensors can be manipulated by servos. On top, there is a video camera. This camera is capable of recording or transmitting audio and video over mobile phone using 3G video calls.

In research and scientific applications Bio-inspired motion detection models have been attracted by small-scale unmanned aerial vehicles (UAVs). UAVs are used for surveillance, security, search and rescue mission. Research on autonomous navigation is greatly influenced by the anatomy of flying insect’s eyes. The basic Reichardt correlator which seems to be the fundamental part in all insect motion processing is used in finding relative motion. An array of Reichardt correlator is used to correlate between each pixel values of successive frames. Responses obtained from each correlator model are aggregated together which give us the information about the direction of motion. The proposed architecture is synthesized and analyzed for timing and power using Xilinx ISE tool. This architecture can have applications for obstacle detection and motion detection in various fields.

Robust and efficient target-tracking algorithms embedded on moving platforms, are a requirement for many computer vision and robotic applications. However, deployment of a real-time system is challenging, even with the computational power of modern hardware. As inspiration, we look to biological lightweight solutions-lightweight and low-powered flying insects. For example, dragonflies pursue prey and mates within cluttered, natural environments, deftly selecting their target amidst swarms. In our laboratory, we study the physiology and morphology of dragonfly 'small target motion detector' neurons likely to underlie this pursuit behaviour. Here we describe our insect-inspired tracking model derived from these data and compare its efficacy and efficiency with state-of-the-art engineering models. For model inputs, we use both publicly available video sequences, as well as our own task-specific dataset (small targets embedded within natural scenes). In the context of the tracking problem, we describe differences in object statistics within the video sequences. For the general dataset, our model often locks on to small components of larger objects, tracking these moving features. When input imagery includes small moving targets, for which our highly nonlinear filtering is matched, the robustness outperforms state-of-the-art trackers. In all scenarios, our insect-inspired tracker runs at least twice the speed of the comparison algorithms.

The stationary background generation problem consists in generating a unique image representing the stationary background of a given video sequence. The LaBGen background generation method combines a pixel-wise median filter and a patch selection mechanism based on a motion detection performed by a background subtraction algorithm. In our previous works related to LaBGen, we have shown that, surprisingly, the frame difference algorithm provides the most effective motion detection on average. Compared to other background subtraction algorithms, it detects motion between two frames without relying on additional past frames, and is therefore memoryless. In this paper, we experimentally check whether the memoryless property is truly relevant for LaBGen, and whether the effective motion detection provided by the frame difference is not an isolated case. For this purpose, we introduce LaBGen-OF, a variant of LaBGen leverages memoryless dense optical flow algorithms for motion detection. Our experiments show that using a memoryless motion detector is an adequate choice for our background generation framework, and that LaBGen-OF outperforms LaBGen on the SBMnet dataset. We further provide an open-source C++ implementation of both methods at http://www.telecom.ulg.ac.be/labgen.

We present an event-based optical flow algorithm for the Davis Dynamic Vision Sensor (DVS). The algorithm is based on the Reichardt motion detector inspired by the fly visual system, and has a very low computational requirement for each event received from the DVS.

In this paper, a novel motion detector is proposed to perceive the weak changes in a image sequence. This is inspired by the mechanism of fixational eye movement and dynamics of vertebrate’s cortex. We realized respectively an artificial model of visual attention selection, called dual-probe adaptive model (DPAM), and an active tremor operation (ATO) approach. It is found that between them there exists a resonance phenomenon. The phenomenon is enhanced when the ATO and the DPAM are in-phase and is suppressed when they are anti-phase. Based on this, we construct a novel motion detector combined by the ATO and the DPAM to resonate with the motion direction. This allows capturing moving edges even in the image sequences with lighting change and noisy background. Simulation and Experimental results demonstrate the effectiveness.

We present a biologically inspired model for estimating the position of a moving target that is invariant to the target’s contrast. Our model produces a monotonic relationship between position and output activity using a divisive normalization between the ‘receptive fields’ of two overlapping, wide-field, small-target motion detector (STMD) neurons. These visual neurons found in flying insects, likely underlie the impressive ability to pursue prey within cluttered environments. Individual STMD responses confound the properties of target contrast, size, velocity and position. Inspired by results from STMD recordings we developed a model using a division operation to overcome the inherent positional ambiguities of integrative neurons. We used genetic algorithms to determine the plausibility of such an operation arising and existing over multiple generations. This method allows the lost information to be recovered without needing additional neuronal pathways.

In nature, it is an important task for animals to detect small targets which move within cluttered background. In recent years, biologists have found that a class of neurons in the lobula complex, called STMDs (small target motion detectors) which have extreme selectivity for small targets moving within visual clutter. At the same time, some researchers assert that lateral inhibition plays an important role in discriminating the motion of the target from the motion of the background, even account for many features of the tuning of higher order visual neurons. Inspired by the finding that complete lateral inhibition can only be seen when the motion of the central region is identical to the motion of the peripheral region, we propose a new lateral inhibition mechanism combined with motion velocity and direction to improve the performance of ESTMD model (elementary small target motion detector). In this paper, we will elaborate on the biological plausibility and functionality of this new lateral inhibition mechanism in small target motion detection.

The process of volcano research through visual monitoring usually takes a long time. There is many time possibility of an occurance of a specific volcano phenomena such as lava and plum. The problem with this monitoring activity can be helped through visual monitoring system that enables constant recording and automatically detects events that can be observed through the visual volcano object. This monitoring system works through an event detection application with static object definitive, fixed focus definitive and sensing area definitive combination method that forms a simple yet effective method called Motion Range Event Detection. This method is a form of volcano object monitoring using camera sensors by defining the form of volcano object as a static object and defining the object phenomena changes based on a predetermined area.

In the human motion measurement, motion capture system is used to record the movement of the human body by using different types of sensors such as a magnetic position sensor, a mechanical motion detector and a vision sensor. The most challenging task in human motion measurement is to achieve the ability and reliability of a motion capture system for tracking and recognizing dynamic gestures, because human body structure has many degrees of freedom. This paper introduces a 3D motion measurement of the human upper body by using an optical motion capture system for the purpose of the estimation of human upper body motions, which is based on the probability distribution of arm trajectories. In this study, by examining the characteristic of the arm trajectory, motion features are selected and classified by using the fuzzy technique. The posture of the occluded body part is probabilistically estimated by using the aggregation of the fuzzy information of arm trajectories and the constructed human upper body model. Experimental results show that the use of the system effectively works for classifying various motion patterns and estimating the occluded posture in the motion.

Motion detection in the fly is extremely fast with low computational requirements. Inspired from the fly's vision system, we focus on a real-time flight control on a miniquadrotor with fast visual feedback. In this work, an elaborated elementary motion detector (EMD) is utilized to detect local optical flow. Combined with novel receptive field templates, the yaw rate of the quadrotor is estimated through a lookup table established with this bioinspired visual sensor. A closed-loop control system with the feedback of yaw rate estimated by EMD is designed. With the motion of the other degrees of freedom stabilized by a camera tracking system, the yaw-rate of the quadrotor during hovering is controlled based on EMD feedback under real-world scenario. The control performance of the proposed approach is compared with that of conventional approach. The experimental results demonstrate the effectiveness of utilizing EMD for quadrotor control.

A series of experiments investigated the extent to which the spatial orientation of a signal line affects discrimination of its trajectory from the random trajectories of background noise lines. The orientation of the signal line was either parallel (iso-) or orthogonal (ortho-) to its motion direction and it was identical in all respects to the noise (orientation, length and speed) except for its motion direction, rendering the signal line indistinguishable from the noise on a frame-to-frame basis. We found that discrimination of ortho-trajectories was generally better than iso-trajectories. Discrimination of ortho-trajectories was largely immune to the effects of spatial jitter in the trajectory, and to variations in step size and line-length. Discrimination of iso-trajectories was reliable provided that step-size was not too short and did not exceed line length, and that the trajectory was straight. The new result that trajectory discrimination in moving line elements is modulated by line orientation suggests that ortho- and iso-trajectory discrimination rely upon two distinct mechanisms: iso-motion discrimination involves a 'motion-streak' process that combines motion information with information about orientation parallel to the motion trajectory, while ortho-motion discrimination involves extended trajectory facilitation in a network of receptive fields with orthogonal orientation tuning. © 2011 Elsevier Ltd.

This paper presents a novel design using a mouse sensor to construct a system for motion detection in normal vision environment. A mouse sensor is packed under an optical mouse for detecting the motion of mouse on desktop and sending out the data and parameters to a controller or a computer directly. This paper introduces this kind of sensor to vision motion detection field by designing and building a circuit system. The feasibility of the design is demonstrated and degree of reliability is measured by experiments performed on the designed system. Additionally the authors point out the advantages of this design in comparison with other traditional methods or devices in brief.

The paper proposes a new method for de-interlacing video. It combines simple techniques such as field insertion and edge-adaptive line averaging as well as motion-compensated (MC) methods by a robust motion detector. The simpler techniques are preferred, and the MC method is limited to places where it is really advantageous to limit processing power without sacrificing too much quality. The final experimental results show that the proposed method needs lower complexity but achieves 2 dB PSNR gain when compared to the previous method.

This paper presents a robust multi-class multi-object tracking (MCMOT) formulated by a Bayesian filtering framework. Multi-object tracking for unlimited object classes is conducted by combining detection responses and changing point detection (CPD) algorithm. The CPD model is used to observe abrupt or abnormal changes due to a drift and an occlusion based spatiotemporal characteristics of track states. The ensemble of convolutional neural network (CNN) based object detector and Lucas-Kanede Tracker (KLT) based motion detector is employed to compute the likelihoods of foreground regions as the detection responses of different object classes. Extensive experiments are performed using lately introduced challenging benchmark videos; ImageNet VID and MOT benchmark dataset. The comparison to state-of-the-art video tracking techniques shows very encouraging results.

Visual motion is fundamentally different from physical motion, because the former represents sequential retinotopic neuronal activations in time generated by a physical moving object. Traditionally, cortical direction-selective neurons are regarded as motion detector while orientation-selective neurons as contour detector. However, orientation-selective neurons also respond rigorously to motion stimuli. What is then the common neural mechanism underlying early motion processing in high mammalians? Here we demonstrate that motion trajectory only at low speed was encoded primarily as direction signal by both direction- and orientation-selective neurons preferring that motion direction, but at high speed, other groups of direction- and orientation-selective neurons with perpendicular preferences were activated to encode the motion trajectory as motion-axis information. Thus, depending on the motion speed, the combined processing of motion direction and axis by neurons with orthogonal direction and orientation preferences may serve as a fundamental principle of visual motion processing in early visual areas of high mammalians.

Event-based sensing, i.e. the asynchronous detection of luminance changes, promises low-energy, high dynamic range, and sparse sensing. This stands in contrast to whole image frame-wise acquisition using standard cameras. Recently, we proposed a novel biologically inspired efficient motion detector for such event-based input streams and demonstrated how a canonical neural circuit can improve such representations using normalization and feedback. In this contribution, we suggest how such a motion detection scheme is defined by utilizing a canonical neural circuit corresponding to the resolution of cortical columns. In addition, we develop a mapping of key computational elements of this circuit model onto neuromorphic hardware. In particular, we focus on the recently developed TrueNorth chip architecture by IBM to realize a real-time, energy-efficient and adjustable neuromorphic optical flow detector. We demonstrate the function of the computations of the canonical model and its approximate neuromorphic realization.

This paper presents a robust multi-class multi-object tracking (MCMOT) formulated by a Bayesian filtering framework. Multi-object tracking for unlimited object classes is conducted by combining detection responses and changing point detection (CPD) algorithm. The CPD model is used to observe abrupt or abnormal changes due to a drift and an occlusion based spatiotemporal characteristics of track states. The ensemble of convolutional neural network (CNN) based object detector and Lucas-Kanede Tracker (KLT) based motion detector is employed to compute the likelihoods of foreground regions as the detection responses of different object classes. Extensive experiments are performed using lately introduced challenging benchmark videos; ImageNet VID and MOT benchmark dataset. The comparison to state-of-the-art video tracking techniques shows very encouraging results.

We present a surveillance camera architecture using a one-bit motion detection algorithm for portable applications. The conventional surveillance system requires too many components and is too complicated for portable applications. In this paper, we propose a simple surveillance camera architecture with a motion detector that can detect changes of an object within its surroundings for low-power and low-cost applications. The proposed motion-detection algorithm uses one-bit images which are selected using a binary search algorithm that is decided by varying the reference voltage. We verify all the functions of the motion detector through the use of a user interface program on an FPGA board. Experimental results are also presented1.

Visual motion contains a wealth of information about self-motion as well as the three-dimensional structure of the environment. Therefore, it is of utmost importance for any organism with eyes. However, visual motion information is not explicitly represented at the photoreceptor level, but rather has to be computed by the nervous system from the changing retinal images as one of the first processing steps. Two prominent models have been proposed to account for this neural computation: the Reichardt detector and the gradient detector. While the Reichardt detector correlates the luminance levels derived from two adjacent image points, the gradient detector provides an estimate of the local retinal image velocity by dividing the spatial and the temporal luminance gradient. As a consequence of their different internal processing structure, both the models differ in a number of functional aspects such as their dependence on the spatial-pattern structure as well as their sensitivity to photon noise. These different properties lead to the proposal that an ideal motion detector should be of Reichardt type at low luminance levels, but of gradient type at high luminance levels. However, experiments on the fly visual systems provided unambiguous evidence in favour of the Reichardt detector under all luminance conditions. Does this mean that the fly nervous system uses suboptimal computations, or is there a functional aspect missing in the optimality criterion? In the following, I will argue in favour of the latter, showing that Reichardt detectors have an automatic gain control allowing them to dynamically adjust their input–output relationships to the statistical range of velocities presented, while gradient detectors do not have this property. As a consequence, Reichardt detectors, but not gradient detectors, always provide a maximum amount of information about stimulus velocity over a large range of velocities. This important property might explain why Reichardt type of computations have been demonstrated to underlie the extraction of motion information in the fly visual system under all luminance levels.

We report on a computational model of retinal motion sensitivity based on correlation-based motion detectors. We simulate object motion detection in the presence of retinal slip caused by the salamander's head movements during locomotion. Our study offers new insights into object motion sensitive ganglion cells in the salamander retina. A sigmoidal transformation of the spatially and temporally filtered retinal image substantially improves the sensitivity of the system in detecting a small target moving in place against a static natural background in the presence of comparatively large, fast simulated eye movements, but is detrimental to the direction-selectivity of the motion detector. The sigmoid has insignificant effects on detector performance in simulations of slow, high contrast laboratory stimuli. These results suggest that the sigmoid reduces the system's noise sensitivity.

Passive infrared (PIR) motion detectors, which can support long-term continuous observation, are widely used for human motion analysis. Extracting all possible trajectories from the PIR sensor networks is important. Because the PIR sensor does not log location and individual information, none of the existing methods can generate all possible human motion trajectories that satisfy various spatio-temporal constraints from the sensor activation log data. In this paper, a geometric algebra (GA)-based approach is developed to generate all possible human trajectories from the PIR sensor network data. Firstly, the representation of the geographical network, sensor activation response sequences and the human motion are represented as algebraic elements using GA. The human motion status of each sensor activation are labeled using the GA-based trajectory tracking. Then, a matrix multiplication approach is developed to dynamically generate the human trajectories according to the sensor activation log and the spatio-temporal constraints. The method is tested with the MERL motion database. Experiments show that our method can flexibly extract the major statistical pattern of the human motion. Compared with direct statistical analysis and tracklet graph method, our method can effectively extract all possible trajectories of the human motion, which makes it more accurate. Our method is also likely to provides a new way to filter other passive sensor log data in sensor networks.

Aimed at the characteristics of infrared images for moving target, such as complicated background, low signal-to-noise ratio and low contrast, a new method for complicated background suppression is proposed based on Large Field (LF) and Small Field (SF) system in the paper. Firstly, the CCD sensors are used to obtain infrared video data for moving target. Then, after analyzing the visual motion detection mechanism of insect compound eye, the paper establishes background suppression model based on LF&SF. In this model, LF is used to suppress noise of edge detection while SF is used to do nonlinear enhancement of target features and estimation of target movement direction. Finally, the paper uses the primary motion detector to integrate LF with SF. Real scene experiment shows that this proposed method has obvious advantages in suppressing the cloud cover, grass lands and other noise and enhancing target.