Search

In the last decade, the indoor localization and positioning has gained attention. The development of location based service, such as activity monitoring during training, has been one of the main drivers for this interest. In this contribution, a tracking system based on inertial measurement is proposed. The tracking system is designed for human tracking. The system has been designed to be low cost and self-configurable. To this aim, the system is able to learn all the parameters exploited in the computation. Moreover, the computational load of the system is low, so it can be implemented over wearable devices. The proposed approach has been tested in a real scenario and the results are reported.

The underground mine environment is dangerous and harsh, tracking and detecting humans based on computer vision is of great significance for mine safety monitoring, which will also greatly facilitate identification of humans using the symmetrical image features of human organs. However, existing methods have difficulty solving the problems of accurate identification of humans and background, unstable human appearance characteristics, and humans occluded or lost. For these reasons, an improved aberrance repressed correlation filter (IARCF) tracker for human tracking in underground mines based on infrared videos is proposed. Firstly, the preprocess operations of edge sharpening, contrast adjustment, and denoising are used to enhance the image features of original videos. Secondly, the response map characteristics of peak shape and peak to side lobe ratio (PSLR) are analyzed to identify abnormal human locations in each frame, and the method of calculating the image similarity by generating virtual tracking boxes is used to accurately relocate the human. Finally, using the value of PSLR and the highest peak point of the response map, the appearance model is adaptively updated to further improve the robustness of the tracker. Experimental results show that the average precision and success rate of the IARCF tracker in the five underground scenarios reach 0.8985 and 0.7183, respectively, and the improvement of human tracking in difficult scenes is excellent. The IARCF tracker can effectively track underground human targets, especially occluded humans in complex scenes.

Millimeter-wave (mmWave) radar, as an emerging technique, is increasing in popularity for human activity recognition. In contrast to traditional sensors and radars, mmWave radars give detailed information on objects from the range domain to the Doppler domain. The short wavelength allows mmWave radars to achieve a high resolution and a small antenna size, but also makes them prone to noise. In this article, we present a system framework for human detection and tracking using mmWave radars. We show that mmWave radars have good performance in indoor environments, with over 90% sensitivity. We show that using a single radar can raise a large number of false alarms due to unstable data and noise. However, with two radars the precision of the system can be improved significantly.

Automating fall risk assessment, in an efficient, non-invasive manner, specifically in the elderly population, serves as an efficient means for implementing wide screening of individuals for fall risk and determining their need for participation in fall prevention programs. We present an automated and efficient system for fall risk assessment based on a multi-depth camera human motion tracking system, which captures patients performing the well-known and validated Berg Balance Scale (BBS). Trained machine learning classifiers predict the patient’s 14 scores of the BBS by extracting spatio-temporal features from the captured human motion records. Additionally, we used machine learning tools to develop fall risk predictors that enable reducing the number of BBS tasks required to assess fall risk, from 14 to 4–6 tasks, without compromising the quality and accuracy of the BBS assessment. The reduced battery, termed Efficient-BBS (E-BBS), can be performed by physiotherapists in a traditional setting or deployed using our automated system, allowing an efficient and effective BBS evaluation. We report on a pilot study, run in a major hospital, including accuracy and statistical evaluations. We show the accuracy and confidence levels of the E-BBS, as well as the average number of BBS tasks required to reach the accuracy thresholds. The trained E-BBS system was shown to reduce the number of tasks in the BBS test by approximately 50% while maintaining 97% accuracy. The presented approach enables a wide screening of individuals for fall risk in a manner that does not require significant time or resources from the medical community. Furthermore, the technology and machine learning algorithms can be implemented on other batteries of medical tests and evaluations.

Industrial growth has increased the number of jobs hence increase the number of employees. Therefore, it is impossible to track the location of all employees in the same building at the same time as they are placed in a different department. In this work, a real-time indoor human tracking system is developed to determine the location of employees in a real-time implementation. In this work, the long-range (LoRa) technology is used as the communication medium to establish the communication between the tracker and the gateway in the developed system due to its low power with high coverage range besides requires low cost for deployment. The received signal strength indicator (RSSI) based positioning method is used to measure the power level at the receiver which is the gateway to determine the location of the employees. Different scenarios have been considered to evaluate the performance of the developed system in terms of precision and reliability. This includes the size of the area, the number of obstacles in the considered area, and the height of the tracker and the gateway. A real-time testbed implementation has been conducted to evaluate the performance of the developed system and the results show that the system has high precision and are reliable for all considered scenarios.

The number of older people needing healthcare is a growing global phenomenon. The assistance in long-term care comprises a complex of medical, nursing, rehabilitation, and social assistance services. The cost is substantial, but technology can help reduce spending by ensuring efficient health services and improving the quality of life. Advances in artificial intelligence, wireless communication systems, and nanotechnology allow the creation of intelligent home care systems avoiding hospitalization with evident cost containment. They are capable of ensuring functions of recognition of activities, monitoring of vital functions, and tracking. However, it is essential to also have information on location in order to be able to promptly intervene in case of unforeseen events or assist people in carrying out activities in order to avoid incorrect behavior. In addition, the automatic detection of physical activities performed by human subjects is identified as human activity recognition (HAR). This work presents an overview of the positioning system as part of an integrated HAR system. Lastly, this study contains each technology’s concepts, features, accuracy, advantages, and limitations. With this work, we want to highlight the relationship between HAR and the indoor positioning system (IPS), which is poorly documented in the literature.

According to Robert McSweeney, in light of a new study: “Conditions in the GCC could become so hot and humid in the coming years that staying outside for more than six hours will become difficult”. He is a climate analyst at CARBON BRIEF, a nonprofit temperature and climate analysis group. He also states that changes there can help give us an idea of what the rest of the world can expect if we do not reduce the emissions that pollute homes and factories. Because of the high temperatures in GCC countries, the effect of heat stress is very high there, which discourages shoppers and pedestrians from shopping in the open area due to the physical exertion and high risks faced by people and workers. Heat stress peaks in most Arab Gulf countries from 11:00 a.m. to 4:00 p.m. during the summer season. Heat stress is increasingly an obstacle to economic efficiency in these countries. This work designs and develops a robot that tracks shoppers and provides a cool stream of air directly around them during shopping in open areas to reduce the effect of heat stress. The robot enables us to cool the temperature around customers in the market to increase comfort. In this project, a robot was designed and manufactured to track a specific person and cool the air around him through a cool stream of air generated by the air conditioner installed inside the robot. We used a Raspberry Pi camera sensor to detect the target person and interact with a single-board computer (Raspberry Pi 3) to accomplish this design and the prototype. Raspberry Pi controls the air-conditioning robot to follow the movement of the target person. We used image processing to discover the target shopper, the control system, and then guide the bot. In the meantime, the robot must also bypass any potential obstacles that could prevent its movement and cause a collision. We made a highly efficient design that can synchronize between the software algorithm and the mechanical platform of the robot. This work is merely the combination of a cool stream of air and a robot that follows a human.

This thesis presents a framework for a mobile robot to escort a human to their destination successfully and efficiently. The proposed technique uses accurate path prediction incorporating human intention to locate the robot in front of the human while walking. Human intention is inferred by the head pose, an effective past-proven implicit indicator of intention, and fused with conventional physics-based motion prediction. The human trajectory is estimated and predicted using a particle filter because of the human's nonlinear and non-Gaussian behavior, and the robot control action is determined from the predicted human pose allowing for anticipative autonomous escorting. Experimental analysis shows that the incorporation of the proposed human intention model reduces human position prediction error by approximately 35% when turning. Furthermore, experimental validation with an omnidirectional mobile robotic platform shows escorting up to 50% more accurate compared to the conventional techniques, while achieving 97% success rate. Master of Science This thesis presents a method for a mobile robot to escort a human to their destination successfully and efficiently. The proposed technique uses human intention to predict the walk path allowing the robot to be in front of the human while walking. Human intention is inferred by the head direction, an effective past-proven indicator of intention, and is combined with conventional motion prediction. The robot motion is then determined from the predicted human position allowing for anticipative autonomous escorting. Experimental analysis shows that the incorporation of the proposed human intention reduces human position prediction error by approximately 35% when turning. Furthermore, experimental validation with an mobile robotic platform shows escorting up to 50% more accurate compared to the conventional techniques, while achieving 97% success rate. The unique escorting interaction method proposed has applications such as touch-less shopping cart robots, exercise companions, collaborative rescue robots, and sanitary transportation for hospitals.

Данная работа посвящена исследованию ÑÐ¾Ð²Ñ€ÐµÐ¼ÐµÐ½Ð½Ñ‹Ñ Ð°Ñ€Ñ Ð¸Ñ‚ÐµÐºÑ‚ÑƒÑ€ Ð½ÐµÐ¹Ñ€Ð¾Ð½Ð½Ñ‹Ñ ÑÐµÑ‚ÐµÐ¹ для решения задачи обнаружения объектов в реальном времени на изображении с камеры при помощи Ð½ÐµÐ¹Ñ€Ð¾Ð½Ð½Ñ‹Ñ ÑÐµÑ‚ÐµÐ¹. В Ñ Ð¾Ð´Ðµ исследования был проведен сравнительный анализ методов оптимизации Ð½ÐµÐ¹Ñ€Ð¾Ð½Ð½Ñ‹Ñ ÑÐµÑ‚ÐµÐ¹ для запуска на Ð²ÑÑ‚Ñ€Ð°Ð¸Ð²Ð°ÐµÐ¼Ñ‹Ñ ÑÐ¸ÑÑ‚ÐµÐ¼Ð°Ñ . На базе ÑÑƒÑ‰ÐµÑÑ‚Ð²ÑƒÑŽÑ‰Ð¸Ñ Ð¼ÐµÑ‚Ð¾Ð´Ð¾Ð² обучена и оптимизирована нейронная сеть с Ð°Ñ€Ñ Ð¸Ñ‚ÐµÐºÑ‚ÑƒÑ€Ð¾Ð¹ SSD MobileNet. Создана система отслеживания положения человека в кадре и продемонстрирована её работа на микрокомпьютере Raspberry Pi 4 Model B 4 Gb. Полученные модель сравнивается на предмет сложности интеграции, скорости работы (количество кадров в секунду), Ð²Ñ‹Ñ‡Ð¸ÑÐ»Ð¸Ñ‚ÐµÐ»ÑŒÐ½Ñ‹Ñ Ð¼Ð¾Ñ‰Ð½Ð¾ÑÑ‚ÐµÐ¹, Ð½ÐµÐ¾Ð±Ñ Ð¾Ð´Ð¸Ð¼Ñ‹Ñ Ð´Ð»Ñ Ð¸Ñ Ñ€Ð°Ð±Ð¾Ñ‚Ñ‹. В разработке приложения использовались компоненты библиотеки TensorFlow Lite., This paper is devoted to the study of modern neural network architectures for solving the problem of detecting objects in real time in the image from a camera using neural networks. A comparative review of methods for optimizing neural networks for embedded systems was carried out as part of the research. Based on existing methods, a neural network with SSD MobileNet architecture was trained and optimized. A system for tracking the position of a person in the frame has been created and its work has been demonstrated on a Raspberry Pi 4 Model B 4GB microcomputer. The resulting model is compared for the complexity of integration, speed of work (number of frames per second), computing power required for their work. In the development of the application, the components of the TensorFlow Lite library were used.

Tracking moving objects has been an issue in recent years of computer vision and image processing and human tracking makes it a more significant challenge. This category has various aspects and wide applications, such as autonomous deriving, human-robot interactions, and human movement analysis. One of the issues that have always made tracking algorithms difficult is their interaction with goal recognition methods, the mutable appearance of variable aims, and simultaneous tracking of multiple goals. In this paper, a method with high efficiency and higher accuracy was compared to the previous methods for tracking just objects using imaging with the fixed camera is introduced. The proposed algorithm operates in four steps in such a way as to identify a fixed background and remove noise from that. This background is used to subtract from movable objects. After that, while the image is being filtered, the shadows and noises of the filmed image are removed, and finally, using the bubble routing method, the mobile object will be separated and tracked. Experimental results indicated that the proposed model for detecting and tracking mobile objects works well and can improve the motion and trajectory estimation of objects in terms of speed and accuracy to a desirable level up to in terms of accuracy compared with previous methods.

We address the use of backscattered mm-wave radio signals to track humans as they move within indoor environments. The common approach in the literature leverages the extended Kalman filter (EKF) method, which however undergoes a severe performance degradation when the system evolution model is highly non-linear or presents long-term time dependencies among the system states. In this work, we propose an original model-free tracking procedure based on denoising autoencoders and sequence-to-sequence neural networks, showing its superior performance with respect to state-of-the-art methods. Our architecture can be trained in either a supervised or unsupervised manner, trading tracking accuracy for flexibility. The proposed system is tested on our own measurements, obtained with a 77 GHz radar on single and multiple subjects simultaneously moving in an indoor space. The results are compared against the ground truth trajectories from a motion tracking system, obtaining average tracking errors as low as 12 cm. This conference paper is subject to ©2020 IEEE Published version of the paper can be found in the proceedings of RadarConf2020.

Action recognition has gained great attention in automatic video analysis, greatly reducing the cost of human resources for smart surveillance. Most methods, however, focus on the detection of only one action event for a single person in a well-segmented video, rather than the recognition of multiple actions performed by more than one person at the same time for an untrimmed video. In this paper, we propose a deep learning-based multiple-person action recognition system for use in various real-time smart surveillance applications. By capturing a video stream of the scene, the proposed system can detect and track multiple people appearing in the scene and subsequently recognize their actions. Thanks to high resolution of the video frames, we establish a zoom-in function to obtain more satisfactory action recognition results when people in the scene become too far from the camera. To further improve the accuracy, recognition results from inflated 3D ConvNet (I3D) with multiple sliding windows are processed by a nonmaximum suppression (NMS) approach to obtain a more robust decision. Experimental results show that the proposed method can perform multiple-person action recognition in real time suitable for applications such as long-term care environments.

Radio Frequency Identification (RFID) technology has been widely used in indoor location tracking, especially serving human beings, due to its advantage of low cost, non-contact communication, resistance to hostile environments and so forth. Over the years, many indoor location tracking methods have been proposed. However, tracking mobile RFID readers in real-time has been a daunting task, especially for achieving high localization accuracy. In this paper, we propose a new Mobile RFID (M-RFID)-based Localization approach for Indoor Human Tracking, named MRLIHT. Based on the M-RFID model where RFID readers are equipped on the moving objects (human beings) and RFID tags are fixed deployed in the monitoring area, MRLIHT implements the real-time indoor location tracking effectively and economically. First, based on the readings of multiple tags detected by an RFID reader simultaneously, MRLIHT generates the response regions of tags to the reader. Next, MRLIHT determines the potential location region of the reader where two algorithms are devised. Finally, MRLIHT estimates the location of the reader by dividing the potential location region of the reader into finer-grained grids. The experimental results demonstrate that the proposed MRLIHT performs well in both accuracy and scalability.

In this work, we investigate the use of backscattered mm-wave radio signals for the joint tracking and recognition of identities of humans as they move within indoor environments. We build a system that effectively works with multiple persons concurrently sharing and freely moving within the same indoor space. This leads to a complicated setting, which requires one to deal with the randomness and complexity of the resulting (composite) backscattered signal. The proposed system combines several processing steps: at first, the signal is filtered to remove artifacts, reflections and random noise that do not originate from humans. Hence, a density-based classification algorithm is executed to separate the Doppler signatures of different users. The final blocks are trajectory tracking and user identification, respectively based on Kalman filters and deep neural networks. Our results demonstrate that the integration of the last-mentioned processing stages is critical towards achieving robustness and accuracy in multi-user settings. Our technique is tested both on a single-target public dataset, for which it outperforms state-of-the-art methods, and on our own measurements, obtained with a 77 GHz radar on multiple subjects simultaneously moving in two different indoor environments. The system works in an online fashion, permitting the continuous identification of multiple subjects with accuracies up to 98%, e.g., with four subjects sharing the same physical space, and with a small accuracy reduction when tested with unseen data from a challenging real-life scenario that was not part of the model learning phase., Comment: 16 pages, 12 figures, 5 tables

Pedestrian detection and tracking have become an important field in the computer vision research area. This growing interest, started in the last decades, might be explained by the multitude of potential applications that could use the results of this research field, e.g. robotics, entertainment, surveillance, care for the elderly and disabled, and content-based indexing. In this survey paper, vision-based pedestrian detection systems are analysed based on their field of application, acquisition technology, computer vision techniques and classification strategies. Three main application fields have been individuated and discussed: video surveillance, human-machine interaction and analysis. Due to the large variety of acquisition technologies, this paper discusses both the differences between 2D and 3D vision systems, and indoor and outdoor systems. The authors reserved a dedicated section for the analysis of the Deep Learning methodologies, including the Convolutional Neural Networks in pedestrian detection and tracking, considering their recent exploding adoption for such a kind systems. Finally, focusing on the classification point of view, different Machine Learning techniques have been analysed, basing the discussion on the classification performances on different benchmark datasets. The reported results highlight the importance of testing pedestrian detection systems on different datasets to evaluate the robustness of the computed groups of features used as input to classifiers.

© 2018 The Authors. Published by Elsevier Ltd. This paper proposes a weighted resampling method for particle filter which is applied for human tracking on active camera. The proposed system consists of three major parts which are human detection, human tracking, and camera control. The codebook matching algorithm is used for extracting human region in human detection system, and the particle filter algorithm estimates the position of the human in every input image. The proposed system in this paper selects the particles with highly weighted value in resampling, because it provides higher accurate tracking features. Moreover, a proportional-integral-derivative controller (PID controller) controls the active camera by minimizing difference between center of image and the position of object obtained from particle filter. The proposed system also converts the position difference into pan-tilt speed to drive the active camera and keep the human in the field of view (FOV) camera. The intensity of image changes overtime while tracking human therefore the proposed system uses the Gaussian mixture model (GMM) to update the human feature model. As regards, the temporal occlusion problem is solved by feature similarity and the resampling particles. Also, the particle filter estimates the position of human in every input frames, thus the active camera drives smoothly. The robustness of the accurate tracking of the proposed system can be seen in the experimental results.

Human segmentation and tracking often use the outcome of person detection in the video. Thus, the results of segmentation and tracking depend heavily on human detection results in the video. With the advent of Convolutional Neural Networks (CNNs), there are excellent results in this field. Segmentation and tracking of the person in the video have significant applications in monitoring and estimating human pose in 2D images and 3D space. In this paper, we performed a survey of many studies, methods, datasets, and results for human segmentation and tracking in video. We also touch upon detecting persons as it affects the results of human segmentation and human tracking. The survey is performed in great detail up to source code paths. The MADS (Martial Arts, Dancing and Sports) dataset comprises fast and complex activities. It has been published for the task of estimating human posture. However, before determining the human pose, the person needs to be detected as a segment in the video. Moreover, in the paper, we publish a mask dataset to evaluate the segmentation and tracking of people in the video. In our MASK MADS dataset, we have prepared 28 k mask images. We also evaluated the MADS dataset for segmenting and tracking people in the video with many recently published CNNs methods.

A human detection and tracking method using a stereo camera is presented in this paper. Two human detection methods are independently implemented, and the results are combined to reduce misdetection and nondetection. The tracking step is based on particle filtering. In the data association phase, we introduce three features: distance, traveling direction, and color. The color feature is obtained from every segmented detection window. Eligibility and co-occurrence of the blocks provide robustness to occlusion. The proposed method is tested by using measurement data at the entrance of a building, where occlusion is frequently observed. The experiments demonstrate improved tracking performance over standard particle filtering.

In recent years, automated human tracking over camera networks is getting essential for video surveillance. The tasks of tracking human over camera networks are not only inherently challenging due to changing human appearance, but also have enormous potentials for a wide range of practical applications, ranging from security surveillance to retail and health care. This review paper surveys the most widely used techniques and recent advances for human tracking over camera networks. Two important functional modules for the human tracking over camera networks are addressed, including human tracking within a camera and human tracking across non-overlapping cameras. The core techniques of human tracking within a camera are discussed based on two aspects, i.e., generative trackers and discriminative trackers. The core techniques of human tracking across non-overlapping cameras are then discussed based on the aspects of human re-identification, camera-link model-based tracking and graph model-based tracking. Our survey aims to address existing problems, challenges, and future research directions based on the analyses of the current progress made toward human tracking techniques over camera networks.

RFID (radio-frequency identification) technology is rapidly emerging for the localization of moving objects and humans. Due to the blockage of radio signals by the human body, the localization accuracy achieved with a single tag is not satisfactory. This paper proposes a method based on an RFID tag array and laser ranging information to address the localization of live moving objects such as humans or animals. We equipped a human with a tag array and calculated the phase-based radial velocity of every tag. The laser information was, first, clustered through the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm and then laser-based radial velocity was calculated. This velocity was matched with phase-based radial velocity to get best matching clusters. A particle filter was used to localize the moving human by fusing the matching results of both velocities. Experiments were conducted by using a SCITOS G5 service robot. The results verified the feasibility of our approach and proved that our approach significantly increases localization accuracy by up to 25% compared to a single tag approach.

International audience; The purpose of this paper is to investigate the problem of 3D human tracking in complex environments using a particle filter with images captured by a catadioptric vision system. This issue has been widely studied in the literature on RGB images acquired from conventional perspective cameras, while omnidirectional images have seldom been used and published research works in this field remains limited. In this study, the Riemannian varieties was considered in order to compute the gradient on spherical images and generate a robust descriptor used along with an SVM classifier for human detection. Original likelihood functions associated with the particle filter are proposed, using both geodesic distances and overlapping regions between the silhouette detected in the images and the projected 3D human model. Our approach was experimentally evaluated on real data and showed favorable results compared to machine learning based techniques about the 3D pose accuracy. Thus, the Root Mean Square Error (RMSE) was measured by comparing estimated 3D poses and truth data, resulting in a mean error of 0.065 m when walking action was applied.

A complete centralized processing framework is proposed for human tracking using multistatic radar in the foliage-penetration environment. The configuration of the multistatic radar system is described. Primary attention is devoted to time of arrival (TOA) estimation and target localization. An improved approach that takes the geomet- rical center as the TOA estimation of the human target is given. The minimum mean square error paring (MMSEP) approach is introduced for multi-target localization in the multistatic radar system. An improved MMSEP algorithm is proposed using the maximum velocity limitation and the global nearest neighbor criterion, efficiently decreasing the computational cost of MMSEP. The experimental re- sults verify the effectiveness of the centralized processing framework.

A human tracking system based on mobile agent technologies has been proposed to achieve automatic human tracking function. In this system, each target person is tracked automatically by its mobile agent moving among cameras in which a person is detected. The current system utilizes an algorithm to predict which camera detects the target person. The algorithm needs information from all cameras about their monitoring ranges. If one monitoring range is updated by a pan / tilt / zoom operation or some other reason, the whole calculation to determine the relationship between camera nodes must be performed accordingly. In order to solve this problem, we propose in this paper an algorithm that uses only localized node information from each camera and its neighboring camera. With this proposed algorithm, each camera is able to calculate its neighbor nodes without obtaining the monitoring ranges of all cameras. This enables the construction of robust human tracking systems.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. At present, traditional visual‐based surveillance systems are becoming impractical, inefficient, and time‐consuming. Automation‐based surveillance systems appeared to overcome these limitations. However, the automatic systems have some challenges such as occlusion and retaining images smoothly and continuously. This research proposes a weighted resampling particle filter approach for human tracking to handle these challenges. The primary functions of the proposed system are human detection, human monitoring, and camera control. We used the codebook matching algorithm to define the human region as a target and track it, and we used the practical filter algorithm to follow and extract the target information. Consequently, the obtained information was used to configure the camera control. The experiments were tested in various environments to prove the stability and performance of the proposed system based on the active camera.

This paper presents a robust and computationally efficient method for human detection and tracking. The unique feature of this method is that it has dedicated threads for human detection and camera control for human tracking. Moreover, it works with infra-red on and infra-red off. The method consists of five parts – training image acquisition, background subtraction, feature extraction, system training, and system testing. Firstly, some sample video clips have been taken with an IP camera for initial system implementation. The clips are then filtered to separate background and foreground. After that, some morphological operations are carried out to identify the most significant motion in the foreground. Those parts are cropped with some extra area and used to train a multiclass support vector machine (SVM) along with an image subset of the people detection dataset of The National Institute for Research in Computer Science and Control (French: Institut National de Recherche en Informatique et en Automatique, INRIA). A total of 597 images have been used as positive images and a total of 662 images have been used as negative images. Average detection accuracy of the system without infra-red is 89.37% and average detection accuracy of the system with infra-red is 72.66%. Therefore the average detection accuracy is 81.1%. We conclude (using dependent probabilistic analysis) that our system performs on an average of 89.37% accuracy based on our frame based analysis of video feeds. Keywords Human tracking; computer vision; surveillance;background subtraction; HOG For More Details: http://it-in-industry.com/itii_papers/2013/1113itii04.pdf

Multiple human tracking (MHT) is a fundamental task in many computer vision applications. Appearance-based approaches, primarily formulated on RGB data, are constrained and affected by problems arising from occlusions and/or illumination variations. In recent years, the arrival of cheap RGB-depth devices has led to many new approaches to MHT, and many of these integrate colour and depth cues to improve each and every stage of the process. In this survey, the authors present the common processing pipeline of these methods and review their methodology based (a) on how they implement this pipeline and (b) on what role depth plays within each stage of it. They identify and introduce existing, publicly available, benchmark datasets and software resources that fuse colour and depth data for MHT. Finally, they present a brief comparative evaluation of the performance of those works that have applied their methods to these datasets.

Human activity regions (HARs) are human-centric semantic partitions where observing and/or interacting with humans is likely in indoor environments. HARs are useful for achieving successful human-robot interaction, such as in safe navigation around a building or to know where to be able to assist humans in their activities. In this thesis, a system is designed for generating HARs automatically based on data recorded by robots. This approach to generating HARs is to cluster the areas that are commonly associated with frequent human presence. In order to detect human positions, we employ state-of-the-art perception techniques. The environment that the robot patrols is assumed to be an indoor environment such as an office. We show how we can generate HARs in correct regions by clustering human position data. The experimental evaluations show that we can do so in different indoor environments, with data acquired from different sensors and that the system can handle noise. Mänskliga aktivitetsregioner, HARs (Human Activity Regions) är människocentreraderegioner som ger en semantisk partitionering av inomhusmiljöer. HARs är användbara för att uppnå väl fungerande människarobot- interaktioner. I denna avhandling utformas ett system för att generera HARs automatiskt baserat på data från robotar. Detta görs genom att klustra observationer av människor för att på så vis få fram de områden som är associerade med frekvent mänsklig närvaro. Experiment visar att systemet kan hantera data som registrerats av olika sensorer i olika inomhusmiljöer och att det är robust. Framförallt genererar systemet en pålitlig partitionering av miljön.

There is an urgent need for intelligent home surveillance systems to provide home security, monitor health conditions, and detect emergencies of family members. One of the fundamental problems to realize the power of these intelligent services is how to detect, track, and identify people at home. Compared to RFID tags that need to be worn all the time, vision-based sensors provide a natural and nonintrusive solution. Observing that body appearance and body build, as well as face, provide valuable cues for human identification, we model and record multi-view faces, full-body colors and shapes of family members in an appearance database by using two Kinects located at a home's entrance. Then the Kinects and another set of color cameras installed in other parts of the house are used to detect, track, and identify people by matching the captured color images with the registered templates in the appearance database. People are detected and tracked by multisensor fusion (Kinects and color cameras) using a Kalman filter that can handle duplicate or partial measurements. People are identified by multimodal fusion (face, body appearance, and silhouette) using a track-based majority voting. Moreover, the appearance-based human detection, tracking, and identification modules can cooperate seamlessly and benefit from each other. Experimental results show the effectiveness of the human tracking across multiple sensors and human identification considering the information of multi-view faces, full-body clothes, and silhouettes. The proposed home surveillance system can be applied to domestic applications in digital home security and intelligent healthcare.

In this paper, we present a statistical model of an indirect path generated in an ultra-wideband (UWB) human tracking scenario. When performing moving target detection, an indirect path signal can generate ghost targets that may cause a false alarm. For this purpose, we performed radar measurements in an indoor environment and established a statistical model of an indirect path based on the measurement data. The proposed model takes the form of a modified Saleh-Valenzuela model, which is used in a UWB channel model. An application example of the proposed model for mitigating false alarms is also presented.

Human detection and tracking is an important task in artificial intelligent robotic systems, which usually require a robust target detector to work in a variety of circumstances. In complex environments where a camera is installed on a wheeled mobile robot, this task becomes much more difficult. In this paper, we present a real-time remote-control system for human detection, tracking, security, and verification in such challenging environments. Such a system is useful for security monitoring, data collection, and experimental purpose. In the proposed system, a Kinect RGB-D camera from Microsoft is used as a visual sensing device for the design of human detection and tracking. We also implemented a remote-control system on a four-wheel mobile platform with a Robot Operating System (ROS). By combining these two designs, a wireless controlled mobile platform having the feature of real-time human monitoring can be realized to handle this task efficiently. Experimental results validate the performance of the proposed system for wirelessly controlling the mobile robot to track humans in a real-world environment.

In this paper, we present a human tracking algorithm that can work robustly in complex environments such that serious occlusion, various appearances and abrupt motion changes occur in the scenario. Our tracking framework is well known particle filter based on Condensation algorithm. In the observation model of the particle filter, we establish RAM(Reliable Appearance Model) which exhibits high discriminative performance in particular for human tracking. The RAM is to describe a target as features from local descriptors. In order to extract practical features from a larger number of local descriptors for robust tracking, the features were employed by boosting algorithm. The components of the features are utilized color and shape based-models. Experimental results demonstrate that our approach tracks the target accurately and reliably when position and scale are changing as well as occurrence of occlusion., SICE Annual Conference 2013 - International conference on Instrumentation, Control, Information Technology and System Integration, September 14-17, 2013, Nagoya University, Nagoya, Japan

This study aimed to identify a specific human with the accurate position measurements by integrating the networked laser range finders in the intelligent space and an wearable acceleration sensor with the human. A human identification method based on associating detections of walking behaviors from both sensors is proposed. Parameter optimization was also performed for improvement of identification. System configurations and implementations using RT component are also described. Some experimental results showed a feasibility of the proposed method. In the proposed system, only walking detection results are communicated among sensors. The proposed system with small data communication is effective for networked sensor systems like intelligent spaces.

International audience; This paper addresses the problem of 3D tracking of human gesture for buying behavior estimation. The top view of the customers, which has been little treated for human tracking, is exploited in this particular context. This point of view avoids occlusion except for those of the arms. We propose an hybrid 3D-2D tracking method based on the particle filtering framework, which uses the exclusion principle to separate the observation related to each customer and deals with multi-person tracking. The head and shoulders are tracked in the 2D space, while the arms are tracked in the 3D space: these are the spaces where they are the most descriptive. We validate our method both experimentally, so as to obtain qualitative results, and on-site. We demonstrated that it makes a good estimation for various cases and situations in real-time ((Formula presented.)40 fps).

Relying on the commonsense knowledge that the trajectory of any physical entity in the spatio-temporal domain is continuous, we propose a heuristic data association technique. The technique is used in conjunction with an Extended Kalman Filter (EKF) for human tracking under occlusion. Our method is capable of tracking moving objects, maintain their state hypothesis even in the period of occlusion, and associate the target reappeared from occlusion with the existing hypothesis. The technique relies on the estimation of the reappearance event both in time and location, accompanied with an alert signal that would enable more intelligent behavior (e.g. in path planning). We implemented the proposed method, and evaluated its performance with real-world data. The result validates the expected capabilities, even in case of tracking multiple humans simultaneously.

Human or face tracking in a diverse environment is very important for various applications in computer vision, especially for video surveillance. Usually, a color cue offers many advantages over motion or geometric information which cannot robustly handle partial occlusion, rotation, scale and resolution changes. In this paper, we present a robust face tracking system by employing a color-based particle filter. The face detection technique is realized based on a Haar-like features algorithm. Here we exploit skin color cues for face-tracking, and it also proposes a body-part particle distribution system. This system is robust against occlusion by human or others and it can perform the tracking in real-time. We conducted experiments in both indoor and outdoor environments, with either a single or multiple persons in a view. Based on the color-based and body-part particle filter, we tracked a person's face satisfactorily by a developed simple robot system.

We analyze the performance of a novel human tracking system, which uses the electric near field to sense human presence. The positioning accuracy with moving targets is measured using raw observations, observation centroids and Kalman filtered centroids. In addition to this, the multi-target discrimination performance is studied with two people and a Rao-Blackwellized Monte Carlo data association algorithm. A reel-based triangulation system is used as the reference positioning system. The mean positioning error for five test subjects walking at different speeds is 21 centimeters. The discrimination performance is 90% when the distance between the two people is over 0.8 meters. With distance over 1.1 meters the discrimination performance is 99%.

本論文は,可変照明下における逐次更新型色ヒストグラムを用いた,人物追跡法を提案する.提案法は,HSV色空間のうちH-S平面上の色ヒストグラムを用いて,対象の色ヒストグラムモデルを取得し,変化する照明のもとでそれを逐次更新しながら対象を追跡する.色モデルを更新する場合,対象色以外の色によって色モデルが発散するという問題がある.提案法では,色モデル上にカーネル関数として非等方性非対称ガウス状分布による重みを設け,逐次更新による色モデルの発散を防ぐ.パン・チルトカメラを用いた色追跡システムを構築し,室内灯の点滅による時間的に不均一な室内照明環境と,日陰が存在する屋外の空間的に不均一な照明環境下で特定色の上着を着用した人物の追跡実験を行い,どちらの環境下においても良好な追跡結果を得た.In this paper, we propose an adaptive color extraction technique employing an update type two-dimensional color histogram model. A two-dimensional color histogram model using an H-S plane in the HSV color space is updated every time, and it extracts object's color that varies under changeable illumination. When updating the color model, color of the background area might be included in a color model as a noise. By such a miss update, a color model will be overflowed with non-object's colors. We define an asymmetric anisotropic Gaussian-like distribution in order to estimate a changeable color area of an object, and employ it as kernel function for providing weight coefficients to the model update. We have developed a color tracking system using a pan-tilt camera, and performed tracking experiments in two different environments, i.e., an indoor environment with changeable illumination and an outdoor environment under sunshine. As a result, the developed system kept tracking an object successfully even though the illumination has changed in both environments.

Tracking-by-detection based on segmentation, Kalman predictions and LSAP association.Occlusion management: SVM kernel metric for GCH+LBP+HOG image features.Overall performance near to 85% while tracking under occlusions in CAVIAR dataset.Human behavior analysis (exits, loitering, etc.) in naturalistic scenes in shops.Real-time multi-camera performance with a processing capacity near to 50fps/camera. Expert video-surveillance systems are a powerful tool applied in varied scenarios with the aim of automatizing the detection of different risk situations and helping human security officers to take appropriate decisions in order to enhance the protection of assets. In this paper, we propose a complete expert system focused on the real-time detection of potentially suspicious behaviors in shopping malls. Our video-surveillance methodology contributes several innovative proposals that compose a robust application which is able to efficiently track the trajectories of people and to discover questionable actions in a shop context. As a first step, our system applies an image segmentation to locate the foreground objects in scene. In this case, the most effective background subtraction algorithms of the state of the art are compared to find the most suitable for our expert video-surveillance application. After the segmentation stage, the detected blobs may represent full or partial people bodies, thus, we have implemented a novel blob fusion technique to group the partial blobs into the final human targets. Then, we contribute an innovative tracking algorithm which is not only based on people trajectories as the most part of state-of-the-art methods, but also on people appearance in occlusion situations. This tracking is carried out employing a new two-step method: (1) the detections-to-tracks association is solved by using Kalman filtering combined with an own-designed cost optimization for the Linear Sum Assignment Problem (LSAP); and (2) the occlusion management is based on SVM kernels to compute distances between appearance features such as GCH, LBP and HOG. The application of these three features for recognizing human appearance provides a great performance compared to other description techniques, because color, texture and gradient information are effectively combined to obtain a robust visual description of people. Finally, the resultant trajectories of people obtained in the tracking stage are processed by our expert video-surveillance system for analyzing human behaviors and identifying potential shopping mall alarm situations, as are shop entry or exit of people, suspicious behaviors such as loitering and unattended cash desk situations. With the aim of evaluating the performance of some of the main contributions of our proposal, we use the publicly available CAVIAR dataset for testing the proposed tracking method with a success near to 85% in occlusion situations. According to this performance, we corroborate in the presented results that the precision and efficiency of our tracking method is comparable and slightly superior to the most recent state-of-the-art works. Furthermore, the alarms given off by our application are evaluated on a naturalistic private dataset, where it is evidenced that our expert video-surveillance system can effectively detect suspicious behaviors with a low computational cost in a shopping mall context., Ministerio de Economía y Competitividad, Comunidad de Madrid

In HRI applications, tracking performance should not be evaluated as a passive sensing behavior, but by considering it as an active process, where the human is involved within the loop. We foresee that the presence of the human being, actively participating in the interaction, improves a tracker performance with a limited additional effort. We tested a tracking approach into a HRI scenario, modeled as a game, measuring both quantitative and qualitative performance.

A method of detecting speech events in a multiple-sound-source condition using audio and video information is proposed. For detecting speech events, sound localization using a microphone array and human tracking by stereo vision is combined by a Bayesian network. From the inference results of the Bayesian network, information on the time and location of speech events can be known. The information on the detected speech events is then utilized in the robust speech interface. A maximum likelihood adaptive beamformer is employed as a preprocessor of the speech recognizer to separate the speech signal from environmental noise. The coefficients of the beamformer are kept updated based on the information of the speech events. The information on the speech events is also used by the speech recognizer for extracting the speech segment.

In the last couple of decades technology has made its way into our everyday lives including our homes, our offices and the vehicles we use for travelling. Many modern devices interact with humans in a more or less intuitive way and some of them use cameras for observing or interacting with humans. Nevertheless, teaching a machine to detect humans in an image or a video is a very difficult task. There are many aspects that contribute to the complexity of this task, such as the many variations in the humans' perceived appearances: their constitution, the clothes they wear and the dynamic nature of the activities performed by humans. The focus of this thesis is on the development of reliable algorithms for real-time vision-based human detection and tracking in indoor as well as in outdoor applications. In order to achieve this, the algorithms presented in this thesis were developed for traditional passive cameras, as they perform well in both environments. The novel approaches for vision-based human detection and tracking are presented for three different applications: gait analysis, pedestrian detection and human-robot interaction. All these approaches have in common the need for real-time human detection and tracking in video sequences in order to extract application-specific data regarding the tracked human. In order to cope with human detection and tracking as a computational expensive task, novel hardware-specific optimizations of the proposed image processing algorithms are presented, that allow the algorithms to run in real-time. For this purpose GPU implementations are presented for pedestrian detection and the processing times are compared to CPU and FPGA implementations. In the case of human-robot interaction the real-time human tracking is achieved by using distributed computing.

Service robots are not only expected to navigate within the environment, as they also will may with people. Human tracking by mobile robots is essential for service robots and human interaction applications. In this work, the goal is to add a more natural robot–human following in front based on the normal human gait model. This approach proposes implementing and evaluating a human–robot interaction strategy, using the integration of a LRF (Laser Range Finder) tracking of human legs with wearable IMU (Inertial Measurement Unit) sensors for capturing the human movement during the gait. The work was carried out in four stages: first, the definition of the model of human–robot interaction and the control proposal were developed. Second, the parameters based on the human gait were estimated. Third, the robot and sensor integration setup are also proposed. Finally, the description of the algorithm for parameters detection is presented. In the experimental study, despite of the continuous oscillation during the walking, the parameters estimation was precise and unbiased, showing also repeatability with human linear velocities changes. The controller was evaluated with an eight-shaped curve, showing the stability of the controller even with sharp changes in the human path during real experiments. Fil: Cifuentes, Carlos. Universidade Federal Do Espirito Santo. Centro Tecnologico. Departamento de Ingenieria Electrica; Brasil Fil: Freire Bastos, Teodiano. Universidade Federal Do Espirito Santo. Centro Tecnologico. Departamento de Ingenieria Electrica; Brasil Fil: Carelli Albarracin, Ricardo Oscar. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico San Juan. Instituto de Automática; Argentina Fil: Frizera Neto, Anselmo. Universidade Federal Do Espirito Santo. Centro Tecnologico. Departamento de Ingenieria Electrica; Brasil

Multiple camera systems may be divided into multiple overlapping and non-overlapping camera systems. Both are widely used for image and video analysis tasks, object analysis, computer graphics and surveillance. Visual matching and tracking of humans in multiple camera systems is an important component of many such systems. For both overlapping and non-overlapping camera systems, recognising the same person in different cameras is the main challenge. Different methods have been developed to match and track humans in video. This thesis proposes novel methods for human matching and tracking in multiple camera systems, both overlapping and non-overlapping. It presents a frame-based human matching method for multiple cameras that combines feature point descriptions and geometric constraints between feature points to achieve better human matching results. It also presents a sequence-based human appearance matching method, based on model-adaptive Mean-Shift tracking and key patch selection to improve the human matching accuracy. Through key patch re-initialization, humans can also be tracked in multiple camera systems. The proposed methods have been tested on videos obtained from the Smart Room in School of Computer Science and Engineering, University of New South Wales. Experimental analysis presented in this thesis shows that the proposed methods can match and track humans stably.

This paper presents a real time approach to track the human body pose in the 3D space. For the buying behavior analysis, the camera is placed on the top of the shelves, above the customers. In this top view, the markerless tracking is harder. Hence, we use the depth cue provided by the kinect that gives discriminative features of the pose. We introduce a new 3D model that are fitted to these data in a particle filter framework. First the head and shoulders position is tracked in the 2D space of the acquisition images. Then the arms poses are tracked in the 3D space. Finally, we demonstrate that an efficient implementation provides a real-time system.

In recent years, the number of the cases of information leaks is increasing. Companies and Research Institutions make various actions against information thefts and security accidents. One of the actions is adoption of the crime prevention system, including the monitoring system by surveillance cameras. In order to solve difficulties of multiple cameras monitoring, we develop the automatic human tracking system using mobile agents through multiple surveillance cameras to track target persons. In this paper, we develop the monitor which confirms mobile agents tracing target persons, and the simulator of video picture analysis to construct the tracking algorithm., {"references":["","D. B. Lange and M. Oshima, \"Seven good reasons for mobile agents,\"\nCommunications of the ACM, vol. 42, no. 3, pp. 88– 89, 1999.","H. Kakiuchi, T. Kawamura, T. Shimizu and K. Sugahara, An Algorithm\nto Determine Neighbor Nodes for Automatic Human Tracking System,\n2009 IEEE International Conference On Electro/Information\nTechnology, Wind- sor, Canada, 2009, pp. 96–102.","Open Service Gateway Initiative Alliance, \"OSGi Alliance Specifications\nOSGi Service Platform Release 1\",\nhttp://www.osgi.org/Specifications/HomePag, last access May 2011.\n[Online].","Y. Yao, C.-H Chen, B. Abidi, D. Page, A. Koschan and M. Abidi, Sensor\nPlanning for Automated and Persistent Object Tracking with Multiple\nCameras, CVPR2008, 2008.","U.M.Erdem,S.Sclaroff,Automatedcameralayoutto satisfy task-specific\nand floor plan-specific coverage requirements, CVIO2006, Vol.103,\nNo.3, 2006, pp. 156–169.","D. G. Lowe, \"Distinctive image features from scale-invariant keypoints,\"\nInternational Journal of Computer Vision, 2004, vol. 60, no. 2, pp.\n91–110"]}

This thesis presents a robust real-time vision-based 3D multiple human tracker capable of automatically identifying, labelling and tracking multiple humans. The primary contribution is a methodology to improve robustness and integration in real-world scenarios. The system consists of 2 stages, 1. a vision based human tracking system using multiple visual cues and a robust occlusion handler, 2. a machine learning based module for intelligent multi-modal fusion and robust to drastic changes in lighting. The fusion module performs an on line analysis of image parameters influencing the performance of the tracker. Thereafter, optimal weights are generated for each visual modality for the current scene. The thesis also proposes a novel approach to validate the 3D multiple human tracking system through zero-error ground truth data. It shows integration into variety of distributed robotic systems being used in real world applications. Diese Dissertation beschreibt ein robustes System zur Verfolgung mehrerer Menschen in 3D mittels Bilddaten. Das System erkennt, markiert und verfolgt Personen automatisch in Echtzeit. Der primäre Beitrag der Arbeit ist eine Methodik zur Steigerung der Robustheit der Verfolgung und die praxisrelevante Umsetzung. Es handelt sich um ein zweistufiges System. Die erste Stufe ist ein bildbasiertes System zur Personenverfolgung durch Verwendung multipler visueller Reize und robuster Verdeckungsbehandlung, die zweite eine Einheit zur intelligenten, multi-modalen Fusion mit Beleuchtungsinvarianz mittels maschinellen Lernens. Die Fusion analysiert die aktuellen Bildparameter, welche die Leistung des Algorithmus beeinflussen und errechnet daraus für jede einzelne Modalität optimale Gewichte. Weiterhin wird ein neuer Ansatz zur Validierung der Ergebnisse, sowie die Anwendung an einer Vielzahl verteilter, praxisnaher Robotiksysteme vorgestellt.

Tracking human is an important and challenging problem in video-based intelligent robot systems. In this paper, a vision-based human tracking system is supposed to provide sensor input for vision-based control of a mobile robot that works in a team helping the human co-worker. A comparison between NARX neural network and Kalman filter in solving the prediction problem of human tracking in robot vision is presented. After collecting video data from a robot, simulation results obtained from the Kalman filter model are used to compare with the simulation results obtained from the NARX Neural network.

A UHF RFID antenna is proposed for human tracking applications. The antenna size is adequately designed in order to satisfy special requirements: compact antenna size, read range greater than 3 m, respect of UHF RFID emission standards avoiding spurious emission. The antenna design is carried out by co-simulation between electromagnetic fields and circuit solvers.

In mobile and ubiquitous computing environments, acquisition of contextual information about a user situation is necessary to provide useful services. Although the definition of user context may change according to the situation or the service used, contextual information about who, where, and when are considered to be essential. We have built a test bed with multiple sensors: floor pressure sensors, RFID (radio frequency identification) tag systems, and cameras, to carry out experiments to detect the positions of users and track their movement. The conventional background subtraction method by using cameras was used for moving object detection and tracking. In this paper, we propose knowledge application and parameter adaptation in the background subtraction method. The results are presented to show that the proposed method decreases the detection errors.

This thesis presents a multi-feature histogram approach to track a person in thermal vision. Illumination variation is a primary constraint in the performance of object tracking in visible spectrum. Thermal infrared (IR) sensor, which measures the heat energy emitted from an object, is less sensitive to illumination variations. Therefore, thermal vision has immense advantage in object tracking in varying illumination conditions. Kernel based approaches such as mean shift tracking algorithm which uses a single feature histogram for object representation, has gained popularity in the field of computer vision due its efficiency and robustness to track non-rigid object in significant complex background. However, due to low resolution of IR images the gray level intensity information is not sufficient enough to give a strong cue for object representation using histogram. Multi-feature histogram, which is the combination of the gray level intensity information and edge information, generates an object representation which is more robust in thermal vision. The objective of this research is to develop a robust human tracking system which can autonomously detect, identify and track a person in a complex thermal IR scene. In this thesis the tracking procedure has been adapted from the well-known and efficient mean shift tracking algorithm and has been modified to enable fusion of multiple features to increase the robustness of the tracking procedure in thermal vision. In order to identify the object of interest before tracking, rapid human detection in thermal IR scene is achieved using Adaboost classification algorithm. Furthermore, a computationally efficient body pose recognition method is developed which uses Hu-invariant moments for matching object shapes. An experimental setup consisting of a Forward Looking Infrared (FLIR) camera, mounted on a Pioneer P3-DX mobile robot platform was used to test the proposed human tracking system in both indoor and uncontrolled outdoor environments. The performance evaluation of the proposed tracking system on the OTCBVS benchmark dataset shows improvement in tracking performance in comparison to the traditional mean-shift tracking algorithm. Moreover, experimental results in different indoor and outdoor tracking scenarios involving different appearances of people show tracking is robust under cluttered background, varying illumination and partial occlusion of target object.