Abstract: To provide a moving object detection apparatus which accurately performs region extraction, regardless of the pose or size of a moving object. The moving object detection apparatus includes: an image receiving unit receiving the video sequence; a motion analysis unit calculating movement trajectories based on motions of the image; a segmentation unit performing segmentation so as to divide the movement trajectories into subsets, and setting a part of the movement trajectories as common points shared by the subsets; a distance calculation unit calculating a distance representing a similarity between a pair of movement trajectories, for each of the subsets; a geodesic distance calculation unit transforming the calculated distance into a geodesic distance; an approximate geodesic distance calculation unit calculating an approximate geodesic distance bridging over the subsets, by integrating geodesic distances including the common points; and a region extraction unit performing clustering on the calculated approximate geodesic distance.

Abstract: In order to implement an unobstructed assessment of three-dimensional (3-D) gait, we developed a mobile force plate and 3-D motion analysis system (M3D) to measure triaxial ground reaction forces (GRF) and 3-D orientations of feet. Calibration and test experiments were conducted to characterize the sensor developed. To test the accuracy of the new measurement system, validation experiments by using the reference measurements of a commercially available measurement system were performed in a gait laboratory, where a stationary force plate, a motion capture system based on high-speed cameras and a motion track system of XSENS were adopted to analyze human movements. Experimental results supported the proposal that the developed system can be used to measure triaxial GRF and orientations with an acceptable precision during successive walking gait.

Abstract: Motion of the carotid artery wall is important for the quantification of arterial elasticity and contractility and can be estimated with a number of techniques. In this paper, a framework for quantitative evaluation of motion analysis techniques from B-mode ultrasound images is introduced. Six synthetic sequences were produced using 1) a real image corrupted by Gaussian and speckle noise of 25 and 15 dB, and 2) the ultrasound simulation package Field II. In both cases, a mathematical model was used, which simulated the motion of the arterial wall layers and the surrounding tissue, in the radial and longitudinal directions. The performance of four techniques, namely optical flow (OFHS), weighted least-squares optical flow (OFLK(WLS)), block matching (BM), and affine block motion model (ABMM), was investigated in the context of this framework. The average warping indices were lowest for OFLK(WLS) (1.75 pixels), slightly higher for ABMM (2.01 pixels), and highest for BM (6.57 pixels) and OFHS (11.57 pixels). Due to its superior performance, OFLK(WLS) was used to quantify motion of selected regions of the arterial wall in real ultrasound image sequences of the carotid artery. Preliminary results indicate that OFLK(WLS) is promising, because it efficiently quantified radial, longitudinal, and shear strains in healthy adults and diseased subjects.

Abstract: Tracking moving objects is one of the most important but problematic features of motion analysis and understanding. The Kalman filter (KF) has commonly been used for estimation and prediction of the target position in succeeding frames. In this paper, we propose a novel and efficient method of tracking, which performs well even when the target takes a sudden turn during its motion. The proposed method arbitrates between KF and Optical flow (OF) to improve the tracking performance. Our system utilizes a laser to measure the distance to the nearest obstacle and an infrared camera to find the target. The relative data is then fused with the Arbitrate OFKF filter to perform real-time tracking. Experimental results show our suggested approach is very effective and reliable for estimating and tracking moving objects.

Abstract: Rushes footages are considered as cheap gold mine with the potential for reuse in broadcasting and filmmaking industries. However, mining “gold” from unedited videos such as rushes is challenging as the reusable segments are buried in a large set of redundant information. In this paper, we propose a unified framework for stock footage classification and summarization to support video editors in navigating and organizing rushes videos. Our approach is composed of two steps. First, we employ motion features to filter the undesired camera motion and locate the stock footage. A hierarchical hidden Markov model (HHMM) is proposed to model the motion feature distribution and classify video segments into different categories to decide their potential for reuse. Second, we generate a short video summary to facilitate quick browsing of the stock footages by including the objects and events that are important for storytelling. For objects, we detect the presence of persons and moving objects. For events, we extract a set of features to detect and describe visual (motion activities and scene changes) and audio events (speech clips). A representability measure is then proposed to select the most representative video clips for video summarization. Our experiments show that the proposed HHMM significantly outperforms other methods based on SVM, FSM, and HMM. The automatically generated rushes summaries are also demonstrated to be easy-to-understand, containing little redundancy, and capable of including ground-truth objects and events with shorter durations and relatively pleasant rhythm based on the TRECVID 2007, 2008, and our subjective evaluations.

Abstract: Optical flow approaches calculate vector fields which determine the apparent velocities of objects in time-varying image sequences. They have been analyzed extensively in computer science using both natural and synthetic video sequences. In life sciences, there is an increasing need to extract kinetic information from temporal image sequences which reveals the interplay between form and function of microscopic biological structures. In this work, we test different variational optical flow techniques to quantify the displacements of biological objects in 2D fluorescent image sequences. The accuracy of the vector fields is tested for defined displacements of fluorescent point sources in synthetic image series which mimic protein traffic in neuronal dendrites, and for GABABR1 receptor subunits in dendrites of hippocampal neurons. Our results reveal that optical flow fields predict the movement of fluorescent point sources within an error of 3% for a maximum displacement of 160 nm. Displacement of agglomerated GABABR1 receptor subunits can be predicted correctly for maximum displacements of 640 nm. Based on these results, we introduce a criteria to derive the optimum parameter combinations for the calculation of the optical flow fields in experimental images. From these results, temporal sampling frequencies for image acquisition can be derived to guarantee correct motion estimation for biological objects.

Abstract: In this paper, we put forward a novel region matching-based motion estimation scheme to detect objects with accurate boundaries from videos captured by moving camera. Here, a fuzzy edge incorporated Markov random field (MRF) model is considered for spatial segmentation. The algorithm is able to identify even the blurred boundaries of objects in a scene. Expectation Maximization algorithm is used to estimate the MRF model parameters. To reduce the complexity of searching, a new scheme is proposed to get a rough idea of maximum possible shift of objects from one frame to another by finding the amount of shift in positions of the centroid. We propose a χ2-test-based local histogram matching scheme for detecting moving objects from complex scenes from low illumination environment and objects that change size from one frame to another. The proposed scheme is successfully applied for detecting moving objects from video sequences captured in both real-life and controlled environments. It is also noticed that the proposed scheme provides better results with less object background misclassification as compared to existing techniques.

Abstract: This work addresses the issue of fire and smoke detection in a scene within a video surveillance framework. Detection of fire and smoke pixels is at first achieved by means of a motion detection algorithm. In addition, separation of smoke and fire pixels using colour information (within appropriate spaces, specifically chosen in order to enhance specific chromatic features) is performed. In parallel, a pixel selection based on the dynamics of the area is carried out in order to reduce false detection. The output of the three parallel algorithms are eventually fused by means of a MLP.

Abstract: This paper introduces a method to track the real-time velocity and capture the dynamic behavior of a person using inertial motion capture system. The velocity of a person while moving is curtail, but existing capture systems do not discuss on this. In this work, a method for monitoring the velocity of a key reference root point on the subject is first proposed. Based on the contact detection, the velocity calculated from lower limb kinematics and the integration of accelerations are fused through Kalman Filters to achieve accurate and drift free estimation results. Based on this velocity estimation and the limb kinematics, the spatial location of the subject can then be updated in real-time. Experiments are conducted to show the capability of the proposed method. The capability and accuracy of the proposed method in both the velocity tracking and localization are verified using the commercial optical motion capture system, Motion Analysis. The velocity and spatial behavior of the person during motions like walking, jumping and running are accurately captured with this self-contained system. This system does not depend on external devices, has little constraint to the human movements and can accurately monitor the moving speed of the person, thus it can show advantage in everyday practice and sport training in large area or home environments.

Abstract: Concurrent advancements in mechanical design and motion planning algorithms allow state-of-the-art humanoid robots to exhibit complex and realistic behavior. In face of this added complexity and the need for humanlike behavior, research has begun to look toward studies in human neuroscience to better organize and guide humanoid robot motion. In this paper, we present one such method of generating anthropomorphic motion by building the “invariants” of human movements and applying them as kinematic tasks. Whole-body motion of 14 healthy participants was recorded during a walking and grasping task. The recorded data were statistically analyzed to extract invariants which best described the observed motion. These invariants were expressed as a set of rules that were used to synthesize the stereotypy in human motion. We propose an algorithm that reproduces the key parameters of motion, taking into account the knowledge from human movement and the limitations of the target anthropomorph. The results are then generalized such that we can generate motion for targets which were not originally recorded. The algorithmic output is applied in a task-based prioritized inverse kinematics solver to generate dynamically stable and realistic anthropomorphic motion. We illustrate our results on the humanoid HRP-2 by making it walk to and grasp objects at various positions. Our approach complements classical optimization or motion-planning-based methods and provides interesting perspectives toward the use of human movements for deducing effective cost functions in optimization techniques or heuristics for planning algorithms.

Abstract: A motion analysis apparatus includes a sensor that is attached to an object and detects a physical value of the object, a holder that holds the object and sets the sensor in a first position, and a motion analyzer that acquires an output from the sensor and analyzes the motion of the object based on the output, the output including first output data from the sensor set in the first position and second output data from the sensor produced after the object is separated from the holder and the sensor is set in at least one known second position.

Abstract: A medical analysis method for estimating a motion vector field of the magnitude and direction of local motion of lung tissue of a subject is described. In one embodiment a first 3D image data set of the lung and a second 3D image data set is obtained. The first and second 3D image data sets correspond to images obtained during inspiration and expiration respectively. A rigid registration is performed to align the 3D image data sets with one another. A deformable registration is performed to match the 3D image data sets with one another. A motion vector field of the magnitude and direction of local motion of lung tissue is estimated based on the deforming step. The motion vector field may be computed prior to treatment to assist with planning a treatment as well as subsequent to a treatment to gauge efficacy of a treatment. Results may be displayed to highlight.

Abstract: Image analysis is generally a process where digital image processing is utilized to process digital images in order to extract significant statistics or information from the images. The analysis process enables to analyze and visualize medical images of numerous modalities. The paper is basically an overview and discussion of the methods and techniques being proposed and developed in regard of brain image analysis. An image is basically analyzed from the perspective of its segmentation, edge detection, registration and morphology or motion analysis. Here in this paper different brain image types i.e., MRI, CT, PET, EEG/MEG are discussed and presented from the point of operations mentioned above. Each method is discussed and analyzed through its applications, advantages, limitations and results.

Abstract: Real-time magnetic resonance imaging is a promising tool for image-guided interventions. For applications such as thermotherapy on moving organs, a precise image-based compensation of motion is required in real time to allow quantitative analysis, retrocontrol of the interventional device, or determination of the therapy endpoint. Reduced field-of-view imaging represents a promising way to improve spatial and / or temporal resolution. However, it introduces new challenges for target motion estimation, since structures near the target may appear transiently due to the respiratory motion and the limited spatial coverage. In this paper, a new image-based motion estimation method is proposed combining a global motion estimation with a novel optical flow approach extending the initial Horn and Schunck (HS & 0.05) of the Dice's similarity criterion computed between the reference and the registered organ positions was achieved.

Abstract: This paper describes an on-board vision sensor system that is developed specifically for small unmanned vehicle applications For small vehicles, vision sensors have many advantages, including size, weight, and power consumption, over other sensors such as radar, sonar, and laser range finder, etc A vision sensor is also uniquely suited for tasks such as target tracking and recognition that require visual information processing However, it is difficult to meet the computing needs of real-time vision processing on a small robot In this paper, we present the development of a field programmable gate array-based vision sensor and use a small ground vehicle to demonstrate that this vision sensor is able to detect and track features on a user-selected target from frame to frame and steer the small autonomous vehicle towards it The sensor system utilizes hardware implementations of the rank transform for filtering, a Harris corner detector for feature detection, and a correlation algorithm for feature matching and tracking With additional capabilities supported in software, the operational system communicates wirelessly with a base station, receiving commands, providing visual feedback to the user and allowing user input such as specifying targets to track Since this vision sensor system uses reconfigurable hardware, other vision algorithms such as stereo vision and motion analysis can be implemented to reconfigure the system for other real-time vision applications

Abstract: This work introduces a novel motion descriptor that enables human activity classification in video-surveillance applications. The method starts by computing a dense optical flow, providing instantaneous velocity information for every pixel. The obtained flow is then characterized by a per-frameorientation histogram, weighted by the norm, with orientations quantized to 32 principal directions. Finally, a set of global characteristics is determined from the temporal series obtained from each histogram bin, forming a descriptor vector. The method was evaluated using a 192-dimensional descriptor with the classical Weizmann action dataset, obtaining an average accuracy of 95%. For more complex surveillance scenarios, the method was assessed with the VISOR dataset, achieving a 96.7% of accuracy in a classification task performed using a Support Vector Machine (SVM) classifier.

Abstract: Motion analysis deals with determining what and how activities are being performed by a subject, through the use of sensors. The process of answering the what question is commonly known as classifi ...

Abstract: Objectives Cadaveric models of the shoulder evaluate discrete motion segments using the glenohumeral joint in isolation over a defined trajectory. The aim of this study was to design, manufacture and validate a robotic system to accurately create three-dimensional movement of the upper body and capture it using high-speed motion cameras. Methods In particular, we intended to use the robotic system to simulate the normal throwing motion in an intact cadaver. The robotic system consists of a lower frame (to move the torso) and an upper frame (to move an arm) using seven actuators. The actuators accurately reproduced planned trajectories. The marker setup used for motion capture was able to determine the six degrees of freedom of all involved joints during the planned motion of the end effector. Results The testing system demonstrated high precision and accuracy based on the expected versus observed displacements of individual axes. The maximum coefficient of variation for displacement of unloaded axes was less than 0.5% for all axes. The expected and observed actual displacements had a high level of correlation with coefficients of determination of 1.0 for all axes. Conclusions Given that this system can accurately simulate and track simple and complex motion, there is a new opportunity to study kinematics of the shoulder under normal and pathological conditions in a cadaveric shoulder model.

Abstract: Hand lay-up is one of methods for FRP molding and it is the most traditional and common molding method that can deal with various shapes according to workmen skill. However, the quality of the composites depends on the techniques of workmen and the judgment with hand lay-up technique is too difficult, because it has not clarified that the difference of that techniques has how influence with manufactured products. In this study, motion analysis was used to compare the difference between hand lay-up techniques by expert and non-expert. Expert’s carrier of hand lay-up was 25 years and non-expert’s carrier was a year. Motion analysis, which is applied to various fields like sports or traditional crafts and so on, can visualize human motion. The Mac 3D System was used as equipment, since it is the most powerful tool for the motion capture and analysis particularly. The sampling rate was 60 Hz. The object of this study was to contribute that technique back to fabrication fields by analyzing and considering what was important factor. Furthermore, this trial is thought that leads to the development of new technology. As the result, the motion of expert’s lower half of body indicated different motion compared with non-expert one. Expert’s centroid moved smoothly and his motion showed constant tendency. On the other hand, non-expert’s motion was awkward in several points and his motion didn’t show the tendency like expert. Furthermore their total processes were almost same; however, non-expert spent more time on finishing touches than expert. Expert spent a third of his total time on finishing touches but non-expert one was a half of his total process. In addition, the quality of manufactured products that they made was different. Hand lay-up technique is discussed based on the motion analysis method, and moreover, it is possible to evolve hand lay-up technique further (We are calling “advanced-Hand lay-up”).Copyright © 2012 by ASME

Abstract: In this paper, a novel vision system is proposed to estimate attention of people from rich visual clues for social robot to perform natural interactions with multiple participants in public environments. The vision detection and recognition modules include multi-person detection and tracking, upper-body pose recognition, face and gaze detection, lip motion analysis for speaking recognition, and facial expression recognition. A computational approach is proposed to generate a quantitative estimation of human attention. The vision system is implemented on a robotic receptionist "EVE" and encouraging results have been obtained.

Abstract: This study quantifies the accuracy of the Microsoft Kinect in two motions. Ten participants were asked to perform reaching and throwing actions which were recorded simultaneously by a Kinect and a motion analysis corporation (MAC) capture system. Elbow and shoulder angles were calculated for both motions. NITE (PrimeSense, USA) and IPI soft tracking algorithms were used. NITE tracking had an average maximum error of 32.4° for the elbow and shoulder in the reach motion and 95.3° in the throwing motion. IPI soft had equivalent maximum error values of 22.3° and 43.0° respectively. While accuracy isn’t high, and suffers in high speed motions, the advantages offered by markerless tracking, low cost and zero calibration make the Kinect potentially valuable for motion analysis in coaching, clinical and educational domains.

Abstract: This paper presents a low-cost motion analysis system consisting of the Microsoft Kinect sensor and a self-developed software for rehabilitation and sports medicine. During therapy or training sessions a person's motion is recorded through video and skeletal data. Subsequently, our software allows the analysis and comparison of all recorded motion data. The combination of motion analysis functionality, a user-friendly setup, and easy handling makes our system well-suited for the development or support of exergames.

Abstract: Motion capture data acquired from high definition cameras creates accurate human motion representation but introduces many redundant frames which pose a problem in data storage and motion retrieval purposes. In this paper, a keyframing approach is proposed to reduce the motion data by extracting keyframes using motion analysis approach in sampling windows. Motion changes in sampling windows for original motion without frame skipping and with frame skipping are computed. The difference in the motion changes is the main aspect in deciding whether the frames in sampling windows are possible candidates for keyframe selection. Simulation results showed that the proposed method is able to achieve an overall good visual quality for different types of motion. It also gives an improvement of up to 52% in terms of mean square error measurement, as compared to the existing keyframe extraction method, which is curve simplification method.

Abstract: Body motion analyzing has always been a useful tool to evaluate patients' joint kinematics. Optical tracking is one of the most accurate dynamic tracking systems, and is commonly used as diagnosis devices for joint motion analysis. However, these systems are usually located in gait analysis laboratory and not readily available in clinic or hospital for day to day diagnostic use. In order to provide an alternative means for joint kinematics assessment, the following method is examined. An inertial sensing node composes of a set of inertial micro-electromechanical system (MEMS) sensors (accelerometers, gyroscopes and magneto- meters) were used to track two adjacent joint segments. This paper presents a preliminary study of motion tracking of the upper and lower extremities of the leg during dynamic activities. Each node communicates wirelessly to the base station with Bluetooth. A quaternion based Extended Kalman Filter (EKF) was implemented to process the data for orientation estimation. Anatomy constraint is applied to the relative orientation of the estimation. The focus of this paper is to introduce the framework of easy to use, high mobility, low cost motion analysis and diagnostic system that can be used in doctor's office.

Abstract: In this paper, a video-surveillance system for the detection of abandoned objects and its owner is proposed. The approach is based on dual background for the detection of static foreground and motion analysis for its owner. The static foreground detection performs in two steps: constructing and updating a pixel-based dual background to obtain the dual foreground, carrying a register for the evidence frame to give an alarm. Motion analysis establishes the correspondence of foreground objects between the successive frames using color histogram, in order to label the objects and record their characteristic and state. When the static foreground detection part find a possible target, the result of the motion analysis can be used to estimate the owner. We demonstrate that this abandoned object detection system solves successfully a high percentage of occlusions.