Narrow Results

This newsletter introduces an automatic real-time data collection method for ice and snow sports activities based on deep mastering. The goal is to cope with the limitations of traditional motion analysis technologies, which includes the inefficiency and excessive prices associated with guide records series. With the developing recognition of ice and snow sports and the increasing competitiveness in these sports, unique evaluation and immediate feedback on athletes’ performance have come to be quintessential. In this look at, we endorse an innovative deep gaining knowledge of framework designed to capture, examine, and offer real-time feedback on athletes’ key performance signs. Our method leverages superior convolutional neural networks (CNNs) to system and examine video records from ice and snow sports activities, which often contain particularly dynamic and complex backgrounds. This machine successfully analyzes athletes’ motion timing facts to offer deeper insights into their overall performance. Moreover, we discover techniques for multi-supply statistics fusion, including combining picture statistics with sensor statistics, to beautify the accuracy and robustness of records collection. Our experimental results demonstrate that our gadget performs noticeably well throughout various ice and snow sports activities occasions. It provides accurate motion monitoring and performance evaluation whilst proving to be more green and fee-powerful as compared to standard methods. This research no longer only advances the utility of deep learning inside the discipline of sports technology however additionally presents a unique technical method for ice and snow sports education and competition. We anticipate that our methodology will significantly enhance athletes’ training outcomes and competitive performance, thereby contributing to the advancement of sports technology in the realm of ice and snow sports.

Knee osteoarthritis (OA) often leads to altered balance control, joint motion and loading patterns during gait. Acupuncture has been proven to be effective in pain relief but its effects on inter-joint load-sharing for body balance have not been reported. The current study bridged the gap by quantifying the immediate effects of acupuncture on the inter-joint and inter-limb load-sharing in patients with knee OA during level walking in terms of the total support moment (Ms) and the contribution of individual joints to the total support moment. Gait analysis was performed on fifteen healthy controls and on fifteen patients with mild to moderate bilateral medial knee OA. After acupuncture treatment the patients with bilateral knee OA walked with significantly increased speed ($p=0.031$), and were able to resist greater and closer-to-normal knee flexion moments ($p=0.038$). They also changed the inter-joint sharing of the support moments with increased knee ($p=0.031$) contributions, but reduced hip contributions ($p=0.022$) to the sagittal Ms during single-limb support. They showed an asymmetric inter-limb load-sharing similar to the normal controls, with increased sharing of the time integral of both the sagittal and frontal whole body support moment by the leading limb during double-limb support ($p=0.024$). The altered intra- and inter-limb contributions to the demand of supporting the body during gait suggests that acupuncture treatment led to immediate changes in the control strategy toward a normal pattern. However, the effects of these changes on the progression of the disease in the long term would require further investigation.

Symmetry of an object on a plane and in a space is an important geometric feature for biology, chemistry, and the understanding of human perception of figures. We propose a randomized method for the detection of symmetry in planar polygons and polyhedrons without assuming the predetermination of the centroids of the objects. Using a voting process, which is the main concept of the Hough transform in image processing, we transform the geometric computation for symmetry detection which is usually based on graph theory and combinatorial optimization, to the peak detection problem in a voting space in the context of the Hough transform. Our algorithm detects the centroid after detecting symmetry of an object for both planar and spatial objects.

In this paper, we present an algorithm for the hierarchical recognition of an environment using independent components of optical flow fields for the visual navigation of a mobile robot. For the computation of optical flow, the pyramid transform of an image sequence is used for the analysis of global and local motion. Our algorithm detects the planar region and obstacles in the image from optical flow fields at each layer in the pyramid. Therefore, our algorithm allows us to achieve both global perception and local perception for robot vision. We show experimental results for both test image sequences and real image sequences captured by a mobile robot. Furthermore, we show some aspects of this work from the viewpoint of information theory.

Motion provides extra information that can aid in the recognition of objects. One of the most commonly seen objects is, perhaps, the human body. Yet little attention has been paid to the analysis of human motion. One of the key steps required for a successful motion analysis system is the ability to track moving objects. In this paper, we describe a new system called Log-Tracker, which was recently developed for tracking the motion of the different parts of the human body. Occlusion of body parts is termed a forking condition. Two classes of forks as well as the attributes required to classify them are described. Experimental results from two gymnastics sequences indicate that the system is able to track the body parts even when they are occluded for a short period of time. Occlusions that extend for a long period of time still pose problems to Log-Tracker.

In this article, we describe some of the algorithms for depth and motion analysis which have been developed within ESPRIT project 940. Specifically we discuss edge detection, token tracking in sequences of images, and trinocular stereo. These processes have been implemented in hardware to form the core of the Depth and Motion Analysis (DMA) machine which has been developed to provide sophisticated real time vision capabilities for a large variety of robotics tasks.

The paper describes a method for detecting 2D straight segments and their correspondences in successive frames of an image sequence by means of a Hough-based matching approach. The main advantage of this method is the possibility of extracting and matching 2D straight segments directly in the feature space, without the need for complex matching operations and time-consuming inverse transformations. An additional advantage is that only four attributes of 2D straight segments are required to perform an efficient matching process: position, orientation, length, and midpoint. Tests were performed on both synthetic and real images containing complex man-made objects moving in a scene. A comparison with a well-known 2D line matching algorithm is also made.

Objective: To assess the accuracy of a skin-mounted device to measure lumbar spinal rotations in rowers.

Methods: Nine subjects were each imaged in 12 extremes of lateral flexion and rotation in an open MR scanner, and the movement between external MR markers fixed to the skin, and the underlying vertebral bodies was quantified.

Results: The average error in measuring the changes in the lateral flexion and rotation between the MR markers and the underlying vertebral bodies was ±5.0° (95% CI 3.5–6.4°, S.E. 0.7°) for lateral flexion and 4.4° (95% CI 3.2–5.6°, S.E 0.6°) for lateral rotation (nine subjects).

Conclusions: This study has demonstrated the accuracy of using an externally mounted device to measure out-of-plane rotations for the extremes of motion achieved during a rowing activity, allowing this system to be used to quantify deviations in rowing technique. The method could be applied to other repetitive activities of the spine.

The successful adaptation and optimization of existing designs play an important role in the design of high speed machinery. Adaptation and extension of existing design principles to achieve some new functional or performance requirement is imperative for the development of the next generation of machines. In addition to this, continual improvement and optimization of existing designs are necessary to sustain a product's commercial success. Where complex mechanical systems are considered such design activities usually require the application of advanced simulation packages or intelligent CAD systems. The widespread utilization of such techniques, for all but the largest enterprises, is severely frustrated by the considerable level of resources required in order to retain a single, proficient user. These resources include extensive initial training as well as ongoing courses, the purchase of commercial licenses and more often than not, the considerable time and effort necessary for the creation and development of complex computer based models. As a consequence, many SMEs demand reliable, efficient and economical methods to generate simulation models for analysis and investigation. One method for delivering this is to provide an approach for the automatic creation of simulation models from data describing an existing design. Such an approach can reduce costs and resources as well as expedite the model building process. This paper presents a method for automatically generating a representative CAD model of a mechanical system. It supports the transformation of digital images into kinematic models for simulation and analysis. The method integrates motion analysis techniques, a parametric description of mechanical systems and a constraint-based CAD system.

Continuous monitoring and automatic detection of crowd activities is extremely helpful for management at public places to avoid any possible disaster. Analysis of crowded scene is a critical task as it typically involves poor resolution of objects, occlusions and complex dynamics. In this paper, we propose a novel, systematic and generalized method based on global motion analysis of people to detect Congestion situation in crowded scenes at entry/exit corridors. Our approach is tested on video footages acquired from surveillance cameras installed at exit corridors of public places. The results show the expediency of our approach.

This study investigates upper limb movement and electromyography (EMG) signals during snatch under various loading conditions and discusses results from six lifting phases. Qualisys motion analysis and Noraxon EMG systems were used to record upper limb movement and muscle activity. When lifting heavy weights, the maximum shoulder flexion angle exceeded 180° in the rise phase and thus, was higher than when lifting lower weight categories. The deltoid and biceps muscles exhibited higher activity during this phase when lifting heavy weights. It can be inferred that the deltoid muscle is activated in this phase in order to maintain the shoulder in an abducted position, and to maintain hyperflexion of the biceps. Muscle activity of the deltoid and biceps in the second pull phase also increased significantly during heavy weight lifting. We infer that the effective use of these two muscles in the second pull phase would produce higher peak barbell vertical velocity, increasing the amount of weight can be lifted. Muscle activity for the latissimus dorsi during first pull showed a statistically significant increase when lifting heavy weights. This ability by the latissimus dorsi to generate higher velocities early in the concentric phase (downswing) possibly contributed to the improved final performance during heavy weight lifting.

This research was carried out to establish the relationship between human anthropometric data and corresponding gait variables. A group comprising 35 participants (18 male and 17 female) was selected for the current study. The study consisted of trials in which each participant was asked to walk the length of the instrumented walkway (Kistler’s force platform inset) at a self-selected speed. Using a four-camera motion analysis system, the kinematic and kinetic parameters of each trial were calculated. The peak values obtained from the data curves were used to generate the necessary regression fits. In order to establish the correlation between the anthropometric data of human and the gait parameters, the univariate, multivariate and stepwise fits were generated. Further, the statistical methods were employed to evaluate the $R^2$, $p$ and $F$-values for each fit. The current multivariate study indicates an increasing trend in $R^2$ values and decreasing trend for $p$-values when compared with the univariate fits and the results follow the expected line.

Background: Although there are a few studies on kinetic and kinematic parameters of scoliotic subjects, it is still controversial whether gait performance of scoliotic subjects differs from that of normal subjects or not. Moreover, there is lack of information regarding joint contact force of scoliotic on convex and concave sides. Therefore, this study examined these issues.

Method: Two groups of children (healthy and children with scoliosis, each group consisting of 5 subjects) participated in this study. The force applied on leg and motions of body parts were evaluated using a Kistler force plate and motion analysis system, respectively. Joint contact forces, muscles length were evaluated in both groups and on both sides (in scoliotic subjects) with OpenSim software. The difference in the parameters between healthy children and scoliotic subjects, and also concave and convex sides, was determined using the independent $t$ test. $P$ value was set at $\le 0.05$.

Results: The results of this study showed that there was no difference between the forces applied on the leg, range of motions of the joints (hip, knee, ankle, pelvic and trunk), muscles length and joint contact force, neither between normal and children with scoliosis, nor between concave and convex sides $(p>0.05)$.

Conclusion: The findings revealed that scoliosis deformity with curve less than $40^{\circ }$ does not have any significant effect on joint contact force, kinetic and kinematic parameters.

This paper considers neurological, formational and functional similarities between gestures and signed verb predicates. From analysis of verb sign movement, we offer suggestions for analyzing gestural movement (motion capture, kinematic analysis, trajectory internal structure). From analysis of verb sign distinctions, we offer suggestions for analyzing co-speech gesture functions.

This project aims to develop an extension to automated gait analysis that makes gait analysis available on smart devices. The alternative may serve as a baseline for future implementations that are cheaper, user-friendly and accessible to an ordinary smartphone or web browser. Accessibility of gait analysis on an application encourages people to check their walking patterns more regularly, and if the issue is very severe, they can take the next step of contacting a specialist. By collaborating with the Podiatry Department of the University of Malta and the Chinese Academy of Sciences Institute of Automation (CASIA), a considerable amount of gait data was acquired. The data consists of videos of people walking regularly or irregularly. But videos are not enough for the development of our system. The videos were inputted into a pose estimator whose goal was to outline the skeleton of the person throughout the video. Additionally, the pose estimator was modified to record the coordinates of the main joints concerning a gait cycle (hip, knee and ankle). These coordinates were then plotted as a scatter plot where the gait cycle is generated. With the coordinates extracted, kinematics were also extracted to create another model which detects different features for gait analysis. After the gait cycle of each video was extracted, the next step was to classify whether that gait cycle was either regular or irregular. This goal is achieved by passing the extracted data through the VGG16 architecture. The application was tested out on people which have either bad, good or slightly bad gaits to investigate the rigidity of the system. After a series of experiments, it can be concluded that the system performs with 94% accuracy just by using a mobile phone.

It has been shown that intraoperative stress can have a negative effect on surgeon surgical skills during laparoscopic procedures. For novice surgeons, stressful conditions can lead to significantly higher velocity, acceleration, and jerk of the surgical instrument tips, resulting in faster but less smooth movements. However, it is still not clear which of these kinematic features (velocity, acceleration, or jerk) is the best marker for identifying the normal and stressed conditions. Therefore, in order to find the most significant kinematic feature that is affected by intraoperative stress, we implemented a spatial attention-based Long Short-Term Memory (LSTM) classifier. In a prior IRB approved experiment, we collected data from medical students performing an extended peg transfer task who were randomized into a control group and a group performing the task under external psychological stresses. In our prior work, we obtained “representative” normal or stressed movements from this dataset using kinematic data as the input. In this study, a spatial attention mechanism is used to describe the contribution of each kinematic feature to the classification of normal/stressed movements. We tested our classifier under Leave-One-User-Out (LOUO) cross-validation, and the classifier reached an overall accuracy of 77.11% for classifying “representative” normal and stressed movements using kinematic features as the input. More importantly, we also studied the spatial attention extracted from the proposed classifier. Velocity and acceleration on both sides had significantly higher attention for classifying a normal movement ($p\le 0.0001$); Velocity ($p\le 0.015$) and jerk ($p\le 0.001$) on nondominant hand had significant higher attention for classifying a stressed movement, and it is worthy noting that the attention of jerk on nondominant hand side had the largest increment when moving from describing normal movements to stressed movements ($p=0.0000$). In general, we found that the jerk on nondominant hand side can be used for characterizing the stressed movements for novice surgeons more effectively.

To determine the dynamic axial rotation movement of human upper extremity, we developed a combined skin- and scapula-based marker system in which the scapula motion was detected with a marker set attached to an intracortical pin on the acromion. The subject performed axial rotation with the arm in 5 different positions. The total rotation range varied greatly if the arm was in different position. The pattern of axial rotation of each segment also changed. The upper extremity rotates total 376° at 90° abduction and only 257° at full abduction. The forearm usually contributed about 140°, so the differences mainly came from the scapulothoracic and the glenohumeral joints. The scapulothoracic joint axially rotated 66° with the arm in backward extension while only rotated 19° with the arm in 90° forward flexion. The glenohumeral joint rotated maximally, 143° with the arm at 90° abduction, while minimally, 69° with the arm at maximal abduction.

Skin marker-based stereophotogrammetry has been widely used in the in vivo, noninvasive measurement of three-dimensional (3D) joint kinematics in many clinical applications. However, the measured poses of body segments are subject to errors called soft tissue artifacts (STA). No study has reported the unrestricted STA of markers on the thigh and shank in normal subjects during functional activities. The purpose of this study was to assess the 3D movement of skin markers relative to the underlying bones in normal subjects during functional activities using a noninvasive method based on the integration of 3D fluoroscopy and stereophotogrammetry. Generally, thigh markers had greater STA than shank ones and the STA of the markers were in nonlinear relationships with knee flexion angles. The STA of a marker also appeared to vary among subjects and were affected by activities. This suggests that correction of STA in human motion analysis may have to consider the multijoint nature of functional activities such as using a global compensation approach with individual anthropometric data. The results of the current study may be helpful for establishing guidelines of marker location selection and for developing STA compensation methods in human motion analysis.

Tai-Chi Chuan (TCC) is thought to be a low-impact and effective exercise to improve balance capability in the elderly. However, the effects of TCC exercise on balance improvement remain controversial. The purpose of the current study was to investigate the effects of long-term Yang-Style TCC training on balance variables such as stable standing time and center of pressure (COP) movement patterns. Fifteen long-term Yang-Style TCC practitioners and fifteen age-matched adults performed different static balance tests each for 30 seconds. For each test, the time-varying COP positions were measured by two forceplates. The sway area of the COP was described by an equivalent ellipse, the two principal axes of which were obtained by using principal component analysis. The results showed that elderly subjects with long-term Yang-Style TCC training were able to maintain stable standing longer than those without TCC training, with reduced COP sway area during challenging tasks such as single-leg stance and tandem stance. It is suggested that long-term TCC exercise is a good choice in a training program aimed at reducing the risk of falling in the elderly.

Few studies have concurrently investigated the accuracy and repeatability of an optical and electromagnetic (EM) system during dynamic motion. The purposes of this study were to: (1) assess the accuracy of both an EM and optical system when compared to a gold standard and (2) to compare the intra- and inter-day repeatability during 3D kinematic motion of both systems. The gold standard used for accuracy assessment was a robot programmed to manipulate a carbon fiber beam through pre-defined motions within the capture volume of both systems at 30, 45 and 60°/s. A total of 12 healthy young adults were tested for intra- and inter-day repeatability of hip, knee and ankle joint angles during a sit-to-stand movement. Marker trajectories were captured using an 8-camera Motion Analysis system and a Polhemus Liberty system. Optical markers for both portions of the study were precisely marked to allow for digitization by the EM system, with collections taken at 120 Hz. Accuracy and repeatability were assessed using the RMS error and coefficient of multiple correlations (CMC), respectively. The optical system demonstrated a 1–2.5° lower RMS error in tracking the robot movements in the transverse and sagittal planes when compared to the EM system. However, it was possible that metal interference affected the accuracy of the EM system. High intra-day and inter-day repeatability was demonstrated by both systems during the sit-to-stand task. The optical system did demonstrate slightly higher CMC values for between day trials, though skin motion artifact might have affected the EM system to a greater extent. Overall, both systems demonstrated an adequate ability to track dynamic motion.