FGFF Descriptor and Modified Hu Moment-Based Hand Gesture Recognition

Gesture recognition has been studied for decades and still remains an open problem. One important reason is that the features representing those gestures are not sufficient, which may lead to poor performance and weak robustness. Therefore, this work aims at a comprehensive and discriminative feature for hand gesture recognition. Here, a distinctive Fingertip Gradient orientation with Finger Fourier (FGFF) descriptor and modified Hu moments are suggested on the platform of a Kinect sensor. Firstly, two algorithms are designed to extract the fingertip-emphasized features, including palm center, fingertips, and their gradient orientations, followed by the finger-emphasized Fourier descriptor to construct the FGFF descriptors. Then, the modified Hu moment invariants with much lower exponents are discussed to encode contour-emphasized structure in the hand region. Finally, a weighted AdaBoost classifier is built based on finger-earth mover’s distance and SVM models to realize the hand gesture recognition. Extensive experiments on a ten-gesture dataset were carried out and compared the proposed algorithm with three benchmark methods to validate its performance. Encouraging results were obtained considering recognition accuracy and efficiency.

Multi-centre, multi-vendor 7 Tesla fMRI reproducibility of hand digit representation in the human somatosensory cortex

Whilst considerable progress has been made in using ultra-high field fMRI to study brain function at fine spatial resolution, methods are generally optimized at a single site and do not translate to studies where multiple sites are required for sufficient subject recruitment. With a recent increase in installations of human 7 T systems, there is now the opportunity to establish a framework for multi-site 7 T fMRI studies. However, an understanding of the inter-site variability of fMRI measurements is required for datasets to be combined across sites. To address this, we employ a hand digit localization task and compare across-site and within-site reproducibility of 7 T fMRI to a hand digit localization task which requires fine spatial resolution to resolve individual digit representations. As part of the UK7T Network “Travelling Heads” study, 10 participants repeated the same hand digit localization task at five sites with whole-body 7T MRI systems to provide a measure of inter-site variability. A subset of the participants (2 per site) performed repeated sessions at each site for measurement of intra-site reproducibility. Dice’s overlap coefficient was used to assess reproducibility, with hand region inter-site Dice = 0.70±0.04 significantly lower than intrasite Dice = 0.76±0.06, with similar trends for the individual digit maps. Although slightly lower than intra-site reproducibility, the inter-site reproducibility results are consistent with previous single site reproducibility measurements, providing evidence that multi-site 7 T fMRI studies are feasible. These results can be used to inform sample size calculations for future multi-site somatomotor mapping studies.

More comprehensive proprioceptive stimulation of the hand amplifies its cortical processing

ABSTRACTCorticokinematic coherence (CKC) quantifies the phase coupling between limb kinematics and cortical neurophysiological signals reflecting proprioceptive feedback to the primary sensorimotor (SM1) cortex. We studied CKC to proprioceptive stimulation (i.e. movement-actuator-evoked movements) of right-hand digits (index, middle, ring and little) performed simultaneously or separately. CKC was computed between magnetoencephalography (MEG) and finger acceleration signals. The strongest CKC was obtained by stimulating the fingers simultaneously at fixed 3-Hz frequency, and can, therefore, be recommended as design for fast functional localization of the hand area in the primary sensorimotor (SM1) cortex using MEG. The peaks of CKC sources were concentrated in the hand region of the SM1 cortex, but did not follow consistent somatotopic order. This result suggests that spatial specificity of MEG is not sufficient to separate proprioceptive finger representations of the same hand adequately or that their representations are overlapping.

Understanding the hand-gestures using Convolutional Neural Networks and Generative Adversial Networks

<p>In this paper, it is introduced a hand gesture recognition system to recognize the characters in the real time. The system consists of three modules: real time hand tracking, training gesture and gesture recognition using Convolutional Neural Networks. Camshift algorithm and hand blobs analysis for hand tracking are being used to obtain motion descriptors and hand region. It is fairy robust to background cluster and uses skin color for hand gesture tracking and recognition. Furthermore, the techniques have been proposed to improve the performance of the recognition and the accuracy using the approaches like selection of the training images and the adaptive threshold gesture to remove non-gesture pattern that helps to qualify an input pattern as a gesture. In the experiments, it has been tested to the vocabulary of 36 gestures including the alphabets and digits, and results effectiveness of the approach.</p>

Understanding the hand-gestures using Convolutional Neural Networks and Generative Adversial Networks

<p>In this paper, it is introduced a hand gesture recognition system to recognize the characters in the real time. The system consists of three modules: real time hand tracking, training gesture and gesture recognition using Convolutional Neural Networks. Camshift algorithm and hand blobs analysis for hand tracking are being used to obtain motion descriptors and hand region. It is fairy robust to background cluster and uses skin color for hand gesture tracking and recognition. Furthermore, the techniques have been proposed to improve the performance of the recognition and the accuracy using the approaches like selection of the training images and the adaptive threshold gesture to remove non-gesture pattern that helps to qualify an input pattern as a gesture. In the experiments, it has been tested to the vocabulary of 36 gestures including the alphabets and digits, and results effectiveness of the approach.</p>

T2 MRI Mapping of Knee Cartilage in Professional Ballet Dancers

AIMS: Professional ballet dancers are at risk for degenerative knee cartilage changes. In the current study, we evaluated the knee cartilage with T2 mapping methods in professional ballet dancers and healthy controls and investigated possible differences of T2 values between these groups. METHODS: We included healthy dancers with 5−20 years of professional ballet dancing experience and sex-matched healthy controls without knee pathology. T2 values of the knee cartilage were measured from axial, coronal, and sagittal images. The values were measured by free hand region of interest (ROI). RESULTS: The study population consisted of 44 people (22 dancers, 22 controls), with mean age of 25.57 ± 5.53 yrs. We found no significant relationship between sex and T2 values. We detected a significant positive correlation between age and T2 values for patellofemoral joint cartilage. T2 values of patellofemoral and tibiofemoral joints of dancers were significantly higher. Mean T2 values of patellofemoral joint were around 32 in all planes in dancers and around 12 in controls (p<0.05). Mean tibiofemoral joint values in dancers were around 39 and around 32 in controls (p<0.05). CONCLUSIONS: T2 values of knee cartilage were higher in professional ballet dancers. T2 mapping method can reveal knee cartilage changes successfully in professional ballet dancers. All anatomical planes (axial, coronal, and sagittal) can be used in order to obtain T2 values.

Real-time virtual mouse system using RGB-D images and fingertip detection

A real-time fingertip-gesture-based interface is still challenging for human–computer interactions, due to sensor noise, changing light levels, and the complexity of tracking a fingertip across a variety of subjects. Using fingertip tracking as a virtual mouse is a popular method of interacting with computers without a mouse device. In this work, we propose a novel virtual-mouse method using RGB-D images and fingertip detection. The hand region of interest and the center of the palm are first extracted using in-depth skeleton-joint information images from a Microsoft Kinect Sensor version 2, and then converted into a binary image. Then, the contours of the hands are extracted and described by a border-tracing algorithm. The K-cosine algorithm is used to detect the fingertip location, based on the hand-contour coordinates. Finally, the fingertip location is mapped to RGB images to control the mouse cursor based on a virtual screen. The system tracks fingertips in real-time at 30 FPS on a desktop computer using a single CPU and Kinect V2. The experimental results showed a high accuracy level; the system can work well in real-world environments with a single CPU. This fingertip-gesture-based interface allows humans to easily interact with computers by hand.

Prevalence of self-reported musculoskeletal disorders of the hand and associated conducted therapy approaches among dentists and dental assistants in Germany

Background Dental professionals are subjected to higher risks for musculoskeletal disorders (MSDs) than other professional groups, especially the hand region. This study aims to investigate the prevalence of hand complaints among dentists (Ds) and dental assistants (DAs) and examines applied therapies. Methods For this purpose, an online questionnaire analysed 389 Ds (240female/149male) and 406 DAs (401female/5male) working in Germany. The self-reported data of the two occupational groups were compared with regard to the topics examined. The questionnaire was based on the Nordic Questionnaire (self-reported lifetime, 12-month and 7-day MSDs prevalence of the hand, the conducted therapy and its success), additional occupational and sociodemographic questions as well as questions about specific medical conditions. Results 30.8% of Ds affirmed MSDs in the hand at any time in their lives, 20.3% in the last twelve months and 9.5% in the last seven days. Among DAs, 42.6% reported a prevalence of MSDs in the hand at any time in their lives, 31.8% in the last 12 months and 15.3% in the last seven days. 37.5% of the Ds and 28.3% of the DAs stated that they had certain treatments. For both, Ds and DAs, physiotherapy was the most frequently chosen form of therapy. 89.7% of Ds and 63.3% of DAs who received therapy reported an improvement of MSDs. Conclusion Although the prevalence of MSDs on the hand is higher among DAs than among Ds, the use of therapeutic options and the success of therapy is lower for DAs compared to Ds.

Hand gesture recognition via image processing techniques and deep CNN

In the present study, we sought to enable instant tracking of the hand region as a region of interest (ROI) within the image range of a webcam, while also identifying specific hand gestures to facilitate the control of home appliances in smart homes or issuing of commands to human-computer interaction fields. To accomplish this objective, we first applied skin color detection and noise processing to remove unnecessary background information from the captured image, before applying background subtraction for detection of the ROI. Then, to prevent background objects or noise from influencing the ROI, we utilized the kernelized correlation filters (KCF) algorithm to implement tracking of the detected ROI. Next, the size of the ROI image was resized to 100×120 and input into a deep convolutional neural network (CNN) to enable the identification of various hand gestures. In the present study, two deep CNN architectures modified from the AlexNet CNN and VGGNet CNN, respectively, were developed by substantially reducing the number of network parameters used and appropriately adjusting internal network configuration settings. Then, the tracking and recognition process described above was continuously repeated to achieve immediate effect, with the execution of the system continuing until the hand is removed from the camera range. The results indicated excellent performance by both of the proposed deep CNN architectures. In particular, the modified version of the VGGNet CNN achieved better performance with a recognition rate of 99.90% for the utilized training data set and a recognition rate of 95.61% for the utilized test data set, which indicate the good feasibility of the system for practical applications.

Dystroglycan regulates proper expression, submembranous localization and subsequent phosphorylation of Dp71 through physical interaction

Dystrophin–dystroglycan complex (DGC) plays important roles for structural integrity and cell signaling, and its defects cause progressive muscular degeneration and intellectual disability. Dystrophin short product, Dp71, is abundantly expressed in multiple tissues other than muscle and is suspected of contributing to cognitive functions; however, its molecular characteristics and relation to dystroglycan (DG) remain unknown. Here, we report that DG physically interacts with Dp71 in cultured cells. Intriguingly, DG expression positively and DG knockdown negatively affected the steady-state expression, submembranous localization and subsequent phosphorylation of Dp71. Mechanistically, two EF-hand regions along with a ZZ motif of Dp71 mediate its association with the transmembrane proximal region, amino acid residues 788–806, of DG cytoplasmic domain. Most importantly, the pathogenic point mutations of Dp71, C272Y in the ZZ motif or L170del in the second EF-hand region, impaired its binding to DG, submembranous localization and phosphorylation of Dp71, indicating the relevance of DG-dependent Dp71 regulatory mechanism to pathophysiological conditions. Since Dp140, another dystrophin product, was also regulated by DG in the same manner as Dp71, our results uncovered a tight molecular relation between DG and dystrophin, which has broad implications for understanding the DGC-related cellular physiology and pathophysiology.