Gesture recognition based on binocular vision

Self Cite

In the current development and design of sports rehabilitation equipment or biomimetic prostheses, in addition to pay attention to the development and design of the structure, the more core is how to realize the accurate and effective control of the rehabilitation equipment or intelligent prosthesis, and the current research is based on data process and pattern recognition. This paper designs 9 kinds of actions that can react effectively to the function of the hand and extracts the original EMG signals, which are based on the sEMG of the forearm muscles of human hand movement, and uses the 20 order comb filter and wavelet threshold to preprocess the signal, and uses the root mean square, wavelength and nonlinear characteristics sample entropy in time domain as three eigenvalues to construct the input feature vectors of the subsequent action classifier. Finally, the recognition of the hand movements is realized successfully through GRNN and SVM. The recognition rate is 98.64% in SVM classifier and 96.27% in GRNN classifier. Experimental results show that the SVM classifier is better than the GRNN classifier.

Section: Noise Reduction Of Original Surface Emg Signalmentioning

confidence: 99%

Surface EMG data aggregation processing for intelligent prosthetic action recognition

Jiang

et al. 2018

Self Cite

“…After the original image is segmented by the depth segmentation and the elliptical skin model, a binary image of the gesture image with a large number of backgrounds is obtained, but there are burrs on the gesture boundary or holes in the gesture area, which will interfere with subsequent feature extraction and classification operations [30][31][32][33]. Therefore, it is necessary to perform morphological processing and image enhancement.…”

Section: Post-processing Of Gesture Imagesmentioning

confidence: 99%

Research on gesture recognition of smart data fusion features in the IoT

Tan

Sun

et al. 2019

Self Cite

With the rapid development of Internet of Things (IoT) technology, the interaction between people and things has become increasingly frequent. Use simple gestures instead of complex operations to interact with the machine, the fusion of smart data feature information and so on has gradually become a research hotspot. Considering that the depth image of the Kinect sensor lacks color information and is susceptible to depth thresholds, this paper proposes a gesture segmentation method based on the fusion of color information and depth information, in order to ensure the complete information of the segmentation image, a gesture feature extraction method based on Hu invariant moment and HOG feature fusion is proposed, by determining the optimal weight parameters, the global and local features are effectively fused. Finally, the SVM classifier is used to classify and identify gestures. The experimental results show that the proposed fusion features method has a higher gesture recognition rate and better robustness than the traditional method.

“…In the case of environmental impacts, the literature [8] extracted the HOG features from pedestrian detection and overcome the environmental change factors. The literature [9] uses LBP features to classify different texture images, and the operation is simple. In literature [10], twelve Fourier descriptors are used as feature vectors for 10 gesture types.…”

Section: Related Workmentioning

confidence: 99%

“…The background is removed to retain only the foreground gesture area, and the feature data of the gesture are extracted to provide sample data for subsequent recognition. Due to the variability of the background, the diversity of gestures, and the interference of the illumination, it makes the gesture segmentation difficult and affects the recognition rate [8][9]. Therefore, it is extremely important to accurately and completely segment the gesture.…”

Section: Gesture Image Preprocessingmentioning

confidence: 99%

Jointly network: a network based on CNN and RBM for gesture recognition

Cheng

Sun

et al. 2018

Self Cite

Hand belongs to non-rigid objects and is rich in variety, making gesture recognition more difficult. The essence of dynamic gesture recognition is the classification and recognition of single-frame still images. Therefore, this paper mainly focuses on static gesture recognition. At present, there are some problems in gesture recognition, such as accuracy, real-time or poor robustness. To solve the above problems, in this paper, the Kinect sensor is used to obtain the color and depth gesture samples, and the gesture samples are processed. On this basis, a jointly network of CNN and RBM is proposed for gesture recognition. It mainly uses superposed network of multiple RBMs to carry out unsupervised feature extraction and combined with supervised feature extraction of CNN. Finally, these two features are combined to classify them. The simulation results show that the proposed jointly network has a better performance in identifying simple background gesture samples, and the recognition capability of gesture samples in complex background needs to be improved.