Gesture recognition based on modified adaptive orthogonal matching pursuit algorithm

Abstract

Aiming at the disadvantages of greedy algorithms in sparse solution, a modified adaptive orthogonal matching pursuit algorithm (MAOMP) is proposed in this paper. It is obviously improved to introduce sparsity and variable step size for the MAOMP. The algorithm estimates the initial value of sparsity by matching test, and will decrease the number of subsequent iterations. Finally, the step size is adjusted to select atoms and approximate the true sparsity at different stages. The simulation results show that the algorithm which has proposed improves the recognition accuracy and efficiency comparing with other greedy algorithms.

Surface EMG data aggregation processing for intelligent prosthetic action recognition

“…In a variety of sEMG noise, power frequency interference is the main noise in the whole frequency range of EMG, because it is first considered. Although most EMG signal acquisition instruments have taken some hardware measures to deal with it, such as trap filter and high mode rejection ratio amplifier, the power frequency noise still exists [29][30]. The power frequency noise is mainly caused by the nonlinear characteristics of electronic devices, such as switching power supply, electronic fluorescent lamp ballast, uninterrupted power supply and so on.…”

“…The support vector machine is mainly used for multi classification, the kernel function type is RBF (radial basis) kernel function, the kernel function parameter is 3.05176*10-5, the total number of classes is 9, the whole SVM model support vector number is 278, the support vector of each category is numbered. Instead of 37,32,14,21,26,37,30,42 and 39, the model will list all the support vectors.…”

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

“…In recent years, the use of data gloves for gesture recognition has triggered a wave of gesture recognition research [4][5]. Some scholars propose a Kinect-based gesture recognition method that uses FEMD algorithm to achieve stable static gesture recognition, but the algorithm requires a large of data for training [6][7]. Some researchers have combined BPNN and PSO algorithms on the basis of traditional algorithms, which greatly reduces the training time and improves the accuracy of gesture recognition [8].…”

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

“…In literature [6], In order to improve the fineness of the description of the features of Hu invariant moments, three characterization vector formulas have been added for recognition research. In literature [7] fuses the information of the first four Hu moments and the contour of the gesture, which improves the accuracy of recognition. In the case of environmental impacts, the literature [8] extracted the HOG features from pedestrian detection and overcome the environmental change factors.…”