Chen Tao. Human Movement Gait Recognition Based on Improved Spatio-Temporal Gait Diagram. Dynamic Systems and Applications 29 (2020) No. 5, 2140 – 2147

https://doi.org/10.46719/dsa202029549

ABSTRACT.
Gait is a new biometric feature interaction recognize the physical distance and the image forming apparatus. In addition, even at low resolve, which makes it ideal for use in many human recognition application, such as video surveillance, monitoring and access control systems good performance. Most existing human gait recognition solutions to extract contours of the body, the body or from an image based on the shape and gait features and Construction. Thus, the performance of this algorithm is highly dependent on the accuracy of the segmentation of the body, but it is still a challenging problem in the literature. In this article, we recommend the use of space-time motion characteristics of individual identification of human gait, contour extraction is not new gait recognition algorithm. The algorithm set of local descriptors extracted from the video sequence in gait. Gaussian mixture model Fisher vector descriptors extracted by the encoding codebook based encoding. Wherein encoding of a simple linear support vector machine to determine the individual classification. The results show that our method shows all five excellent database performance and the most advanced gait recognition method is better than the country.

Keywords: Recognition Algorithms. Gait, Spatio-Temporal, Conventional Identification.