Methods of Biometric Identification Based on Gait in the Smart City

The chapter presents the methods of biometric identification based on parallel shift technology. With the help of video sensors, a video sequence was formed, which was divided into video frames. The image of each frame of the video sequence was processed using the parallel shift technology and the function of area of intersection was formed, according to the shape of which identification is made. With the help of the obtained functions of the areas of intersection, the acceleration is determined, and various confidence intervals are used, which made it possible to achieve a high degree of identification. The technique of searching for the human motion vector, as well as determining the area of analysis of the silhouette of an object in the image, is described. Based on the parallel shift technology, the method of superimposing adjacent video frames is used. For this, binary images of the resulting human silhouettes are used on each video frame.