Optical flow method based on video retrieval system

Abstract

An optical flow method based on video retrieval system includes the steps: preprocessing a video lens, performing frame segmentation on the given video lens, performing convolution smoothing on each frame of image, calculating optical flow value of a pixel point through every two adjacent frames of images after smoothing, then obtaining modular length and argument information of optical flow, utilizing the modular length and the argument information of the optical flow to distinguish frame images in the video lens, classifying the frame images with the same modular length size and argument direction as a category, constructing classified image frames of each category to optical flow histograms based on the modular length and the argument, respectively obtaining video lens optical flow histogram arrays based on the modular length and the argument, defining distance function based on the optical flow histogram arrays, matching the video lens similar to the given video lens from a video database, and completing a retrieval process.

Description

technical field [0001] The invention relates to computer pattern recognition processing technology, in particular to an optical flow method based on a video retrieval system. Background technique [0002] Content-based video retrieval technology has always been a hot issue in the field of multimedia information research. The general workflow is to divide a given original video into several related independent unit shots according to the mutation or gradient between image frames; then, extract certain features for each segmented shot as the index of the shot structure; finally, calculate the similarity between the shot to be tested and the target shot according to the established index structure, so that the corresponding result can be matched when the user submits a browsing or query request. According to the different extracted features, the existing methods of video retrieval can be subdivided into methods based on color features, methods based on domain space features, m...

AI Technical Summary

Problems solved by technology

However, the method of clipping the color histogram based on Alpha has two disadvantages: first, it discards the spatial information in the frame; The size is large, which makes the retrieval process extremely time-consuming

These methods have the characteristics of low computational cost, but the defects of such methods are also obvious, because what they represent is not the real motion of moving objects in the video content, but only the motion changes between shots, and the human visual system tend to be more easily influenced and attracted by the former

Motion features of moving objects in video content are the second important feature, Tahayna et al. (see B. Tahayna, M. Belkhatir, and S. Alhashmi. Motion Information for Video Retrieval. IEEE international conference on Mult imedia and Expo, pp. 870-873, 2009) and Feng et al. (see B.L. Feng, J. Cao, S.X. Lin, Y.D. Zhang, and K. Tao. Motion Region-based Trajectory Analysis and Re-ranking for Video Retrieval. IEEE international conference on Multimedia and Expo , pp.378-381, 2009) in their proposed method focused on retrieval through the motion trajectory of moving objects in the video. Experiments have proved that using motion trajectory is far superior to using lens motion features in terms of retrieval accuracy and recognition rate. , but the description of the motion trajectory also needs to establish a corresponding model, which will increase the calculation cost more than the first type of motion features, and spend more running time

Images