Image classification method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides an image classification method and device, equipment and a medium, and the method comprises the steps: employing a first classification model, respectively detecting the types of a plurality of image frames of a target video, obtaining a first classification result and first feature data of each image frame, and carrying out the first feature data for the image frames of which the first classification results are the target types, performing feature extraction on the image frame by adopting a feature extraction network of a second classification model to obtain second feature data; and according to the first feature data and the second feature data, detecting the category of the image frame by a classification network of a second classification model to obtain a second classification result replacing the first classification result of the image frame. And filtering by using the second classification result on the basis that the first classification result is the target category. Since the first feature data and the second feature data are directly used for fusion, feature complementation is realized at a low cost, so that the detection accuracy of the second classification result is improved, and the classification accuracy of the category of the image frame is improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image classification method, device, electronic equipment and storage medium. Background technique [0002] With the continuous development of video technology, especially the rise of short videos, there are more and more videos of various types, which inevitably brings a lot of pressure to the video review work. [0003] At present, the mainstream video classification technology is to use the convolutional neural network to achieve frame-level image classification, and then use the frame-level classification results to obtain the classification results of the video, so as to realize the classification of the video. However, since the video is composed of many frames, the workload of classifying image frames is very large, so the processing speed will be very slow. In the process of business application, it is found that in order to meet the speed requirement...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiments of the present invention is to provide an image classification method, device, electronic equipment, and storage medium, which solve the problem of poor recognition effect and false detection to a certain extent in order to meet the speed requirements of actual online use. high rate of technical issues

Images