Video playback method and device, electronic equipment and storage medium

Abstract

The invention relates to a video playback method and device, an electronic equipment and a storage medium. The method comprises the following steps: acquiring an interested image from a video; performing target identification on the interested image to obtain feature information of the interested image; obtaining a query condition, querying a picture with the interested image in the video according to the query condition and the feature information of the interested image, and determining a first picture with the interested image appearing for the first time and a second picture with the interested image appearing for the last time under the query condition; and adding the first video clip containing the first picture and the second video clip containing the second picture into a video playback list, and playing back the video clips in the video playback list. Through adoption of the method and the device, the problem that the behavior track of the target in the video is difficult to obtain in the related technology is solved, and the technical effect of quickly obtaining the behavior track of the interested target in the video is realized.

Description

technical field [0001] The present application relates to the technical field of video monitoring, in particular to a video playback method, device, electronic device and storage medium. Background technique [0002] Surveillance systems usually include: multiple network cameras and network hard disk video recorders. Each video file captured by the IPC corresponds to a video channel, and a video channel can be called a video. NVR has the ability to store and playback video files. [0003] With the maturity of image analysis algorithms and the improvement of chip computing capabilities, intelligent video analysis is more and more widely used in the field of video surveillance. Intelligent analysis in the field of surveillance can be divided into front-end intelligence and back-end intelligence according to the location of the algorithm. category. Among them, the front-end intelligence is mainly run on the network camera, and the back-end intelligence is mainly run on the s...

AI Technical Summary

Problems solved by technology

[0005] At present, the existing intelligent video analysis often collects the video image information of each physical area through acquisition equipment such as network cameras, transmits the video image data to the server, and stores and analyzes the data transmitted by the acquisition equipment through the server. The terminal displays the original data collected by the acquisition device and the data in the video after analysis. The videos under each channel displayed on the client are often sorted by channel and time. If someone stays in a certain shooting area for a long time, and Due to the large number of people in this area, occlusion is more likely to occur; at the same time, if the user wants to view the behavior trajectory information of the target in different shooting areas, it is often necessary to select all channels for one search or for a single channel for multiple searches. , To judge the activity of the target object in the event, it is necessary to manually click on the recordings corresponding to the event one by one to play back or to play back all the event recordings in sequence, and it is impossible to only play back the event recordings that only contain the target specified by the user

[0006] At present, there is no effective solution to the problem that it is difficult to obtain the behavior trajectory of the target in the video in related technologies.

Images