A video surveillance image storage and retrieval method

Abstract

The invention discloses a method for storing and retrieving video monitoring images. The method includes the following steps: synchronously recording the position parameters and optical parameter information of the camera in the video monitoring image files; after the video monitoring image files are transmitted to the server, the associated server According to the position parameter and the optical parameter information, the time-space coverage corresponding to the video surveillance image and the camera motion state are obtained; wherein, the X-axis of the time-space coverage is longitude, the Y-axis is latitude, and the Z-axis is time; The camera movement state mentioned above refers to rotation and stretching operations. A surveillance video spatio-temporal database is constructed from the spatio-temporal coverage, the motion state and corresponding video surveillance images; the time- and space-based query of the surveillance video is realized through the surveillance video spatio-temporal database. The realization of the scheme of the present invention makes up for the deficiency of video data retrieval based on image content, and realizes retrieval based on time and space coverage.

Description

technical field [0001] The invention relates to the field of video monitoring, and more specifically relates to a method for storing and retrieving video monitoring images. Background technique [0002] With the development and deepening of various policies such as safe city construction, many cities have begun to deploy a large number of monitoring equipment and have generated massive video data. How to retrieve these video data to obtain valuable information is still the key to safe city construction. a challenge. In order to solve the problems of difficult data organization and low retrieval efficiency in existing video data organization and retrieval, some people proposed to combine video data with camera location information, organize video data through maps, and index event occurrence time , Location search for video clips of interest, so as to realize the search and analysis of massive video data, and promote the construction process of safe cities. [0003] Toyama ...

AI Technical Summary

Problems solved by technology

[0005] Surveillance cameras do not have a positioning function, and placing key frames on the map is only a rough positioning, with a low degree of accuracy. It is necessary to manually place the video data on the map according to the recorded location information. There are human errors, which will also bring certain Accuracy is affected, and it is not suitable for large-scale promotion;

[0006] It only locates the position of the camera, and does not consider the temporal and spatial coverage of the video content. It can simply place the key frames of the video on the map, and cannot accurately save and reconstruct the surveillance video coverage. bring inconvenience;

[0007] It is only placed for the key frames of the video without considering the content of the entire video, so it is impossible to accurately infer the occurrence of continuous events in the surveillance video, which brings some troubles to the subsequent data analysis work

Images