Monitoring video target concentration expression method and system in three-dimensional geographic scene model

Abstract

The embodiment of the invention discloses a monitoring video target concentration expression method and system in a three-dimensional geographic scene model, and relates to the technical field of visual fusion, and the method comprises the steps: reading a monitoring video for a period of time, obtaining image coordinates and geographic coordinates of a homonymous point pair, and establishing a mapping relation between a video image and a three-dimensional geographic space; detecting and processing the video target to obtain a video target subgraph and an image space position in the image; reordering the playing time of the video targets; and finally, dynamically mapping the reordered video target subgraphs into a geographic scene model according to the mapping relationship to realize video target concentration expression. The mapping relation between the video target and the geographic scene is established, the expression duration of the video target is compressed, the expression efficiency of the video moving target in the three-dimensional environment is improved, the fusion expression effect of the monitoring video in the geographic scene is enhanced, and great convenience is provided for integrated rapid retrieval and efficient understanding of video geographic scene information.

Description

Technical field [0001] The embodiment of the present invention relates to the technical field of visualization fusion, and specifically relates to a method and system for condensed expression of surveillance video targets in a three-dimensional geographic scene model. Background technique [0002] Surveillance video is widely used in security, traffic, environmental monitoring and other fields. In the past ten years, the number of new cameras in my country has reached millions of new cameras each year, and it has grown rapidly year by year. While the video data has grown substantially, the pressure on the effective processing and analysis of the video targets is also increasing. The video target in surveillance video has the characteristics of large amount of data and sparse distribution, which makes manual retrieval and analysis very difficult. In order to extract video targets from massive video data and play them together to assist retrieval and analysis, video concentration...

AI Technical Summary

Problems solved by technology

[0004] For this reason, the embodiment of the present invention provides a method and system for condensing and expressing surveillance video objects in a three-dimensional geographic scene model to solve the Due to the large quantity and the inability of existing video enrichment methods to realize the fusion and expression of video objects and 3D geospatial information, users cannot fully and efficiently understand the content of surveillance video

Images