Multipath analysis method and device

Abstract

The invention relates to the technical field of data processing, and particularly provides a multi-channel analysis method, which is characterized in that a data acquisition unit acquires image frames of multi-channel video channels in a camera equipment cluster and sends data to a data receiving and caching unit, a plurality of artificial intelligence analysis tasks of multi-channel video data are set by a dynamic configuration service, and the multi-channel video data is subjected to multi-channel analysis. And the scheduling center continuously performs monitoring configuration by adopting a long polling method, senses the change of the analysis task and dynamically adjusts the analysis capability of the analysis unit in the calculation center. Compared with the prior art, the method has the advantages that the high efficiency and the flexible expansibility of multiple paths of videos during execution of multiple artificial intelligence analysis tasks are improved.

Description

technical field [0001] The invention relates to the technical field of data processing, and specifically provides a multi-path analysis method and device. Background technique [0002] With the development of artificial intelligence technology, the analysis of video data has become one of the important branches of computer vision. Compared with images, videos have more abundant information, and at the same time, the challenges are also greater, especially under the domestic CPU and OS, artificial intelligence Video analysis still has the challenge of weak ecology. For video processing services, one video usually corresponds to an artificial intelligence analysis person, for example, it is only used for face recognition. The function of this service is relatively simple. However, with the continuous enrichment of scenes, more and more detection and Recognition tasks need to be added continuously and dynamically. For example, when a road of video data is performing face recog...

AI Technical Summary

Problems solved by technology

[0002] With the development of artificial intelligence technology, the analysis of video data has become one of the important branches of computer vision. Compared with images, videos have more abundant information, and at the same time, the challenges are also greater, especially under the domestic CPU and OS, artificial intelligence Video analysis still has the challenge of weak ecology

For video processing services, one video usually corresponds to an artificial intelligence analysis person, for example, it is only used for face recognition. The function of this service is relatively simple. However, with the continuous enrichment of scenes, more and more detection and Recognition tasks need to be continuously and dynamically added. For example, a video data is performing face recognition tasks. When a new analysis task mask detection needs to be added, it needs to be shut down and deployed to perform multiple analysis tasks at the same time.

[0003] The existing artificial intelligence analysis service of multi-channel video adopts the method of pre-determining the detection or identification task to load the service, which cannot effectively and dynamically cope with the continuously enriched video analysis capabilities. This method limits the expansion ability of multi-channel video for multiple analysis tasks. lack of flexibility

Images