Static portrait data scheduling method, device and system and storage medium

Abstract

The invention provides a static portrait data scheduling method, device and system and a storage medium, and the method comprises the steps: obtaining a portrait pull request of a near-end business system, with the portrait pull request comprising identification information of a static portrait or the static portrait; judging whether a target image matched with the static portrait exists in the portrait cache or not according to the portrait pull request; if the target image exists in the portrait cache, returning the target image in the portrait cache to the near-end service system; if the target image does not exist in the portrait cache, sending a base map pull request to a remote service system, and obtaining a first response of the base map pull request; and when the first response comprises the target base map matched with the static portrait, returning the target base map to the near-end service system. The request pressure on the distributor is reduced through a caching strategy and a task retry mechanism. A good guarantee is provided for a failed task, and efficiency of data scheduling is greatly improved.

Description

technical field [0001] The present invention relates to the technical field of image processing, and more particularly relates to the processing of data scheduling of static portraits. Background technique [0002] In recent years, driven by the accelerated integration of security and AI and the application of safe cities and other industries, the security industry has rapidly moved towards high-definition, intelligence, and scale. As intelligence information becomes more and more important, the collection of important information such as the identity of the application personnel of the portrait data analysis system will be faster and more effective. However, the single-network portrait analysis system itself can only use data in its own system, and cannot effectively use resource information in other regions, which is not conducive to the timeliness of work development. At present, the data distribution system has realized data scheduling between different systems, but the...

AI Technical Summary

Problems solved by technology

However, the single-network portrait analysis system itself can only use the data in its own system, and cannot effectively use resource information in other regions, which is not conducive to the timeliness of work development

At present, the data distribution system has realized data scheduling between different systems, but the existing data distribution system terminal has no caching strategy. On the one hand, repeated requests will cause excessive pressure on the distribution machine, and on the other hand, it will take time to pull portrait pictures. Relatively long; in addition, the data distribution system terminal itself does not have a task retry mechanism, resulting in the lack of processing of failed tasks

[0003] Therefore, the data scheduling in the prior art has the problems of excessive pressure on the distributor, long data scheduling time, and no processing mechanism after scheduling failure, resulting in low efficiency of data scheduling between different systems

Images