Offline image structured analysis method, device, system and storage medium

Abstract

The invention provides an offline image structured analysis method, a device, a system and a storage medium. The method comprises the steps that an offline task service module creates a task for performing structural analysis on an offline image; the offline task service module registers each subtask on the input / output multiplexer, transmits the plurality of subtasks to the algorithm service module, and releases CPU and memory resources after registration; the algorithm service module carries out structural analysis on a plurality of subtasks, and transmits structural data of the subtasks carrying identifiers of the subtasks to the input and output multiplexer every time the structural analysis of one subtask is completed. After receiving the structured data, the input / output multiplexertransmits a notification that the structured data is readable to an offline task service module, which reads the structured data from the input / output multiplexer in response to the notification. According to the offline image structural analysis scheme, IO blocking time is shortened, and the performance of a structural analysis task is optimized.

Description

technical field [0001] The present invention relates to the technical field of image structuring, and more particularly relates to an off-line image structuring analysis method, device, system and storage medium. Background technique [0002] Video structuring is a technology for structurally extracting important information in video content. Using it, video content can be organized into computer-readable formats by using time-space segmentation, feature extraction, object recognition, deep learning and other processing methods according to the semantic relationship. and human-understandable textual information or visual graphical information. Wherein, the offline image structured analysis refers to the structured analysis and processing of offline pictures and videos. [0003] Existing offline image structural analysis methods usually use synchronous Hypertext Transfer Protocol (Http) requests to transmit offline task data to the algorithm service module. However, synchro...

AI Technical Summary

Problems solved by technology

However, synchronous Http requests are based on the synchronous input and output (IO) blocking model, from datagram preparation to datagram copy, blocking the whole process; how many IO events there are, how many threads will be blocked; and, with the sudden increase in the number of requests , the overhead of thread context switching will become larger and larger, the garbage collection (GC) time of the entire service will be longer and longer, and the processing speed will be slower and slower. In addition, the number of IO events that this model can handle is also very limited; moreover, using When synchronizing Http requests, from the request to the data response to the system kernel, the synchronous IO is in the blocking stage, and the central processing unit (CPU) resources will be occupied

Images