Method, device, system, device and storage medium for analyzing equipment comprehensive efficiency based on attention mechanism target detection

Abstract

The invention discloses an analysis method, device, system, equipment and storage medium for comprehensive efficiency of equipment based on attention mechanism target detection. The analysis method includes: first, using the method of attention mechanism target detection to convert the effective Equipment and worker behaviors are identified, and irrelevant factory environments and invalid equipment are eliminated; secondly, appropriate video analysis methods are used to conduct targeted analysis on identified different equipment and worker behaviors, which improves the analysis speed and accuracy; Finally, the analyzed results are used for efficiency calculation and abnormal event analysis, realizing intelligent analysis of equipment comprehensive efficiency applied to factories, with strong self-adaptive ability, good precision and high efficiency.

Description

technical field [0001] The present invention relates to the technical field of video image analysis, in particular to an analysis method, device, system, equipment and storage medium for comprehensive efficiency of equipment based on attention mechanism target detection. Background technique [0002] Intelligent manufacturing is a general term for advanced manufacturing processes, systems and models with functions such as information self-perception, self-decision-making, and self-execution. Big data, artificial intelligence, etc. Smart manufacturing generally has four characteristics: smart factories as the carrier, the intelligence of key manufacturing links as the core, end-to-end data flow as the basis, and Internet connectivity as the support. [0003] With the development of intelligent manufacturing, the factory workshop tends to be unmanned and fully automatic management. How to manage the factory workshop in an unmanned environment to ensure the normal operation an...

AI Technical Summary

Problems solved by technology

[0005] First of all, there are many types of machines, transport vehicles, auxiliary equipment and other objects in the production workshop, which are often blocked by each other. The similarity of different process operations and frequent changes in light intensity in the workshop all bring new challenges to video and image analysis and recognition. To meet the challenge, the traditional video image analysis method processes the entire frame, and the adaptability is not strong, and the direct application effect in the factory environment is not good.

[0006] Secondly, the characteristics of each process in the manufacturing process are different, and there are disadvantages such as unclear working hours and complex workpieces. The traditional single-model video and image analysis models are weak in robustness and low in efficiency when directly applied

[0007] In the workshop environment of the factory, due to many problems in intelligent monitoring, when analyzing the comprehensive efficiency of equipment, manual recording is usually used, which is inefficient and unreliable

Images