Machine vision-based system for detecting abnormal state of scraper

Abstract

The invention discloses a machine vision-based system for detecting an abnormal state of a scraper. The system is characterized by at least comprising an image collection unit and a data processing unit, wherein the image collection unit comprises a video camera and an auxiliary light source; the video camera is used for collecting a video image of a running state of a scraper plate; the auxiliarylight source is used for illuminating the surface of the scraper plate; the data processing unit is used for receiving and processing video image data output by the video camera, and when the scraperis stopped due to tilting and chain scission, storing and recording a field video image, and sending a fault alarm signal; the data processing unit performs tagging on the video image output by the video camera by utilizing an image coordinate system; an image detection region is manually set in the video image; at least two scraper plates are shot in the image detection region; a virtual standard line is set in the image detection region; and the virtual standard line is parallel to the longest edges of the upper surfaces of the shot scraper plates. An equipment stop fault due to the scraperplate tilting and the chain scission can be detected in a scraper running process.

Description

technical field [0001] The invention relates to the technical field of machine vision applications, in particular to a machine vision-based abnormal state detection system for scraper machines. Background technique [0002] The chain scraper machine used in the coal washing plant has the problem that the scraper is inclined and the chain is broken during production and operation, forcing the equipment to stop. Accurate fault detection and timely alarm can avoid equipment damage and reduce economic losses. At present, the monitoring of the scraper machine mainly relies on workers' inspections. This traditional method of workers' inspections consumes a lot of manpower, and the processing efficiency is relatively low. In recent years, with the rapid development of science and technology, some people have proposed to use mechanical and other electronic sensors to monitor the operation status of the scraper machine. For example, the Chinese utility model patent: the patent appli...

AI Technical Summary

Problems solved by technology

At present, the monitoring of the scraper machine mainly relies on workers' inspections. This traditional method of workers' inspections consumes a lot of manpower, and the processing efficiency is relatively low.

In recent years, with the rapid development of science and technology, some people have proposed to use mechanical and other electronic sensors to monitor the operation status of the scraper machine. For example, the Chinese utility model patent: the patent application number is CN200720173760.8, and the patent name is: comprehensive detection of scraper machine device, but through the process of practice, it is found that the comprehensive detection device of the scraper machine with the above structure has the defects of complex structure and cumbersome use process, and it is also not conducive to installation and maintenance.

Images