Coal dust image recognition method for underground coal mine explosion-proof detection

Abstract

The invention discloses a coal dust image recognition method for coal mine underground explosion-proof detection. The method comprises the following steps: 1, collecting and transmitting a coal dust infrared image; 2, image enhancement processing; 3, carrying out image segmentation processing; and 4, separating overlapped coal dust particles, and recognizing the coal dust particles. The method is simple in step, novel and reasonable in design and convenient to implement, coal dust overlapped particles can be reasonably separated out by adopting the fruit fly algorithm to separate the coal dust overlapped particles, good robustness is obtained, the recognition effect is good and stable, the speed is high, the image compatibility is extremely high, the recognition precision of the coal dust overlapped particles is improved, and the method is suitable for popularization and application. The method is good in effectiveness and robustness, high in adaptability, flexible and convenient to use, high in practicability, good in using effect and high in application and popularization value.

Description

technical field [0001] The invention belongs to the technical field of coal mine environment monitoring, and in particular relates to a coal dust image recognition method for underground explosion-proof detection of coal mines. Background technique [0002] China is a large coal-producing country, and coal will still be the main energy structure for a long time in the future. However, most of the coal mines in my country are mined underground, and there are many unsafe factors. Disaster accidents such as gas, coal dust and fire occur frequently, causing serious damage, many injuries, long interruption of production, and damage to shaft engineering or production equipment. Therefore, how to strengthen mine disaster prevention and control work, how to correctly handle the relationship between safety and production, safety and benefit, how to accurately, real-time, and quickly perform coal mine safety monitoring functions, and ensure the efficient operation of emergency rescue ...

AI Technical Summary

Problems solved by technology

Obviously, the application of edge tracking algorithm cannot well solve the problem of coal dust particle identification. Although the method of mathematical morphology is not limited by the shape of the analyzed object, the accuracy of its segmentation algorithm is low; Particles; while the area outline obtained by the watershed transform algorithm has airtightness, connectivity, single pixel width and precise position, but the watershed algorithm has the problem of over-segmentation

Images