Coal and rock identification method based on generalized gamma distribution in transform domain

Abstract

The invention discloses a coal rock identification method based on the generalized gamma distribution in the transform domain. In the sample training stage, multiple coal training sample images and multiple rock training sample images that meet certain conditions are obtained, and each image is subjected to two-way wavelet transformation, and the root mean square wavelet coefficients arranged according to certain sorting rules are obtained after correlation processing. band, calculate the parameters of these root mean square wavelet coefficient subbands under the condition of generalized gamma distribution, and use these parameters to construct the feature column vector of training samples; in the stage of coal and rock identification, collect test sample images of unknown categories that meet certain conditions , similar to the processing method of the training sample image, a series of processing is performed on the test sample image, and finally the test sample feature column vector is obtained. By comparing the similarity between the feature column vector of the test sample and the feature column vector of the training sample, the coal rock type of the test sample is judged. The invention has good reliability, high correct recognition rate and easy software and hardware realization.

Description

technical field [0001] The invention relates to a coal rock recognition method based on a generalized gamma distribution in a transform domain, and belongs to the technical field of image recognition. Background technique [0002] Coal and rock identification refers to the automatic identification of coal and rock through various technical means. In the process of coal resource mining and transportation, there are many production links that need to distinguish between coal and rock, such as the adjustment of the height of the shearer drum, the process control of top coal caving in fully mechanized mining, and the selection of raw coal in coal preparation plants. Since the 1950s, South Africa, Australia, Germany, the United States, China and other major coal-producing countries in the world have carried out a series of research on coal rock identification methods, and have produced some representative research results, such as natural γ-ray detection method, Radar detection ...

AI Technical Summary

Problems solved by technology

However, these methods all have the following common problems: (1) It is necessary to install and deploy various sensors on existing equipment, resulting in complex structure and high manufacturing cost; The force is complex, the vibration is severe, the wear is serious, the deployment of the sensor is relatively difficult, the electronic circuit is also easily damaged, and the reliability of the device is poor; (3) For different types of mechanical carrier equipment, the selection of the sensor and the selection of the installation position are difficult. There is a big difference, which requires personalized customization, so its universality is not good

The existing coal and rock recognition methods based on image processing still have a lot of room for improvement in terms of robustness and recognition rate.

Images