Coal gangue identification method based on cross algorithm edge detection theory and visual features

Abstract

The invention relates to a coal gangue identification method based on a crossover algorithm edge detection theory and visual features, and the method comprises steps of providing the crossover algorithm edge detection theory according to the gray values of gray images of coal and coal gangue and the phenomenon of monotonically increasing along the oblique-45-degree and 45-degree directions near the edge, and carrying out edge detection on the coal and coal gangue images by using a crossover algorithm, and carrying out morphological technology processing on a detection result to obtain a segmentation result of a single coal and coal gangue image. And extracting a mean value, a contrast ratio and an entropy value of the coal and coal gangue single images as identification features to construct a support vector machine classification model. The problems of edge detection errors, image segmentation distortion, low recognition precision and the like existing in coal and gangue sorting under the complex background condition are solved, image segmentation is accurate, recognition features are few, and the recognition accuracy is high.

Description

Technical field [0001] The present invention relates to a method of coal and rock gangue sorting field, in particular a crossing edge detection algorithm based on the theory and visual features. Background technique [0002] The efficient and clean use of coal is one of the themes of the coal industry, environmental protection and air pollution control more and more attention to social and scientific researchers. Refuse to reduce the rate of coal is an effective way to reduce solid waste generation and emissions, coal can improve quality, reduce transportation costs and the liberation of labor, while promoting the gangue solid waste recycling. [0003] Various characteristics of coal and gangue differences exist in physics and chemistry, many scholars by differential characteristics of coal and gangue in physics binding lossless NN model identification. Common characteristics of the physical aspects of the differences are mainly characterized in density, hardness characteristics,...

AI Technical Summary

Problems solved by technology

[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a coal gangue recognition method based on the crossover algorithm edge detection theory and visual features; solve the edge detection errors, image segmentation distortion and Problems such as low recognition accuracy, accurate image segmentation, few recognition features, and high recognition accuracy

Images