Coal and rock recognition method based on random local image features

Abstract

The invention discloses a coal and rock recognition method based on random local image features. According to the method, the random local image features are used for describing coal images and rock images, the random local image features of coal samples and rock training samples are selected through a clustering algorithm to serve as a primitive dictionary, then the selected random local image features of the coal sample images and the rock sample images are labeled by the primitive dictionary according to the nearest-neighbor rule, a primitive frequency counting regular histogram of one coal sample image or one rock sample image shows one mode of coal or rock, and the coal features and the rock features are expressed in multiple modes; when the coal and the rock are recognized, random local image features are extracted and a histogram is built for images to be recognized according to the same method with the training images, and then the modes are compared with the modes learnt in the training stage, are measured through the Bhattacharyya coefficient and are recognized according to the nearest-neighbor rule. According to the method, images, under different light rays and different viewpoints, of different kinds of coal and rock serve as the training samples, and therefore influence by changes of light and imaging viewpoints is small, influence by changes of varieties of coal and rock is avoided, the recognition rate is high, and the stability is good.

Description

technical field [0001] The invention relates to a coal rock recognition method based on random local image features, belonging to the technical field of image recognition. Background technique [0002] Coal rock identification is to use a method to automatically identify coal rock objects as coal or rock. In the process of coal production, coal rock identification technology can be widely used in the production links such as drum coal mining, tunneling, caving coal mining, and raw coal gangue selection. The safe and efficient production of coal mines is of great significance. [0003] There are many coal rock identification methods, such as natural γ-ray detection method, radar detection method, stress pick method, infrared detection method, active power monitoring method, vibration detection method, sound detection method, dust detection method, memory cutting method etc., but these methods have the following problems: ① It is necessary to install various sensors on the e...

AI Technical Summary

Problems solved by technology

[0003] There are many coal rock identification methods, such as natural γ-ray detection method, radar detection method, stress pick method, infrared detection method, active power monitoring method, vibration detection method, sound detection method, dust detection method, memory cutting method etc., but these methods have the following problems: ① It is necessary to install various sensors on the existing equipment to obtain information, resulting in complex structure and high cost of the device

② Shearer drums, roadheaders and other equipment are subjected to complex forces, severe vibrations, severe wear, and large dust during the production process. It is difficult to deploy sensors, which easily leads to damage to mechanical components, sensors, and electrical circuits, and poor device reliability.

③ For different types of mechanical equipment, there is a big difference in the optimal type of sensor and the selection of signal pickup points, which requires personalized customization and poor adaptability of the system

[0004] Existing image-based coal and rock recognition methods are sensitive to imaging conditions such as illumination and viewpoint. If the imaging conditions of the coal or rock image to be recognized are different from those of the coal or rock sample image during training, the recognition rate will be greatly reduced; in addition , if the type of coal and rock to be recognized changes, it is necessary to re-sample coal and rock sample images to train the recognizer

Images