Method for recognizing lithology through reconstructed textures

Abstract

The invention discloses a method for recognizing lithology through reconstructed textures. The method comprises the following steps: establishing an original specimen image library; establishing a standardized image library; extracting features of the rock specimen image by using discrete convolution and maximum pooling operation of a convolutional neural network MobileNet-V2 convolutional model,and effectively preventing the problems of gradient explosion and gradient disappearance in the image feature extraction process by using a new algorithm; and selecting the rock category with the maximum probability to output a result, and completing the recognition process. According to the method, the problems of low precision and low recognition efficiency caused by the fact that an existing algorithm is limited to two dimensions in the field of randomly recognizing rock lithology under the conditions of particle distribution and section texture development and imaging quality including rock color, definition and the like are solved. The method has the advantages that various lithology characters can be widely identified without being limited by external factors such as backgrounds andcolors, and the identification accuracy is improved; intelligent identification operation speed improvement.

Description

technical field [0001] The invention relates to the field of intelligent processing and identification of rock lithology by computer or mobile intelligent terminal, more specifically, it is a convenient and efficient method for reconstructing texture and identifying lithology. Background technique [0002] The existing convenient identification algorithm of rock lithology is the traditional machine learning algorithm. First, preprocess the data, determine the classification method according to the preprocessed data characteristics, then analyze and visualize the data, determine which distribution the data conforms to, and finally test on the existing algorithm (decision tree, Bayesian), Model integration. There are four major defects in the existing algorithm: first, it can only identify rock types with obvious characteristic differences, such as the identification of coal rocks and other single rocks; second, it is greatly affected by rock color and imaging quality, and th...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the inadequacies of the above-mentioned background technology, and propose a method for reconstructing texture to identify lithology, which solves the problem of random identification of rock lithology in the field by existing algorithms. Restricted by texture development, the imaging quality, including rock color and clarity, is limited to two dimensions, resulting in low precision and low recognition efficiency

Images