Sedimentary rock category identification method and device, electronic equipment and storage medium

Abstract

The invention provides a sedimentary rock category automatic identification method, which comprises the steps of obtaining a rock image of a to-be-identified sedimentary rock, and performing expansion processing on the rock image to obtain an image set corresponding to the rock image; inputting the image set into an extraction layer in a category recognition model to obtain feature information of the image set; wherein the category recognition model is obtained by transforming and training a ResNet50 model; inputting the feature information into a self-built convolutional neural network in the category recognition model to obtain a classification score value corresponding to each category of rock; and determining a prediction category of the sedimentary rock according to the classification score value corresponding to each category of rock. The problem that the recognition accuracy is unstable due to an artificial experience recognition method is avoided, meanwhile, the problems that the accuracy of a traditional laboratory recognition method is guaranteed, but the cost is high and the period is long are avoided, and the problems that in the prior art, an adopted VGG model cannot well aim at a small data set, and the feature extraction effect is poor are also avoided.

Description

technical field [0001] The present application relates to the technical field of rock identification, in particular to a sedimentary rock type identification method, device, electronic equipment and storage medium. Background technique [0002] Field geological survey and drilling are important tasks in geology, mineral resources, hydrogeology and engineering geology. In this work, the identification of sedimentary rock types is the top priority of the work. Being able to quickly and accurately identify the types of sedimentary rocks in the field is the cornerstone for the normal development and advancement of field work. [0003] At present, the identification of sedimentary rock categories mainly relies on manual experience identification and traditional laboratory identification. Manual experience recognition is to judge the type of sedimentary rocks based on the experience of workers. Due to the interference of various factors such as experience differences, environment...

AI Technical Summary

Problems solved by technology

Manual experience recognition is to judge the type of sedimentary rock based on the experience of the worker. It is interfered by various factors such as experience differences, environmental factors, and subjective vision, and its recognition accuracy is unstable.

Traditional laboratory identification is to send sedimentary rock samples to the laboratory, and the laboratory staff will observe and analyze the samples in slices. The accuracy is guaranteed, but the cost is high and the cycle is long. It solves the problem that the VGG and other models used in the existing technology cannot be well targeted. Small data sets, poor feature extraction, and inaccurate identification of sedimentary rocks

Images