Lithium-containing pegmatite vein extraction method based on full convolutional neural network

Abstract

The embodiment of the invention discloses a lithium-containing pegmatite vein extraction method based on a full convolutional neural network, and the method comprises the steps: obtaining a remote sensing image of a lithium-containing pegmatite vein region, and carrying out the processing of the remote sensing image; constructing a full convolutional neural network model, using the processed remote sensing image for model training and parameter adjustment, wherein the full convolutional neural network is provided with a loss function, inputting the processed remote sensing image into the trained full convolutional neural network model, carrying out lithium-containing pegmatite vein marking on the processed image to obtain an output result of the full convolutional neural network model; and splicing output results of the full convolutional neural network model to obtain an extraction result image of the lithium-containing rock vein in the region.

Description

technical field [0001] The invention relates to the field of mineral resource extraction. More specifically, it relates to a lithium-bearing pegmatite vein extraction method based on a fully convolutional neural network. Background technique [0002] Lithium metal is a metal resource with extremely high strategic value. It is widely used in the fields of atomic energy, special alloys, special glass and new energy batteries. The current social production has a great demand for lithium. Limited by nature and mining conditions, my country's Lithium ore is mostly produced in hard rock spodumene deposits. Most of this type of deposits are located in relatively remote areas with few people, complex terrain and harsh environment, so it is difficult to carry out regional geological survey research and ore prospecting work. [0003] For hard rock spodumene deposits, spodumene-containing pegmatite veins are important targets for lithium ore prospecting, and in various mineral explora...

AI Technical Summary

Problems solved by technology

At present, the identification of lithium-bearing pegmatite dikes in remote sensing images is still mainly done by manual interpretation. Faced with remote sensing images containing huge amounts of complex surface information, manual interpretation is becoming more and more difficult in many cases. Although based on The research on remote sensing image extraction technology of traditional methods started earlier, but because the interpretation of remote sensing images is affected by the complex geographical environment, traditional methods do not make full use of image information, mostly based on the underlying features of images such as spectral features, texture features, etc. , geometric features, etc., the feature selection is relatively single, and it is easy to have the phenomenon of "same object with different spectrum, different object with same spectrum", so the extraction results sometimes cannot meet expectations

Images