Intelligent Interpretation Technology of Remote Sensing Images based on Deep Learning Algorithms

In recent years, high-resolution remote sensing images have been increasingly used in the field of geoscience. Therefore, how to interpret high-resolution remote sensing images with high quality is an urgent problem to be solved. This paper takes high-resolution remote sensing images as the research object, uses deep convolutional neural network (CNN) technology, conducts remote sensing image segmentation and feature extraction based on deep convolutional neural network, establishes a high-precision remote sensing image interpretation model, and improves the accuracy and computing speed of intelligent recognition of remote sensing images. This paper uses residual network (ResNet) as a learning sample to compare its performance with the classical algorithm on the same set of samples. Preliminary studies of this project have shown that the accuracy of convolutional neural networks in image analysis can reach 97.1%, and it shows obvious superiority at multiple spatial scales, and can effectively improve the efficiency of image interpretation.