Remote sensing sea ice image classification method based on convolutional neural network

Abstract

The invention discloses a remote sensing sea ice image classification method based on a convolutional neural network. The method solves the problems that a traditional method cannot fully excavate hyperspectral remote sensing sea ice image spatial-spectral characteristics and cannot effectively distinguish different spectral characteristic contribution degrees in combination with a classificationtarget. According to the technical scheme, the method is characterized by comprising the following steps that original data are obtained through an original remote sensing image; a part of samples aremanually marked from the original data as a sample library; a training sample is randomly selected for the input data according to a set strategy; other samples are taken as test samples; training and feature extraction are carried out on a pre-built three-dimensional convolutional neural network model through the training sample, weight adjustment is carried out on extracted features through anextrusion excitation network, and finally a support vector machine classifier is selected to complete classification; and the hyperspectral remote sensing images are detected and classified through the trained and tested three-dimensional convolutional neural network model. According to the method, the existing difficulties can be effectively overcome, and the classification precision of the remote sensing sea ice images is improved.

Description

technical field [0001] The invention relates to the field of sea ice detection, in particular to a sea ice image classification method based on convolutional neural network remote sensing. Background technique [0002] Sea ice is one of the most prominent marine hazards in high-latitude regions. Its freezing, thawing and drifting will have a significant impact on production operations in coastal areas and oceans. Therefore, in order to quickly and accurately assess sea ice conditions, timely forecast sea ice disasters, and ensure personal and property safety, sea ice detection research is of great significance, and sea ice classification is an important content of sea ice detection. [0003] Remote sensing technology is an effective means of sea ice detection, which can obtain sea ice data in a timely and large scale, and has been widely used in sea ice detection. In recent years, commonly used remote sensing data include aperture radar, multispectral satellite imagery wit...

AI Technical Summary

Problems solved by technology

However, most of these methods improve network performance through spatial dimensions, without considering the complex correlation between spectra.

Images