Sea ice recognition method and system based on multi-source optical remote sensing image

Abstract

The invention provides a sea ice recognition method and system based on a multi-source optical remote sensing image, and the method comprises the steps: inputting the optical remote sensing image into a sea ice semantic segmentation model, and obtaining a sea ice recognition result; wherein the sea ice semantic segmentation model is obtained by training based on a sample optical remote sensing image and a corresponding sea ice mask image; wherein the sample optical remote sensing image is determined based on original remote sensing images acquired by various optical remote sensing sensors. The system comprises: importing an optical remote sensing image to be recognized into the system for sea ice recognition, after a recognition result is obtained, manually editing the recognition result in a sea ice pattern spot editing module, and generating a sea ice monitoring report through one key after editing is completed. According to the method and system provided by the invention, the application of the deep learning technology can overcome the problem of poor generalization ability of a traditional algorithm, and the pattern spot editing function in the system can further improve the precision and reliability of sea ice monitoring. In addition, the sea ice semantic segmentation model can be compatible with most medium-high resolution satellite remote sensing data.

Description

technical field [0001] The invention relates to the technical field of remote sensing monitoring, in particular to a sea ice identification method and system based on multi-source optical remote sensing images. Background technique [0002] Sea ice identification is one of the important applications in the field of remote sensing monitoring. The accuracy of sea ice identification is of great significance for evaluating sea ice conditions, dealing with sea ice disasters, and ensuring navigation safety. [0003] Remote sensing technology can quickly obtain information on large areas of sea ice and ice conditions. At present, the use of remote sensing means to monitor sea ice at sea mainly starts from two aspects: data and algorithms: [0004] In terms of data, sea ice identification is mainly based on optical images and radar images. Among them, radar images have the characteristics of being unaffected by weather and all-weather and all-weather earth observation. limitations...

AI Technical Summary

Problems solved by technology

[0006] The present invention provides a sea ice recognition method and system based on multi-source optical remote sensing images to solve the problems of poor generalization ability and complicated recognition schemes in the existing sea ice recognition

Images