Sea chart aquiculture area recognition method based on multispectral remote sensing image

Abstract

The invention discloses a sea chart aquiculture area recognition method based on a multispectral remote sensing image. The method comprises the following steps of 1, performing image fusion on the multispectral remote sensing image and a high-resolution panchromatic image of a shallow water area of a coastal zone; 2, using a local space correlation statistical index to describe aquiculture area textures of the fused high-resolution multispectral image acquired in the step 1; 3, building a support vector machine classification model, using an operation result acquired in the step 2 to serve asa texture band and input into a classification model together with the fused high-resolution multispectral image for classification, and thus recognizing offshore floating rafts and net cage aquiculture areas; and 4, registering the remote sensing image with the sea chart after classification, performing pattern spot embedding on an aquiculture area extraction result by comparing with the sea chart, thus achieving identification of the aquiculture area on the sea chart. According to the method provided by the invention, the offshore aquiculture area can be objectively, accurately and dynamically monitored, and the method and the technical support are provided for further developing navigation monitoring and investigation work of distribution of the coastal aquiculture areas.

Description

Technical field: [0001] The invention relates to a sea chart breeding area identification method based on multi-spectral remote sensing images, which is used for multi-spectral remote sensing image analysis and target identification, and belongs to the field of image processing and computer vision. Background technique: [0002] With the development of my country's aquaculture industry, the scope of aquaculture areas in coastal areas has continued to expand, which has produced huge economic benefits, but also caused a lot of negative impacts on the ecological environment and maritime traffic. Every year, there are many incidents where merchant ships stray into breeding areas. Merchant ships entering the aquaculture area will, on the one hand, affect or partially destroy aquaculture, resulting in economic disputes; on the other hand, accidents of fishing nets being entangled with propellers may also occur, which will affect the safe navigation of ships. At present, the offsh...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to propose a sea chart based on multi-spectral remote sensing images for the current problem that the distribution and location range information of large-scale offshore aquaculture areas cannot be dynamically obtained, and the information on offshore aquaculture areas is not marked on the current sea chart. Identification method of breeding area

Images