Raft culture area extraction method based on SAR image of Gaofen-3 satellite

Abstract

The invention discloses a raft culture area extraction method based on a Gaofen-3 satellite SAR image. The method comprises the following steps: firstly, acquiring Gaofen-3 satellite SAR image data including a research area; performing feature extraction on the SAR image data to obtain a plurality of extracted features; performing feature screening on the extracted features to obtain a plurality of effective features capable of reflecting the difference between the raft culture area and seawater; based on the plurality of effective features, using an iterative threshold algorithm to perform preliminary extraction of the raft culture area; and performing post-processing on the preliminary extraction result by using a density clustering algorithm, intersection and union set reconstruction and a morphological processing method, and extracting the raft culture area. According to the method, multiple features extracted from the SAR image of the Gaofen-3 satellite are used for extracting the raft culture area together, and then the iterative threshold method and the density clustering are combined, so that the influence of SAR image speckle noise can be reduced, and large continuous rows of raft culture areas can be accurately extracted.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for extracting a raft farming area based on the SAR image of the Gaofen-3 satellite. Background technique [0002] Mariculture is the dominant form of aquaculture in many coastal areas and is of high commercial value. Timely and accurate acquisition of the area of ​​mariculture area and its spatial distribution is of great significance for the fishery management department to provide auxiliary decision-making information for rationally planning the sea for aquaculture, controlling the aquaculture density, curbing the deterioration of the aquaculture environment, and preventing aquaculture diseases. Compared with traditional field surveys, remote sensing technology has irreplaceable advantages of high frequency, wide coverage, and low cost, and is a very potential evaluation method. In particular, the meter-level high-resolution remote sensing developed in recen...

AI Technical Summary

Problems solved by technology

[0004] However, the above methods often only rely on a single feature segmentation method to classify ground objects, and it is difficult to distinguish raft farming areas with small spectral differences from seawater.

Images