A SAR ship recognition method and system combining saliency and neural network

Abstract

The present invention provides a SAR ship recognition method and system combining saliency and neural network, including data preprocessing, removing coherent speckle noise of SAR image through Lee filtering, while maintaining the edge information of the image, and then cropping the SAR image of each scene Obtain image blocks; construct a data set, including selecting SAR image blocks containing ships in different scenes, and mark the position information of ships with a rectangular minimum bounding box to obtain a SAR image ship data set containing label information; construct a fusion saliency perception Convolutional neural network, including extracting features through the Darknet53 network, calculating ship candidate frames and confidence based on the obtained multi-scale feature map, obtaining the salient feature map in the candidate frame based on the global contrast method, and taking the outer rectangle of the salient area as the detection Results; training network, for the SAR image to be recognized, the image block is obtained after preprocessing by row, and then the network model obtained by training is used for prediction, and the SAR image is re-stitched based on the predicted image block.

Description

technical field [0001] The invention belongs to the field of ship recognition processing at sea, and in particular relates to a SAR ship recognition method and system combining saliency and neural network. Background technique [0002] Maritime vessel identification plays a very important role in maritime traffic management, oil spill pollution control, maritime safety management, and maintenance of marine rights and interests. Remote sensing images currently used for ship identification at sea include optical remote sensing images, reflective infrared remote sensing images, hyperspectral remote sensing images, thermal infrared remote sensing images, and radar images. Objects emit radio waves, and the receiver receives the image formed by the scattered echoes, which has the characteristics of all-day and all-weather. Among them, synthetic aperture radar (SAR) is the most suitable radar for ship target detection. With the successful launch of China's Gaofen-3, Japan's ALOS-...

AI Technical Summary

Problems solved by technology

[0005] (1) Affected by sea clutter and coherent speckle noise, some ships are too affected by noise to be correctly identified

[0006] (2) Ships in the SAR image are deformed and affected by noise, which leads to inaccurate bounding box positions of the recognition results, and the intersection ratio is not high

[0007] (3) The lack of color and other features in the SAR image leads to indistinct distinction between coastal vessels and the highlighted areas of land buildings, resulting in a high detection false alarm rate

[0008] (4) The geometric deformation caused by the incident angle will affect the size of the ship in the SAR image and the diversity of the ship will cause the ship to be multi-scale in the SAR image

Images