Remote sensing image sea surface ship detection method

Abstract

The invention provides a remote sensing image sea surface ship detection method, which comprises the following steps of 1, performing preprocessing operation on a remote sensing image to be detected,and improving an image visual effect while filtering image noise; 2, performing multi-feature saliency analysis on the preprocessed image by using a GBVS model to obtain a fused saliency map; 3, completing sea surface ship segmentation of the remote sensing image by taking a saliency analysis result as an initial condition of the improved Chan-Vse model; and 4, selecting the ROI as the minimum circumscribed rectangle area according to the TBR criterion, and finally realizing sea surface ship detection. The remote sensing image sea surface ship detection method effectively realizes remote sensing image sea surface ship detection under a complex background.

Description

technical field [0001] The invention relates to the field of remote sensing image sea surface ship detection, in particular to a remote sensing image sea surface ship detection method. Background technique [0002] Ships are key targets at sea. As an important research direction in the field of remote sensing image processing, ship detection is of great value in the fields of maritime traffic supervision and marine space planning. Using remote sensing images to automatically detect sea surface ship targets can avoid the defects of traditional manual interpretation, greatly improve efficiency, and reduce costs and labor. Most of the existing sea ship detection methods in remote sensing images are based on the method of threshold segmentation. This algorithm is simple and fast. average, and the gray scale of the ship is similar to that of the sea surface. The traditional sea ship detection method can easily lead to missed detection and false alarm. Therefore, how to effecti...

AI Technical Summary

Problems solved by technology

In view of the problems of large amount of data, noise interference, and unsatisfactory contrast in remote sensing images, the remote sensing images are grayscaled, median filtered, and linearly stretched to filter out useless information such as noise while emphasizing the region of interest; then GBVS is used The model performs multi-feature saliency analysis on the preprocessed image to generate a comprehensive saliency map; in order to avoid the defect that the Chan-Vese model is sensitive to initial conditions, the saliency analysis results are used as the initial area of ​​the improved Chan-Vese model to realize the Segmentation; finally, considering the number of pixels in the target and background areas comprehensively, based on the minimum circumscribed rectangle method, the sea surface ship target detection is finally completed

Images