Optical remote sensing image ship target detection method fusing space-frequency domain features

Abstract

The invention provides an optical remote sensing image ship target detection method fusing space-frequency domain features. The method comprises the following steps: firstly, constructing an image feature map by utilizing a first-order gradient combination of a brightness feature map, a color feature map and brightness features of an image, carrying out non-overlapping partitioning on the image toobtain a plurality of image blocks, calculating a feature covariance matrix of each image block, and calculating a feature value between the feature covariance matrix of each image block and the feature covariance matrix of the whole image, and obtaining a normalized feature value graph as a spatial domain feature graph of the image; secondly, constructing quaternion features of the image, performing Fourier transform on the quaternion features, setting a transformed amplitude spectrum as 1 to leave phase information, performing inverse Fourier transform on the amplitude spectrum, and smoothing an inverse transform result by using a Gaussian filter to obtain a frequency domain feature map; and finally, fusing the feature maps of the spatial domain and the frequency domain by using a cellular automaton to obtain a final detection result. According to the method, the ship target can be rapidly and accurately detected from the remote sensing image with complex sea surface landform and cloud and fog backgrounds, the serious interference problem of the complex backgrounds on ship detection is solved, the ship detection effect under different complex backgrounds is good, and the application range is large.

Description

technical field [0001] The invention belongs to the field of target detection of remote sensing images, and more specifically relates to a ship target detection method of optical remote sensing images by integrating space-frequency domain features. Background technique [0002] As an important carrier of maritime shipping, ships can be effectively monitored to understand the shipping conditions in sea areas. In recent years, the continuous improvement of remote sensing technology has made optical remote sensing images more and more widely used in application scenarios. Optical remote sensing images contain a large amount of visual information. The special structural design of the ship that hides the radar can complete the ship detection task that the synthetic aperture radar could not complete in the past; it can also provide a large amount of image detail information similar to visual perception, which is convenient for manual interpretation. The development of optical rem...

AI Technical Summary

Problems solved by technology

[0009] Aiming at the fact that the existing visual saliency technology only considers the spatial domain or frequency domain features when detecting ship targets in complex backgrounds, and cannot suppress the interference of complex backgrounds while highlighting ship target features, the present invention proposes an optical Ship target detection method in remote sensing images to solve these problems

Images