Marine target significance detection method based on spectrum singular value decomposition

Abstract

The invention relates to a marine target significance detection method based on spectrum singular value decomposition. The method comprises the following steps of utilizing the brightness and color channels of a marine image CIELab space to respectively carry out Fourier transform; according to set threshold values, selecting the non-main components of amplitude spectrum, and combining with original phase spectrum to carry out Fourier inverse transform, to obtain the significance map of each feature; and combining the color significance maps, and then combining with brightness significance maps to obtain a total significance map. The method has the advantages that a significance area in the marine scene is quickly extracted, so the target detection of a marine scene is favorably realized, the interference of marine clutters is better inhibited, the combination of significance maps with a plurality of dimensions is not needed, and the method can be realized on original image dimension; the method provides machine vision auxiliary means for target detection in marine peril searching and rescuing, marine monitoring, port video monitoring, detection of various ships in marine enforcement evidence collection and the like.

Description

technical field [0001] The invention relates to a sea target saliency detection method based on frequency spectrum singular value decomposition (singular value decomposition, SVD). Background technique [0002] In the past ten years, many scholars at home and abroad have conducted in-depth research on the attention mechanism of human vision. Scholars in the field of computer vision have done a lot of research work in proposing various computational models of visual attention. At present, the research results of visual attention have been applied in image retrieval, image quality evaluation, image and video coding, object detection and tracking and other fields. Saliency detection is the core problem of visual attention models, that is, how to measure the saliency of images. According to the image space division utilized, it can be divided into spatial domain method and frequency domain method. The essence of frequency-domain saliency detection methods that have been propo...

AI Technical Summary

Problems solved by technology

[0005] However, since the single-scale spectral residual method, phase spectrum method, and discrete cosine transform method are sensitive to the image scale, different saliency maps will be obtained when the image scale is changed, and the internal area of ​​the salient target cannot be completely extracted.

The method of using multiple scale fusion will increase the complexity of the algorithm, especially without any prior information, how to choose the fusion scale is also a problem that needs additional consideration

Images