Target characteristic extraction method, identification method, extraction device and identification system for synthetic aperture sonar image

Abstract

The invention discloses a target characteristic extraction method, identification method, extraction device and identification system for a synthetic aperture sonar image. The target characteristic extraction method includes the steps of extracting shadow geometrical characteristics which include the target shadow zone area, the perimeter zone contour length, the eccentricity ratio, shape parameters, direction values and the ratio of the largest width of a target to the largest width of a shadow, obtaining a characteristic space formed by the shadow geometrical characteristics, extracting the target normalization central moment, obtaining a characteristic space formed by the target normalization central moment, extracting statistical property parameter difference characteristics of a background, the target and the shadow, obtaining shape parameters and dimension parameters, then obtaining a characteristic space formed based on the statistical property parameter difference characteristics of the background, the target and the shadow, and forming a new characteristic space formed by the three characteristic spaces. According to the target characteristic extraction method and identification method for the synthetic aperture sonar image, an underwater object can be effectively represented, and detection and identification of the synthetic aperture sonar image are facilitated.

Description

technical field [0001] The present invention relates to the field of synthetic aperture sonar (Synthetic Aperture Sonar, referred to as SAS) image detection and recognition, in particular, the present invention relates to a synthetic aperture sonar image target feature extraction method and recognition method, extraction device and recognition system. Background technique [0002] Synthetic aperture sonar is a high-resolution underwater imaging sonar that can obtain high-quality underwater images. Synthetic aperture technology uses multiple echoes to coherently superimpose to obtain an aperture, so that images with high precision in azimuth and range can be obtained. In recent years, the automatic detection and classification of synthetic aperture sonar images has important research value. The processing steps of the underwater imaging sonar recognition system (that is, the synthetic aperture sonar image target recognition system) include: image preprocessing, image segment...

AI Technical Summary

Problems solved by technology

Therefore, the traditional method is not conducive to the detection and recognition of synthetic aperture sonar images because it does not consider the shadow information of the target

Images