Target detection method based on region growth

Abstract

The invention discloses a target detection method based on region growth. The method comprises steps that S1, based on the morphological filtering mechanism, progressive scanning is carried out through utilizing the overall texture distribution characteristics of an image, valid maximum and minimum points are selected, and the size of structural elements is adaptively determined for image filtering; and S2, based on morphological target region detection, the automatic selection principle of the seed point constraint region is defined, an empty queue is established in the region growth process,based on the sequential traversal method, region growth starts from the seed region, the size of a target shape template prepared in advance is taken as the growth criterion reference, threshold adaptive adjustment of determination of the entire growth is made, a final growth result map is acquired, and the to-be-detected target is determined.

Description

technical field [0001] The present invention specifically relates to a method for object detection based on region growing. Background technique [0002] Infrared imaging terminal guidance is a core technology of aircraft imaging terminal guidance, which has an important influence on the visual navigation technology of the new generation of weapon systems. In the complex infrared background, the target is often embedded in the background component whose spatial distribution is not stable, and is easily interfered by background clutter and decoys. This brings great difficulties to infrared imaging guidance. In recent years, in the research of detecting the area containing the target in the image from various background clutter, many methods have been adopted, such as threshold segmentation method, edge segmentation method, gray histogram method, but in the complex background And the effect is not ideal in the case of low contrast; the mathematical morphology method has a pe...

AI Technical Summary

Problems solved by technology

This brings great difficulties to infrared imaging guidance

In recent years, in the research of detecting the area containing the target in the image from various background clutters, many methods have been adopted, such as threshold segmentation method, edge segmentation method, gray histogram method, but in the complex background And the effect is not ideal in the case of low contrast; the mathematical morphology method has a perfect mathematical foundation, is a nonlinear filtering method, and has been widely used in computer vision, image processing, image segmentation, medical image analysis and other fields