基于局部直方图的红外目标分割算法

摘要

In infrared remote target detection system,the sky-background changes slowly,but the target shows the par-tial corner. It has a broad gray range and is brighter than the background. According to these characteristics,a target segmentation algorithm based on partial histogram is presented. The distribution characteristics of multi-target histo-gram are analyzed. According to gray level distribution rule and the image pixels restriction,the image segmentation is realized. Experiment results show that this method can segment flying targets rapidly and effectively,and it has strong practicability.%在远程红外探测系统中，背景为缓慢变化的天空，而目标则表现为局部奇异点。目标灰度分布范围大，局部较亮，边缘与背景对比度低。根据这一特性，提出了一种基于局部直方图的目标分割算法。文中分析了多个目标的直方图分布特性，根据其灰度分布规律和像素个数判决条件实现了目标的有效分割。该算法适用于空域背景下的飞行目标分割。经过仿真验证表明，本文所提出的算法能快速有效地分割出红外飞行目标，有很强的实用性。