摘要
空域远距离红外目标探测系统中,飞行目标多表现为点状或面状的小目标,像素数少,且常伴有低空地面物体的干扰。根据空域和地面在梯度变化上的不同和目标本身的特性,提出了一种基于地面背景抑制和特征点检测的红外空中目标检测算法。分析了地面和空域在梯度变化上的特点,根据梯度变化大的像素的整体统计信息划分了空域和地面在图像中的分布,再通过特征点检测实现了候选红外飞行目标的检测。该算法适用于纯空域和低空背景,经过对实际采集的大量红外图像的仿真表明,本文提出的算法具有很强的实用性和鲁棒性。
In infrared remote target detection system,the aeral targets containing little piexls looks just like a spot.Objects on the ground have much influence on the detection of the targets.A infrared aeral target detection algorithm based on background suppression and feature points detection is presented.This method analyses the difference of gradient between ground and air and then partition them.Subsequently finding out the feature points which represent the location of the targets.This method is applicable for the spatial domain and low altitude target detection.Experiment results shows that the proposed algorithm is effective and convergent.
出处
《激光与红外》
CAS
CSCD
北大核心
2013年第4期457-460,共4页
Laser & Infrared
关键词
红外目标检测
背景抑制
特征点检测
图像梯度
infrared target detection
background suppression
feature points detection
gradient calculation
