基于Harris算法的自适应双极性红外舰船目标检测  

Adaptive infrared bipolar warship target detection based on Harris algorithm

作　　者：徐倩[1] 陈咸志[1] 罗镇宝[1] 金代中[1] 胡俊杰[1] 曲锐[1] 

机构地区：[1]西南技术物理研究所,四川成都610041

出　　处：《激光与红外》2014年第12期1364-1368,共5页Laser & Infrared

基　　金：国家"863"计划重大项目(No.2012AAXXX2012X)资助

摘　　要：红外舰船目标检测是舰船目标自动识别与跟踪的关键技术之一。针对现有方法在舰船目标成像灰度呈现"双极性"(舰船亮度高于或低于背景)或多极性(舰船目标灰度分布不均匀)的情况下适应能力不足,提出了一种新的红外舰船目标检测方法。为抑制海空杂波干扰提高检测概率,对图像进行均值漂移滤波,利用行垂直方向梯度投影获得感兴趣区域,通过检测区域内目标角点实现舰船目标的自动检测。实验结果表明,该算法能够克服传统算法基于舰船亮度高于背景的假设,在舰船灰度呈现双极性或多级性的情况下获得较好的检测结果。Infrared warship target detection is one of the most important techniques in ship automatic recognition and tracking. Because adaptive capacity of the existing method is poor when warship target imaging gray appears bipolar（ the intensity of warship is higher or lower than background）,a new infrared warship target detection method is proposed. In order to suppress background interference,an image is processed by using mean shift filter. Compared to the traditional method,the ROI is obtained by vertical gradient projection. Finally,automatic ship target detection is realized by suppressing false corners and extracting warship corners. Experiment results show that the proposed algorithm can get good performance in bipolar target detection especially when the warship has different gray scale values.

关 键 词：红外图像 舰船目标检测 双极性目标 HARRIS算法 梯度特征计算 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]