摘要
在光学对空探测领域,强噪声环境下的红外弱小目标检测既是关键问题,也是挑战难度很大的瓶颈问题。提出一种基于新形态学算法的弱小目标检测算法,通过形态学增强及非局部均值(NLM)算法,抑制背景杂波。最后通过自适应双阈值方法进行二值化处理,捕获并跟踪目标。在传统方法及问题的基础上进行改进,为充分验证提出算法的有效性,将提出的算法与近年来一些先进算法进行比对。测试手段采用受试者测试曲线,以及在不同图像上的检测指标进行处理。实验结果表明,与其他先进算法相比,提出的算法检测性能更好,检测率平均提高了11.5%。
In the field of optical air detection,infrared weak and small target detection in strong noise environment is not only the key problem,but also the bottleneck problem with great challenge.In this paper,a dim small target detection algorithm based on new morphological algorithm is proposed.The background clutter is suppressed by morphological enhancement and NLM algorithm.Finally,the adaptive double threshold method is used for binarization to capture and track the target.Based on the full introduction of traditional methods and problems,this paper describes the improved algorithm in detail.In order to fully verify the effectiveness of the proposed algorithm,the proposed algorithm is compared with some advanced algorithms in recent years.The test method adopts the test curve of subjects and the detection indexes on different images for processing.Experimental results show that compared with other advanced algorithms,the proposed algorithm has better detection performance,and the detection rate is increased by 11.5%on average.
作者
康利娟
陈先桥
Kang Lijuan;Chen Xianqiao(Information Engineering College,Zhengzhou Technology and Business University,Zhengzhou 450000,China;College of Computer Science and Technology,Wuhan University of Technology,Wuhan 430000,China)
出处
《国外电子测量技术》
北大核心
2022年第4期49-54,共6页
Foreign Electronic Measurement Technology
基金
国家重点研发计划基金(2018YFC0810400)
2022年度河南省高等学校重点科研项目(22B520039)
郑州工商学院科研创新项目(2021-KYZD-03)资助
关键词
红外弱小目标检测
增强形态学
NLM算法
自适应双阈值
infrared dim small target detection
enhanced morphology
NLM algorithm
adaptive double threshold
作者简介
康利娟,硕士,讲师,主要研究方向为图形图像处理。E-mail:klj15938766500@163.com;陈先桥,博士,教授,主要研究方向为图形图像处理,计算机仿真。E-mail:chenxianqiao1961@163.com
