红外线列探测器闪元噪声分析与抑制方法
Received:October 16, 2017  Revised:March 13, 2018  点此下载全文
引用本文:刘高睿,孙胜利,林长青,吕凭乐.红外线列探测器闪元噪声分析与抑制方法[J].Journal of Infrared and Millimeter Waves,2018,37(4):421~426
Hits: 198
Download times: 140
Author NameAffiliationE-mail
LIU Gao-rui Key Laboratory of Infrared Detecting Imaging Technology,Chinese Academy of Sciences
Shanghai Institute of Technical Physics,Chinese Academy of Sciences
University of Chinese Academy of Sciences 
1173142897@qq.com 
SUN Sheng-li Key Laboratory of Infrared Detecting Imaging Technology,Chinese Academy of Sciences
Shanghai Institute of Technical Physics,Chinese Academy of Sciences 
palm_sun@163.com 
LIN Chang-qing Key Laboratory of Infrared Detecting Imaging Technology,Chinese Academy of Sciences
Shanghai Institute of Technical Physics,Chinese Academy of Sciences 
 
Lv Ping-yue Key Laboratory of Infrared Detecting Imaging Technology,Chinese Academy of Sciences
Shanghai Institute of Technical Physics,Chinese Academy of Sciences
University of Chinese Academy of Sciences 
 
基金项目:中国科学院上海技术物理研究所创新专项(No.CX-63)
中文摘要:闪元是红外探测器固有特性,在红外线列探测器扫描图像中,闪元噪声以条状形式存在,对目标检测性能影响很大,多种检测方法在闪元位置产生大量虚警。本文分析了闪元现象在红外扫描图像中的特点,设计了基于时序多帧最大值投影的闪元检测与补偿方法。考虑到闪元仍然保留了一定的信息获取能力,采用了一维中值滤波的方法单独对闪元位置进行背景抑制,对闪元补偿后的图像进行二维目标检测。通过实验验证表明:经背景抑制后,闪元位置平均目标信杂比可以达到常规位置的96%,目标信息得到较好的保留。常规目标检测方法不再到受闪元噪声的干扰。
中文关键词:闪元噪声  红外线列探测器  矩匹配法  弱小目标
 
Analysis and Suppression method of flickering pixel noise in Images of infrared linear detector
Abstract:The flickering pixel noise is the inherent characteristic of the infrared detector. In the scanning image of the infrared linear detector, flickering pixel noise exists in the form of a strip. It has a great impact on the performance of target detection, and many detection methods produce a large number of false alarm in the location of flickering pixels. In this paper, the characteristics of the flickering pixel phenomenon in the infrared scanning image are analyzed, and the detection and compensation method of the flickering pixel based on the time series multi-frame maximum projection is designed. Considering that the flickering pixels still retain a part of the information acquisition ability, the one-dimensional median filter is used to suppress the background for the position where flickering pixels exist. Traditional methods of target detection are carried out for the image after flickering pixels compensated. Experiment results show that after background suppression, the target average signal to clutter ratio in the flickering pixel location can reach 96% of the normal position, and the target information is well preserved. The traditional method of target detection is no longer interfered by flickering noise.
keywords:flickering pixel noise  Infrared linear detector  Moment matching method  Low SNR small target
View Full Text  View/Add Comment  Download reader

Copyright:《Journal of Infrared And Millimeter Waves》