基于CFAR-DCRF红外遥感舰船单帧目标检测方法
投稿时间:2019-01-10  修订日期:2019-03-08  点此下载全文
引用本文:
摘要点击次数: 46
全文下载次数: 0
作者单位E-mail
宋文韬 中国科学院上海技术物理研究所 songwentao@whu.edu.cn 
胡勇 中国科学院上海技术物理研究所 huyong@mail.sitp.ac.cn 
匡定波 中国科学院上海技术物理研究所  
巩彩兰 中国科学院上海技术物理研究所  
张文奇 中国科学院上海技术物理研究所  
黄硕 中国科学院上海技术物理研究所  
基金项目:国家重点研发计划(2017YFC0602103)
中文摘要:红外舰船目标的检测与精确提取在军事和民用领域有着重要的价值,由于现阶段天基红外高分辨率舰船数据的缺乏,红外舰船目标检测只能在中小尺度下开展研究。针对红外舰船小目标图像复杂背景弱信号,虚警率较高且难以被精确检测的问题,提出了一种CFAR(Constant False-Alarm Rate)-DCRF(Dense Conditional Random Fields)舰船目标检测算法。该算法针对小目标与虚警信号变化特征相似但结构特征不同的特点,利用CRF的多维上下文(空间、辐射)表达的优势,实现虚警特征抑制,并引入CFAR对模型进行改进,提高了DCRF对于弱信号目标的检出能力,实现舰船小目标的精确检测与分割。实验结果表明:该算法能够充分利用海域的全局上下文信息,能够在保持较高检出率同时,有效降低虚警率,实现单帧端到端的小目标检测。
中文关键词:遥感;全连接条件随机场  红外小目标;恒虚警率
 
Detection of ship targets based on CFAR-DCRF in single infrared remote sensing images
Abstract:The detection of infrared ship targets is of great important value in the military and civilian fields. However, due to the lack of space-based infrared high-resolution ship data, there are only few studies on space-based in small and medium scale. This paper focuses on the problem of low detection accuracy and low pixel image extraction accuracy of traditional small target detection and ship detection methods. An improved target detection algorithm based on CFAR(Constant False-Alarm Rate)-DCRF(Dense Conditional Fandom Fields)is proposed. The algorithm is based on the characteristics of small target and false alarm signal changes but different structural features. It uses the advantages of CRF multi-dimensional context (space, radiation) to achieve false alarm feature suppression, and introduces CFAR to improve the model and improve DCRF. Based on this model, experiments were performed under different conditions. The analysis results show that the algorithm can make full use of the global context information of the sea area, and can reduce the false alarm rate while maintaining a high detection rate.
keywords:Remote sensing  Dense conditional random fields  infrared dim target  CFAR
  查看/发表评论  下载PDF阅读器

版权所有:《红外与毫米波学报》编辑部