Review of Dim Small Target Detection Research in Single Infrared Image
-
摘要: 远距离广视角场景中由于红外热成像仪成像原理的局限性、大气环境的干扰、远距离传输介质对红外辐射的衰减,检测目标面临巨大挑战。本文在详细分析了图像背景复杂、目标特性弱小、图像对比度低和结构特性缺失等红外弱小目标图像特性的基础上,从基于目标突显和背景预测两大类概述了单帧红外图像弱小目标检测技术的研究现状,并探讨了红外弱小目标检测研究的发展趋势。Abstract: For long-distance and wide field-of-view scenes, infrared target detection has significant challenges owing to the principle of a thermal imager, interference of the atmospheric environment, and attenuation of infrared radiation by long-distance transmission media. Based on the characteristic analysis of small-target infrared images, such as complex background, dim and small targets, low image contrast, and lack of image structures, we reviewed the research status of infrared dim small-target detection from target highlight and background estimation and discussed the development trend of infrared dim small-target detection.
-
0. 引言
在红外热成像设备的设计和应用中,经常针对不同厂家或者不同型号的红外探测器定制非标的红外图像处理算法,以及标定不同的算法参数。由于算法模型或者算法参数经常变化,导致设计算法、仿真、在硬件上调试算法的时间很多,开发周期很长,开发效率低下等问题。而常规的红外图像处理算法,包括数据采集、时域滤波、非均匀性校正、空域滤波、锐化增强、坏点校正以及调光映射等。设计的初期,需要从算法原理设计、算法模型的建立、数学过程的计算,转化成可实现的编程语言,然后实现和显示。
随着小型化、低功耗、低成本的红外热成像设备应用的普及和民用。FPGA处理器非常适合于该种应用需求,FPGA是高速并行的处理器,适合于图像阵列的算法处理和加速,并且功耗低,可反复编程设计,所以常应用于红外热成像设备中。
1. 传统的红外图像仿真方法
基于FPGA的红外图像处理算法,仿真流程示意图如图 1所示。
从图 1中可知,利用FPGA处理器进行图像处理,需要两步的仿真结果。
第一步:根据算法需求,设计算法原理,在Matlab软件上编程实现算法过程,先用浮点数进行设计和计算,代码设计和编写完成后,导入测试图像,进行仿真和显示,在显示窗口中显示出来[1-4]。对显示出来的图像效果进行评估,看看图像效果有没有达到算法原理期望的结果,如果没有,调整算法参数和实现过程,反复迭代调整直至图像效果达到期望的结果。达到期望结果后,进行定点化设计,将算法实现过程中的浮点数,全部转化成定点化数,比如INT8、INT16、INT32等,并且转化完成后,保证算法实现的结果仍然达到期望值,不损失精度,不丢失细节。然后再更换测试图像,测试数据设计成依次递增或者递减的特征数据,再进行仿真。
第二步:文献[5]、[6]提出定点化之后的算法原理和实现过程,编写FPGA的硬件描述语言,设计FPGA的逻辑,将第一步过程中的算法原理实现出来,再进行ModelSim仿真,仿真测试的输入激励是第一步中同样依次递增或者递减的特征数据,得到仿真结果,与第一步中每个关键节点(关键步骤)的结果进行比对,保证每一步的结果都与Matlab仿真的结果相同,才能确保算法效果的达到。
传统的红外图像仿真方法,分成上述两个步骤,先要进行Matlab仿真,显示评判效果达到之后,再启动定点化,将定点化之后的算法过程转化成FPGA中可以实现的过程,并用FPGA的硬件描述语言实现相关的逻辑,再进行ModelSim仿真,由于ModelSim不能像Matlab工具一样方便导入一张图像(图片数据)进行仿真,仿真结果不能直观地可视化显示。所以在第一步、第二步过程中,输入激励的时候制作一些递增、递减的相同特征数据进行仿真。对每一步关键步骤和最后结果的仿真比对,保证数据正确,才能保证算法原理的实现和算法效果的达到。
2. 基于ModelSim可视化的红外图像仿真方法
从上述传统的红外图像仿真过程看,既要实现Matlab代码编写、仿真、还要做定点化设计。等FPGA代码编写完成后,在ModelSim仿真的时候还要进行每一步过程的计算结果比对,并且不能直接可视化的方式看到算法运行的结果。存在实现过程复杂、开发效率低下,开发周期长等问题。针对这些问题,本文提出了一种新的基于ModelSim可视化仿真的红外图像仿真方法。如图 2所示为本文的实现框图。
从图 2中可知,基于ModelSim可视化的红外图像仿真方法相比传统的红外图像仿真方法,仿真过程简单很多,开发效率也会提高很多。在算法原理设计完成后,直接编写FPGA的硬件描述语言实现算法计算过程,再进行ModelSim仿真,仿真的时候,不再用特定的递增数据或者递减的特征数据做输入激励,而是用之前原始采集到的红外图像,如果算法原理是针对16 bit的原始红外图像处理,可以是Y16数据,针对8 bit的调光之后红外图像处理可以是Y8数据,之前采集到的红外图像Y16或者Y8数据,存放在RAW文件中。而RAW文件中的图像数据是可以用ImageJ查看的,ImageJ工具可以直接显示16位或者8位的RAW图像。
对于红外图像算法所用到的数据文件——RAW文件,是不能直接导入ModelSim进行仿真的,需要经过binary工具将raw文件转换成txt文件,在ModelSim中通过系统调用$readmemb/ readmemh函数读入TXT文件的方式读入到内存中,然后进行图像算法处理,算法处理完成,调用$writememb/ $writememh函数将算法处理之后的图像写入txt文件中,通过txt2bin工具,将txt格式的数据转换成bin文件,再通过直接改后缀名,将.bin改成.raw文件。这样输入输出都是raw格式的数据文件,导入ImageJ工具中显示,以可视化的方式显示出了图像。
所以,这样红外图像算法处理的输入是图像数据,输出是处理之后的图像数据,可以直观地在ImageJ工具中显示出来对比,分析红外图像算法的处理过程是否正确。并且每一步的计算过程和关键步骤的结果也可以通过如图 3内存查看的方式、图 4内存读写过程仿真的方式查看数据在内存中存储、写入和读出是否异常。并通过如图 5红外图像算法处理仿真过程来查看数据计算过程正确与否。
这样数据输入,写入内存,然后从内存中读出,再进行红外图像算法处理,每一步数据的处理都是可以查询和追溯的,所以既方便又快捷。如图 6所示为本文进行红外图像滤波降噪算法处理过程中的输入源图和处理完成之后输出结果对比图Image显示,图 7为本文中基于ModelSim仿真过程中输入输出文件结构示意图。
在反复调整算法过程,以及算法参数的情况下,能够很快得到想要的算法结果,比如图 6中右边是降噪之后的效果图,降噪之后,图像噪声能消除了,但是图像变得模糊了,说明降噪算法参数设置不合适,通过调整滤波器的系数和阈值之后,降噪效果有明显的改善,如图 8所示。
上述图 6与图 8中的红外图像滤波降噪算法为非局部平均滤波降噪算法,算法流程如图 9所示。这种降噪算法是一种利用图像自相似性的空域滤波算法,相比传统高斯平滑、双边滤波等方法,保护更多的图像细节。图像各点之间可以同时进行运算,非常适合于在FPGA并行处理器上运算。根据相似性计算权重的时候,如果相似性阈值选择太大,算法滤波出来的效果如图 6所示,过度平滑了,图像变模糊了,而选择合适的相似性阈值,得到滤波出来的效果如图 8所示,既能滤除噪声,又能保留图像细节。在该种红外图像仿真方法中能很快得到一组适合的算法参数。
基于该种ModelSim可视化的红外图像仿真方法,能够快速地修改算法参数和调整算法计算过程,运行之后得到图像算法的仿真结果,并且输入、算法计算、输出等环节里的每一步数据都是可以通过内存或者波形查看的方式追溯数据处理过程,保证数据向着期望的方向计算和处理。
3. 分析讨论
本文提出两种红外图像的仿真方法,传统的红外图像仿真方法,需要进行Matlab仿真,仿真结束后,再进行定点化设计,才能指导FPGA进行编程和实现,最后启动ModelSim的仿真,并且传统的ModelSim仿真并不能直接对图像数据的输入进行仿真并且可视化,而是制作相关的递增或者递减的特征数据作为输入,并且Matlab仿真同样以该特征数据作为输入,进行仿真,最后对2组仿真数据结果进行比对,比对结果一致代表该算法在FPGA中运行正常,能达到期望的算法效果。该过程复杂、繁琐、开发效率低下、开发周期长等问题不可忽视。所以本文在传统的红外图像仿真方法上进行了改进和优化,提出了一种新的基于ModelSim可视化的红外图像仿真方法与系统。该种方法在传统方法上省去了Matlab仿真的过程和环节,同时也省去了定点化的设计过程。在算法原理设计完成后,启动FPGA硬件描述语言的编写和仿真,只是在仿真的时候,输入图像可以用之前采集好的红外图像或者用PS工具生成一张图片,保存为.raw文件,在仿真输入的时候用binary工具将raw格式文件转换成txt文件,就可以启动仿真过程了,仿真结束后,生成的结果TXT,再通过txt2bin工具,将txt格式的数据转换成bin文件,再通过直接改后缀名的方式,将.bin改成.raw文件。用ImageJ工具打开source.raw和result.raw文件就可以直观地对比显示在图像窗口上了。并且对于算法参数、计算过程的调整,也是很快得到相应的输出结果。该过程相对传统的Matlab仿真和ModelSim联合仿真要方便很多,尤其是在算法调整和参数的调整的情况下,又要重复一遍Matlab仿真,再去ModelSim仿真的过程。但是该种方法对于红外图像算法原理的理解和FPGA逻辑实现图像算法的能力要求比较高。并且本文针对红外图像的仿真方法与传统的红外图像仿真方法,在适用范围上是一样的,对象都是红外图像的算法仿真。针对具体的算法模型,需要将算法分解成各计算步骤,然后针对各计算步骤进行数据仿真,所以对于其他文献[7]中不同的算法,比如弱小目标的背景的抑制,突出目标的增强算法等都是一样的仿真过程。
4. 结论
对于红外图像处理算法在FPGA的实现过程中,本文阐述了传统的仿真方法,先进行Matlab仿真,仿真效果达到的前提下,再进行定点化设计,指导FPGA进行逻辑设计,完成算法原理的实现,最后启动ModelSim仿真,而且仿真结果的正确评判标准是与Matlab仿真结果比对一致,不一致的话,要优化定点化过程,再进行仿真,重复直至结果一致。而本文的仿真方法基于ModelSim的可视化仿真,省去了上述繁琐的开发过程,直接将图像源文件转化后导入ModelSim仿真,算法处理完成后,转成图像结果文件进行比对,用可视化的方式评判图像经过算法处理之后是否达到算法原理期望达到的算法效果。同时如果算法未到达期望效果,通过调整参数、算法实现过程,以及追溯输入图像、写入、读出内存过程,判断图像算法计算过程中数据的正确与否。所以该方法与传统的红外图像仿真方法相比,评估算法效果以可视化显示输入、输出对比图差异性来达到。评估算法计算过程的正确性,同样可以追溯数据处理过程的每一步的正确性来达到。所以该仿真方法的仿真效果是与传统的方法一致的。但是开发过程缩减了和效率上大大提升了,开发周期短。尤其是在反复调参和优化过程中,显得更为明显和方便便捷。
-
[1] BAI X, ZHOU F. Analysis of new top-hat transformation and the application for infrared dim small target detection[J]. Pattern Recognition, 2010, 43(6): 2145-2156. DOI: 10.1016/j.patcog.2009.12.023
[2] Jakubowicz J, Lefebvre S, Maire F, et al. Detecting aircraft with a low-resolution infrared sensor[J]. IEEE Transactions on Image Processing, 2012, 21(6): 3034-3041. DOI: 10.1109/TIP.2012.2186307
[3] 凡遵林, 管乃洋, 王之元, 等. 红外图像质量的提升技术综述[J]. 红外技术, 2019, 41(10): 941-946. http://hwjs.nvir.cn/article/id/hwjs201910009 FAN Zunlin, GUAN Naiyang, WANG Zhiyuan, et al. Infrared image quality improvement technology: a review[J]. Infrared Technology, 2019, 41(10): 941-946. http://hwjs.nvir.cn/article/id/hwjs201910009
[4] 杨俊彦, 吴建东, 宋敏敏. 红外成像制导技术发展展望[J]. 红外, 2016, 37(8): 1-6. DOI: 10.3969/j.issn.1672-8785.2016.08.001 YANG Junyan, WU Jiandong, SONG Minmin. Development and prospect of infrared imaging guidance technology[J]. Infrared, 2016, 37(8): 1-6. DOI: 10.3969/j.issn.1672-8785.2016.08.001
[5] Ahmadi K, Salari E. Small dim object tracking using frequency and spatial domain information[J]. Pattern Recognition, 2016, 58(3): 227-234.
[6] JIA Y, GAO T, ZHAO J. Texture based infrared military target extraction[J]. Applied Mechanics & Materials, 2011, 44-47: 2489-2493.
[7] LAW W, XU Z, YONG K. Manganese-doped near-infrared emitting nanocrystals for in vivo biomedical imaging[J]. Optics Express, 2016, 24: 17553-17561. DOI: 10.1364/OE.24.017553
[8] 任向阳, 王杰, 马天磊, 等. 红外弱小目标检测技术综述[J]. 郑州大学学报: 理学版, 2020, 52(2): 1-21. https://www.cnki.com.cn/Article/CJFDTOTAL-ZZDZ202002001.htm REN Xiangyang, WANG Jie, MA Tianlei, et al. Review on infrared dim and small target detection technology[J]. Journal of Zhengzhou University: Natural Science Edition, 2020, 52(2): 1-21. https://www.cnki.com.cn/Article/CJFDTOTAL-ZZDZ202002001.htm
[9] ZHENG C, LI H. Small infrared target detection based on harmonic and sparse matrix decomposition[J]. Optical Engineering, 2013, 52(6): 066401. DOI: 10.1117/1.OE.52.6.066401
[10] ZHANG B, ZHANG T, CAO Z, et al. Fast new small-target detection algorithm based on a modified partial differential equation in infrared clutter[J]. Optical Engineering, 2007, 46(10): 106401. DOI: 10.1117/1.2799509
[11] BAI X, ZHOU F, XUE B. Infrared dim small target enhancement using toggle contrast operator[J]. Infrared Physics & Technology, 2012, 55: 177-182.
[12] Bae T, ZHANG F, Kweon I. Edge directional 2D LMS filter for infrared small target detection[J]. Infrared Physics & Technology, 2012, 55(1): 137-145.
[13] Karalı A, Okman O, Aytaç T. Adaptive enhancement of sea-surface targets in infrared images based on local frequency cues[J]. J. Opt. Soc. Am. A. , 2010, 27(3): 509-517. DOI: 10.1364/JOSAA.27.000509
[14] Angaitkar P, Saxena P. Enhancement of infrared image: a review[J]. International Journal of Engineering Research and Applications, 2012, 2(2): 1186-1189.
[15] Janani V, Dinakaran M. Infrared image enhancement techniques - a review[C]//IEEE International Conference on Current Trends in Engineering and Technology, 2014: 167-173.
[16] Morris N, Avidan S, Matusik W, et al. Statistics of infrared images[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2007: 1-7.
[17] Nanhakumar N, Aggarwal J. Integrated analysis of thermal and visual images for scene interpretation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1988, 10(4): 469-481. DOI: 10.1109/34.3911
[18] 穆为磊, 高建民, 陈富民, 等. 符合人眼视觉特性的焊缝射线数字图像增强方法[J]. 西安交通大学学报, 2012, 46(3): 90-93. DOI: 10.3969/j.issn.1008-245X.2012.03.014 MU Weilei, GAO Jiammin, CHEN Fumin, et al. Weld radiographic image enhancement conforming to human visual system [J]. Journal of Xi'an Jiaotong University, 2012, 46(3): 90-93. DOI: 10.3969/j.issn.1008-245X.2012.03.014
[19] 龚昌来, 罗聪, 杨冬涛, 等. 基于正弦灰度变换的红外图像增强算法[J]. 激光与红外, 2013, 43(2): 200-203. https://www.cnki.com.cn/Article/CJFDTOTAL-JGHW201302019.htm GONG Changlai, LUO Cong, YANG Dongtao, et al. Infrared image enhancement based on sine gray level transformation[J]. Laser & Infrared, 2013, 43(2): 200-203. https://www.cnki.com.cn/Article/CJFDTOTAL-JGHW201302019.htm
[20] 汪国有, 陈振学, 李乔亮. 复杂背景下红外弱小目标检测的算法研究综述[J]. 红外技术, 2006, 28(5): 287-292. DOI: 10.3969/j.issn.1001-8891.2006.05.010 WANG Gouyou, CHEN Zhenxue, LI Qiaoliang. A review of infrared weak and small targets detection under complicated background[J]. Infrared Technology, 2006, 28(5): 287-292. DOI: 10.3969/j.issn.1001-8891.2006.05.010
[21] Mahata S, Kar R, Mandal D. Optimal fractional-order highpass Butterworth magnitude characteristics realization using current-mode filter[J]. AEUE-International Journal of Electronics and Communications, 2019, 102: 78-89.
[22] YANG L, YANG J, YANG K. Adaptive detection for infrared small target under sea-sky complex background[J]. Electronics Letters, 2004, 40(17): 1083-1085. DOI: 10.1049/el:20045204
[23] DONG X, HUANG X, ZHENG Y. Infrared dim and small target detecting and tracking method inspired by human visual system[J]. Infrared Physics & Technology, 2014, 62: 100-109.
[24] HAN J, MA Y, HUANG J. An infrared small target detecting algorithm based on human visual system [J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(3): 452-456.
[25] ZHANG Y, ZHENG L, ZHANG Y. Small infrared target detection via a Mexican-hat distribution[J]. Applied Sciences, 2019, 5: 5570.
[26] 周姣, 辛云宏. 基于显著性与尺度空间的红外弱小目标检测[J]. 激光与红外, 2015, 45(4): 452-456. DOI: 10.3969/j.issn.1001-5078.2015.04.022 ZHOU Jiao, XIN Yunhong. Infrared dim small target detection based on saliency and scale-space[J]. Laser & Infrared, 2015, 45(4): 452-456. DOI: 10.3969/j.issn.1001-5078.2015.04.022
[27] Srivastava H, Khatterwani K, Upadhyay S. A certain family of fractional wavelet transformations[J]. Mathematical Methods in the Applied Sciences, 2019, 42(9): 3103-3122. DOI: 10.1002/mma.5570
[28] Boccignone G. Small target detection using wavelets[C]//IEEE Fourteenth International Conference on Pattern Recognition, 1998: 1776-1778.
[29] Davidson G, Griffiths H. Wavelet detection scheme for small target in sea clutter[J]. Electronics Letters, 2002, 38(19): 1128-1130. DOI: 10.1049/el:20020790
[30] Candes E. Ridgelets: Theory and Application[D]. Stanford: Stanford University, 1998.
[31] Starck J, Candes E, Donoho D. The curvelet transform for image denoising[J]. IEEE Transactions on Image Processing, 2002, 11(6): 131-141.
[32] Do M, Vetterli M. The Contourlet transform: an efficient directional multi-resolution image representation[J]. IEEE Transactions on Image Processing, 2005, 14(12): 2091-2106. DOI: 10.1109/TIP.2005.859376
[33] Easley G, Labate D, Lim W. Sparse directional image representation using the discrete shearlet transforms[J]. Applied and Computational Harmonic Analysis, 2008, 25: 25-46. DOI: 10.1016/j.acha.2007.09.003
[34] Lim W. Nonseparable shearlet transform[J]. IEEE Transactions on Image Processing, 2013, 22(5): 2056-2065. DOI: 10.1109/TIP.2013.2244223
[35] PENG L, ZHANG T, LIU Y, et al. Infrared dim target detection using shearlet's kurtosis maximization under non-uniform background[J]. Symmetry, 2019, 11(5): 723. DOI: 10.3390/sym11050723
[36] CHEN C, LI H, WEI Y. A local contrast method for small infrared target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(1): 574-581. DOI: 10.1109/TGRS.2013.2242477
[37] BAI X, BI Y. Derivative entropy-based contrast measure for infrared small-target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(4): 2452-2466. DOI: 10.1109/TGRS.2017.2781143
[38] LI S, LI C, YANG X, et al. Infrared dim target detection method inspired by human vision system[J]. Optik, 2020, 206: 164167. DOI: 10.1016/j.ijleo.2020.164167
[39] 危水根, 王程伟, 张聪炫, 等. 多信息融合的红外弱小目标检测[J]. 红外技术, 2019, 41(9): 857-865. http://hwjs.nvir.cn/article/id/hwjs201909010 WEI Shuigen, WANG Chengwei, ZHANG Congxuan, et al. Infrared dim target detection based on multi-information fusion[J]. Infrared Technology, 2019, 41(9): 857-865. http://hwjs.nvir.cn/article/id/hwjs201909010
[40] YI X, WANG B, ZHOU H, et al. Dim and small infrared target fast detection guided by visual saliency[J]. Infrared Physics & Technology, 2019, 97: 6-14.
[41] 黄苏琦. 时空谱多特征联合红外弱小目标检测方法研究[D]. 成都: 电子科技大学, 2020. HUANG Suqi. Infrared Dim Small Target Detection Based on Joint Temporal-Spatial-Spectral Features[D]. Chengdu: University of Electronic Science and Technology of China, 2020.
[42] Serra J. Image Analysis and Mathematical Morphology[M]. New York: Academic Press, 1982.
[43] BAI X, ZHOU F. Analysis of new top-hat transformation and the application for infrared dim small target detection[J]. Pattern Recognition, 2010, 43(6): 2145-2156. DOI: 10.1016/j.patcog.2009.12.023
[44] ZHU H, LIU S, DENG L. Infrared small target detection via low-rank tensor completion with top-hat regularization[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 58(2): 1004-1016.
[45] BAI X, ZHOU F. Analysis of new top-hat transformation and the application for infrared dim small target detection[J]. Pattern Recognition, 2010, 43(6): 2145-2156. DOI: 10.1016/j.patcog.2009.12.023
[46] 刘源, 汤心溢, 李争. 基于新Top Hat变换局部对比度的红外小目标检测[J]. 红外技术, 2015, 37(7): 544-552. http://hwjs.nvir.cn/article/id/hwjs201507002 LIU Yuan, TANG Xinyi, LI Zheng. A new top hat local contrast based algorithm for infrared small target detection[J]. Infrared Technology, 2015, 37(7): 544-552. http://hwjs.nvir.cn/article/id/hwjs201507002
[47] Ohki S. Two-dimensional LMS adaptive filters[J]. IEEE Trans. Consum Electr, 1991, 37(1): 66-73. DOI: 10.1109/30.73648
[48] LV P, SUN S, SUN C, et al. Space moving target detection and tracking method in complex background[J]. Infrared Physics & Technology, 2018, 91: 107-118.
[49] 张艺璇, 李玲, 辛云宏. 基于自适应双层TDLMS滤波的红外小目标检测[J]. 光子学报, 2019, 48(9): 186-198. https://www.cnki.com.cn/Article/CJFDTOTAL-GZXB201909023.htm ZHANG Yixuan, LI Ling, XIN Yunhong. Infrared small target detection based on adaptive double-layer TDLMS filter[J]. Acta Photonica Sinica, 2019, 48(9): 186-198. https://www.cnki.com.cn/Article/CJFDTOTAL-GZXB201909023.htm
[50] Bae T, Sohng K. Small target detection using bilateral filter based on edge component[J]. Journal of Infrared Millimeter and Terahertz Waves, 2010, 31(6): 735-743.
[51] ZHAO J, CHEN Y, FENG H, et al. Infrared image enhancement through saliency feature analysis based on multi-scale decomposition[J]. Infrared Physics & Technology, 2014, 62: 86-93.
[52] LU Y, HUANG S, ZHAO W. Sparse representation based infrared small target detection via an online learned double sparse background dictionary[J]. Infrared Physics & Technology, 2019, 99: 14-27.
[53] GAO C, MENG D, YANG Y, et al. Infrared patch-image model for small target detection in a single image[J]. IEEE Transactions on Image Processing, 2013, 22(12): 4996-5009. DOI: 10.1109/TIP.2013.2281420
[54] WANG X, PEN Z, KONG D, et al. Infrared dim target detection based on total variation regularization and principal component pursuit[J]. Image and Vision Computing, 2017, 63: 1-9. DOI: 10.1016/j.imavis.2017.04.002g/guestquery?queryType=xml&restype=unixref&xml=|Metaphilosophy||38|2-3|264|2007|||
[55] ZHANG L, PENG L, ZHANG T, et al. Infrared small target detection via non-convex rank approximation minimization Joint/2,1 norm[J]. Remote Sensing, 2018, 10(11): 1821. DOI: 10.3390/rs10111821
[56] DAI Y, WU Y, SONG Y. Infrared small target and background separation via column-wise weighted robust principal component analysis[J]. Infrared Physics and Technology, 2018, 77: 421-430.
[57] 熊斌, 黄心汉, 王敏. 基于自适应目标图像恢复的红外弱小目标检测[J]. 华中科技大学学报: 自然科学版, 2017, 45(10): 25-30. https://www.cnki.com.cn/Article/CJFDTOTAL-HZLG201710005.htm XIONG Bin, HUANG Xinhan, WANG Min. Infrared small target detection based on adaptive double-layer TDLMS filter[J]. Journal of Huazhong University of Science and Technology: Natural Science Edition, 2017, 45(10): 25-30. https://www.cnki.com.cn/Article/CJFDTOTAL-HZLG201710005.htm
[58] Bouwmans T, Javed S, Sultana M, et al. Deep neural network concepts for background subtraction: A systematic review and comparative evaluation [J]. Neural Networks, 2019, 117: 8-66. DOI: 10.1016/j.neunet.2019.04.024
[59] Sheri A, Rafique M, Jeon M, et al. Background subtraction using Gaussian-Bernoulli restricted Boltzmann machine[J]. IET Image Processing, 2018, 12(9): 1646-1654. DOI: 10.1049/iet-ipr.2017.1055
[60] Farnoosh A, Rezaei B, Ostadabbas S. DeepPBM: Deep probabilistic background model estimation from video sequences [J/OL]. Computer Vision and Pattern Recognition, 2019, https://arxiv.org/abs/1902. 00820v1.
[61] Akilan T, WU J. sEnDec: An improved image to image CNN for foreground localization[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(10): 4435-4443.
[62] ZHENG W, WANG K, WANG F. A novel background subtraction algorithm based on parallel vision and Bayesian GANs[J]. Neurocomputing, 2020, 394: 178-200. DOI: 10.1016/j.neucom.2019.04.088
[63] Lim L, Keles H. Foreground segmentation using convolutional neural networks for multiscale feature encoding [J]. Pattern Recognition Letters, 2018, 112: 256-262. DOI: 10.1016/j.patrec.2018.08.002
[64] FAN Z, BI D, XIONG L, et al. Dim infrared image enhancement based on convolutional neural network[J]. Neurocomputing, 2018, 272: 396-404. DOI: 10.1016/j.neucom.2017.07.017
[65] ZHANG T, CAO S, PU T, et al. AGPCNet: attention-guided pyramid context networks for infrared small target detection[J/OL]. Computer Vision and Pattern Recognition, 2021, https://arxiv.org/abs/2111.03580.
[66] WANG H, ZHOU L, WANG L. Miss detection vs. false alarm: Adversarial learning for small object segmentation in infrared images[C]//IEEE/CVF International Conference on Computer Vision (ICCV), 2019: 8508-8517.
[67] DAI Y, WU Y, ZHOU F, et al. Asymmetric contextual modulation for infrared small target detection[C]//IEEE Winter Conference on Applications of Computer Vision (WACV), 2021: 949-958.
[68] LI B, XIAO C, WANG L, et al. Dense nested attention network for infrared small target detection[J/OL]. IEEE Transactions on Image Processing, 2022, Doi: 10.1109/TIP.2022.3199107.
[69] WANG K, DU S, LIU C, et al. Interior attention-aware network for infrared small target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-13.
[70] YAN P, HOU R, DUAN X, et al. STDMANet: spatio-temporal differential multiscale attention network for small moving infrared target detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5602516.
[71] HU Y, MA Y, PAN Z, et al. Infrared dim and small target detection from complex scenes via multi-frame spatial–temporal patchtensor model[J]. Remote Sensing, 2022, 14(9): 2234. DOI: 10.3390/rs14092234
[72] WU T, LI B, LUO Y, et al. MTU-Net: multilevel transUNet for space-based infrared tiny ship detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5601015.
[73] SONG Z, HUI B. A dataset for infrared image dim-small aircraft target detection and tracking underground/air background[DB/OL]//SCI Data Bank, [2019-10-28]. https://www.scidb.cn/en/detail?dataSetId=720626420933459968&dataSetType=journal.
[74] SUN X, GUO L, ZHANG W, et al. A dataset for small infrared moving target detection under clutter background [DB/OL]. Science Data Bank, [2023-10-24]. https://cstr.cn/31253.11.sciencedb.j00001.00231.