复杂背景下的红外弱小目标的检测
|
摘要
作为红外搜索与跟踪领域很重要的一个课题,在复杂背景的红外图像中对小目标检测的一系列瓶颈问题和关键技术受到了国内外专家学者的高度重视。本文研究的即是这个领域的两项关键技术：背景抑制技术和基于多帧的弱小目标检测。
本文首先提出两种改进的滤波算法。双边滤波,去噪同时也能较好地保留边缘,结合这一优点,本文对其滤波算子添加一个能保护小目标的模板；规整化的各向异性自适应滤波,对不同的结构具有自适应性,结合Robinson滤波的判据条件,二者互补。分别利用这两种算法针对不同的场景进行仿真,发现都有很好的鲁棒性。且与传统算法相比有较低的虚警率、较高的信杂比增益和背景抑制因子。
在图像分割阶段,提出一个基于平均方差加权熵的自适应阈值,利用大量实验数据讨论参数的选取来得到最好的效果。将滤波后的图与原图像的特征相结合,与其他分割算法相比,能更好地去除虚假目标。
接下来,结合Kalman滤波来对帧与帧之间的数据关联进行维持与更新,检测过程中,利用航迹评价因子进一步剔除虚假航迹,降低计算复杂程度,并最终得到真实目标的轨迹。最后给出整个算法流程的实验证明,发现整个算法确实能正确有效的检测出红外弱小目标的正确轨迹。
As an important topic in infrared(IR) search and tracking, a series of bottleneck problems and key techniques of IR small target detection in complex background are highly valued by the experts at home and abroad. This dissertation concentrates just in the two key techniques:the suppression of background clutter and the detection of the dim small moving targets in multi-frames.
At first, this paper proposed two improved filter algorithm. Bilateral filtering, de-noising and effectively keeping the edge, is added a template which can protect small target with its filter operator; Regularizing filtering(RegAF) is adaptive to different structures.A integrated and complementary of RegAF and Robinson filtering is proposed by this dissertation. According to different scene, the simulations both show good robustness. And compared with the traditional methods the two have lower false alarm rate, higher gain of Signal-to-Clutter Ratio and Background Suppression Factor(BSF).
In the image segmentation stage, an adaptive threshold value based on mean variance weighted information entropy is proposed with its parameters obtained by a lot of experimental data. With combination of characteristics of original image and filtered image, it is better able to remove the false target compared other segmentation algorithms.
Next, maintenance and update the data connection of sequential images via kalman filter. During detection, track evaluation factors further eliminate false track, reduce the complexity of the calculation, and finally the real goal of the track is obtained.
The results of whole algorithm are given in the end, which show that the proposed algorithm can detect right track of infrared dim-small target validly.
本文首先提出两种改进的滤波算法。双边滤波,去噪同时也能较好地保留边缘,结合这一优点,本文对其滤波算子添加一个能保护小目标的模板；规整化的各向异性自适应滤波,对不同的结构具有自适应性,结合Robinson滤波的判据条件,二者互补。分别利用这两种算法针对不同的场景进行仿真,发现都有很好的鲁棒性。且与传统算法相比有较低的虚警率、较高的信杂比增益和背景抑制因子。
在图像分割阶段,提出一个基于平均方差加权熵的自适应阈值,利用大量实验数据讨论参数的选取来得到最好的效果。将滤波后的图与原图像的特征相结合,与其他分割算法相比,能更好地去除虚假目标。
接下来,结合Kalman滤波来对帧与帧之间的数据关联进行维持与更新,检测过程中,利用航迹评价因子进一步剔除虚假航迹,降低计算复杂程度,并最终得到真实目标的轨迹。最后给出整个算法流程的实验证明,发现整个算法确实能正确有效的检测出红外弱小目标的正确轨迹。
As an important topic in infrared(IR) search and tracking, a series of bottleneck problems and key techniques of IR small target detection in complex background are highly valued by the experts at home and abroad. This dissertation concentrates just in the two key techniques:the suppression of background clutter and the detection of the dim small moving targets in multi-frames.
At first, this paper proposed two improved filter algorithm. Bilateral filtering, de-noising and effectively keeping the edge, is added a template which can protect small target with its filter operator; Regularizing filtering(RegAF) is adaptive to different structures.A integrated and complementary of RegAF and Robinson filtering is proposed by this dissertation. According to different scene, the simulations both show good robustness. And compared with the traditional methods the two have lower false alarm rate, higher gain of Signal-to-Clutter Ratio and Background Suppression Factor(BSF).
In the image segmentation stage, an adaptive threshold value based on mean variance weighted information entropy is proposed with its parameters obtained by a lot of experimental data. With combination of characteristics of original image and filtered image, it is better able to remove the false target compared other segmentation algorithms.
Next, maintenance and update the data connection of sequential images via kalman filter. During detection, track evaluation factors further eliminate false track, reduce the complexity of the calculation, and finally the real goal of the track is obtained.
The results of whole algorithm are given in the end, which show that the proposed algorithm can detect right track of infrared dim-small target validly.
引文
[1]E.T. Lim, S.D. Deshpande, C.W. Chan, R. Venkateswarlu, Adaptive spatial filtering techniques for the detection of targets in infrared imaging seekers, In:Proceedings of SPIE, vol.4025,2000, pp.194-202.
[2]A.P. Tzannes, D.H. Brooks. Detection of point targets in image sequences by hypothesis testing:a temporal test first approach, In:Proceedings of ICASSP,1999.
[3]武斌,姬红兵,李鹏,基于三阶累积量的红外弱小运动目标检测新方法.红外与毫米波学报,2006,25(5)：364-367.
[4]L. Yang, J Yang, K Yang. Adaptive detection for infrared small target under sea-sky complex background. ELECTRONICS LETTERS,2004,40(17):1083-1085.
[5]Boaz Porat, Benjamin Friedlander. A Frequency Domain Algorithm for Multi-Frame Detection and Estimation of Dim Targets. IEEE Trans On Pattern Analysis and Machine Intelligence,1990,12(4):398-401.
[6]M.S. Longmire, E.H. Takken. Lms and matched digital filters for optical clutter suppression, Applied Optics,1988,27(6):1141-1159.
[7]J.Y. Chen, I.S. Reed. A detection algorithm for optical targets in clutter, IEEE Transactions on Aerospace and Electronic Systems,1987,23(1):46-59.
[8]Manuel Fernandez, Tom Aridgides. A complex-domain adaptive order statistic filter and its application to signal detection in non-Gaussian noise and clutter. Signal and Data Processing of Small Targets 2010, Proc. of SPIE Vol.7698.
[9]Zhenxue Chen, Guoyou Wang, Jianguo Liu, Chengyun Liu. Small Target Detection Algorithm Based on Average Absolute Difference Maximum and Background Forecast. Int J Infrared Milli Waves,2007,28:87-97.
[10]Suyog D. Deshpande, M. H. Er, V. Ronda, et al. Max-Mean and Max-Median filters for detection of small targets//Denver, Colorado; SPIE,1999,3809:74-83.
[11]Ffrench P A, Zeidler J H, Ku W H. Enhanced detectability of small objects in correlated clutter using an improved 2-D adaptive lattice algorithm. IEEE Transactions on Image Processing,1997,6(3):383-397.
[12]P Wang, J W Tian, Ch Q Gao. Infrared Small Target Detection Using Directional Highpass Filters based on LS-SVM. Electronics Letters,2009,45(3):156-158.
[13]聂洪山,沈振康.一种基于Winer滤波的红外背景抑制方法.国防科技大学学报,2003,25(3)：51-54.
[14]李杰,马强.一种新的红外图像背景滤除算法研究.航空兵器,2006,1：29-31.
[15]RONDA V, NEW W L, TAN M H, et al.. Adaptive threshold-based spatio-temporal filtering techniques for detection of small targets. IEEE Transactions on Aerospace and Electronic Systems,2001,37(3):832-848.
[16]SUN Yu-qiu, ZHENG Sheng, TIAN Jin-wen, et al.. Infrared image dim small target detection based-on double energy accumulation. SPIE,2005,5985:3S-1-3S-5.
[17]Reed I S, Gagliardi R M, and Shao H. M. Application of three-dimensional filtering to moving target detection. IEEE Trans.Aeros.Electron. Sys.,1983,19(6):898-905.
[18]Reed I S, Gagliardi R M, Stotts L. Optical Moving Target Detection with 3-D Match Filtering. IEEE Trans. onAeros. Electron. Sys.,1998,24(4):327-336.
[19]Barniv Y. Dynamic Programming Solution for Detecting Dim Moving Targets. IEEE Trans onAES,1985,21(1):144-156.
[20]Blostein S D, Huang T S. Detecting Small Moving Objects in Image Sequences Using Sequential Hypothesis Testing. IEEE Trans on Signal Processing,1991,39(7):1611-1629.
[21]Liou R, Azimi Sadjadi M R. Detection of Dim Target in High Cluttered Background Using High Order Correlation Method. IEEE Trans on Aerospace and Electronic System, 1993,28(3):841-856.
[22]Roth M W. Neural Network for Extraction of Weak Target in High Clutter Environments. IEEE Trans. on Sys. Man. Cyber.,1989,19(5):1210-1217.
[23]D J C Andrew, et al. Inexact Graph Matching Using Genetic Search. Pattern Recognition,1997,30(7):953-970.
[24]姚连兴,仇维礼,王福恒.目标和环境的光学特性.北京：宇航出版社.1995.
[25]谭晓宇.复杂环境下弱小目标检测与识别技术研究.哈尔滨：哈尔滨工程大学博士学位论文.2006.
[26]胡永生.复杂背景中红外弱小目标探测方法研究.南京：南京理工大学博士学位论文.2008.
[27]Eisner Leonard et al. Spectral Radiation of Sky and terrain at Wavelengths between 1 and 20u. II. Terrain Measurements. Journal of the Optical Society of America.1962, 52:201-209.
[28]韩客松.复杂背景下红外点目标检测的预处理.系统工程与电子技术,2000,22(1)：52-54.
[29]Barnett J. Statistical analysis of median subtraction filtering with application to point target detection in inrfared background. Proc. SPIE.1989,1050:10-18.
[30]黄新栋,张涌.多波段线扫红外图像的实时预处理系统.激光与红外,2008,38(8)： 818-821.
[31]冈萨雷斯等.数字图像处理.第2版.北京：电子工业出版社.2005.
[32]杨磊.复杂背景条件下的红外弱小目标检测与跟踪算法研究.武汉：华中科技大学博士论文,2006.
[33]姜丹.信息论与编码.中国科学技术大学出版社,2001.
[34]Tomasi C, Manduchi R. Bilateral filtering for gray and color images//Proc Sixth Inter Conf Computer Vision.1998,839-846.
[35]秦翰林,周慧鑫,刘上乾,等.基于双边滤波的弱小目标背景抑制.强激光与粒子束,2009,21(1)：25-28.
[36]J. Immerkaer. Fast Noise Variance Estimation. Computer Vision and Image Understanding,1996,64(2):300-302.
[37]张必银,张天序,桑农,等.红外弱小运动目标实时检测的规整化滤波方法.红外与毫米波学报,2008,27(2)：95-100.
[38]Perona P, Malik J. Scale-space and edge detection using anisotropic diffusion. IEEE Trans. Pattern Anal. Mach. Intell.,1990,12(7):629-639.
[39]钱维贤.复杂背景下红外小目标跟踪若干关键技术研究.南京：南京理工大学博士学位论文,2010.
[40]许彬,郑链,王永学,宋承天.红外序列图像小目标检测与跟踪技术综述.红外与激光工程,2004,33(5)：482-487.
[41]Ostu N A. Threshold Selection Method from Gray-Level Histograms. IEEE Trans, on System, Man, and Cybernetics.1979,9(1):62-66.
[42]何斌,马天予等.Visual C++数字图像处理.第二版.北京：人民邮电出版社,2002：459～471.
[43]李俊山,杨威,张雄美.红外图像处理、分析与融合.北京：科技出版社,2009.
[44]Cassasent D. Sub-pixel target detection and tracking. Proc of the SPIE.1986,726. 1206-1210.
[45]Iranni M, Rousso B, Peleg S. Detection and tracking multiple moving objects using temporal integration. Second European Conference on Computer Vision.1992.282-287.
[46]Horn B K, Schunck B G. The image flow constraint equation. CVGIP,1986,35(3): 26231.
[47]Wang Yang, Zhen Qinbo, Zhang Junping. Real-time detection of small target in IR grey image based on mathematical morphology. Infrared and Laser Engineering,2003, 32(1):28-31.
[48]Ye Bin, Peng Jiaxiong.Small target detection based on energy accumulation and order morphology filtering in infrared image. Journal of Image and Graphics,2002,7(A) (3): 251-255.
[49]Ye Bin, Peng Jiaxiong. Moving small target detection based on order morphology filtering in infrared image sequences. Journal of Data Acquisition &, Processing,2001, 16(3):315-319.
[50]Li Guokuan, Peng Jiaxiong, Li Hong. Small target detection based on multi-wavelet transform. J Huazhong Univ of Sci &, Tech,2000,28(1):72-75.
[51]Li Guokuan, Peng Jiaxiong, Li Hong. Infrared imaging dim target detection based on wavelet transform. J Huazhong Univ of Sci&, Tech,2000,28(5):69-71.
[52]Li Hong, Zheng Chengyong, Gao Jingli. Detecting small moving object based on wavelet multi-scale image registration. Infrared and Laser Engineering,2003,32(3): 268-270.
[53]Bruton L. T and Bartley N. The Enhancement and Tracking of Moving Object in Digital Images Using Adaptive 3D Rercursive Filters. IEEE Trans,1986,33(6).
[54]B. Porat and B. Friendlander. A Frequency Domain Approch to MultiFrame Detection and Estimation of Dim Targets. Proc. On ASSP,1987.
[2]A.P. Tzannes, D.H. Brooks. Detection of point targets in image sequences by hypothesis testing:a temporal test first approach, In:Proceedings of ICASSP,1999.
[3]武斌,姬红兵,李鹏,基于三阶累积量的红外弱小运动目标检测新方法.红外与毫米波学报,2006,25(5)：364-367.
[4]L. Yang, J Yang, K Yang. Adaptive detection for infrared small target under sea-sky complex background. ELECTRONICS LETTERS,2004,40(17):1083-1085.
[5]Boaz Porat, Benjamin Friedlander. A Frequency Domain Algorithm for Multi-Frame Detection and Estimation of Dim Targets. IEEE Trans On Pattern Analysis and Machine Intelligence,1990,12(4):398-401.
[6]M.S. Longmire, E.H. Takken. Lms and matched digital filters for optical clutter suppression, Applied Optics,1988,27(6):1141-1159.
[7]J.Y. Chen, I.S. Reed. A detection algorithm for optical targets in clutter, IEEE Transactions on Aerospace and Electronic Systems,1987,23(1):46-59.
[8]Manuel Fernandez, Tom Aridgides. A complex-domain adaptive order statistic filter and its application to signal detection in non-Gaussian noise and clutter. Signal and Data Processing of Small Targets 2010, Proc. of SPIE Vol.7698.
[9]Zhenxue Chen, Guoyou Wang, Jianguo Liu, Chengyun Liu. Small Target Detection Algorithm Based on Average Absolute Difference Maximum and Background Forecast. Int J Infrared Milli Waves,2007,28:87-97.
[10]Suyog D. Deshpande, M. H. Er, V. Ronda, et al. Max-Mean and Max-Median filters for detection of small targets//Denver, Colorado; SPIE,1999,3809:74-83.
[11]Ffrench P A, Zeidler J H, Ku W H. Enhanced detectability of small objects in correlated clutter using an improved 2-D adaptive lattice algorithm. IEEE Transactions on Image Processing,1997,6(3):383-397.
[12]P Wang, J W Tian, Ch Q Gao. Infrared Small Target Detection Using Directional Highpass Filters based on LS-SVM. Electronics Letters,2009,45(3):156-158.
[13]聂洪山,沈振康.一种基于Winer滤波的红外背景抑制方法.国防科技大学学报,2003,25(3)：51-54.
[14]李杰,马强.一种新的红外图像背景滤除算法研究.航空兵器,2006,1：29-31.
[15]RONDA V, NEW W L, TAN M H, et al.. Adaptive threshold-based spatio-temporal filtering techniques for detection of small targets. IEEE Transactions on Aerospace and Electronic Systems,2001,37(3):832-848.
[16]SUN Yu-qiu, ZHENG Sheng, TIAN Jin-wen, et al.. Infrared image dim small target detection based-on double energy accumulation. SPIE,2005,5985:3S-1-3S-5.
[17]Reed I S, Gagliardi R M, and Shao H. M. Application of three-dimensional filtering to moving target detection. IEEE Trans.Aeros.Electron. Sys.,1983,19(6):898-905.
[18]Reed I S, Gagliardi R M, Stotts L. Optical Moving Target Detection with 3-D Match Filtering. IEEE Trans. onAeros. Electron. Sys.,1998,24(4):327-336.
[19]Barniv Y. Dynamic Programming Solution for Detecting Dim Moving Targets. IEEE Trans onAES,1985,21(1):144-156.
[20]Blostein S D, Huang T S. Detecting Small Moving Objects in Image Sequences Using Sequential Hypothesis Testing. IEEE Trans on Signal Processing,1991,39(7):1611-1629.
[21]Liou R, Azimi Sadjadi M R. Detection of Dim Target in High Cluttered Background Using High Order Correlation Method. IEEE Trans on Aerospace and Electronic System, 1993,28(3):841-856.
[22]Roth M W. Neural Network for Extraction of Weak Target in High Clutter Environments. IEEE Trans. on Sys. Man. Cyber.,1989,19(5):1210-1217.
[23]D J C Andrew, et al. Inexact Graph Matching Using Genetic Search. Pattern Recognition,1997,30(7):953-970.
[24]姚连兴,仇维礼,王福恒.目标和环境的光学特性.北京：宇航出版社.1995.
[25]谭晓宇.复杂环境下弱小目标检测与识别技术研究.哈尔滨：哈尔滨工程大学博士学位论文.2006.
[26]胡永生.复杂背景中红外弱小目标探测方法研究.南京：南京理工大学博士学位论文.2008.
[27]Eisner Leonard et al. Spectral Radiation of Sky and terrain at Wavelengths between 1 and 20u. II. Terrain Measurements. Journal of the Optical Society of America.1962, 52:201-209.
[28]韩客松.复杂背景下红外点目标检测的预处理.系统工程与电子技术,2000,22(1)：52-54.
[29]Barnett J. Statistical analysis of median subtraction filtering with application to point target detection in inrfared background. Proc. SPIE.1989,1050:10-18.
[30]黄新栋,张涌.多波段线扫红外图像的实时预处理系统.激光与红外,2008,38(8)： 818-821.
[31]冈萨雷斯等.数字图像处理.第2版.北京：电子工业出版社.2005.
[32]杨磊.复杂背景条件下的红外弱小目标检测与跟踪算法研究.武汉：华中科技大学博士论文,2006.
[33]姜丹.信息论与编码.中国科学技术大学出版社,2001.
[34]Tomasi C, Manduchi R. Bilateral filtering for gray and color images//Proc Sixth Inter Conf Computer Vision.1998,839-846.
[35]秦翰林,周慧鑫,刘上乾,等.基于双边滤波的弱小目标背景抑制.强激光与粒子束,2009,21(1)：25-28.
[36]J. Immerkaer. Fast Noise Variance Estimation. Computer Vision and Image Understanding,1996,64(2):300-302.
[37]张必银,张天序,桑农,等.红外弱小运动目标实时检测的规整化滤波方法.红外与毫米波学报,2008,27(2)：95-100.
[38]Perona P, Malik J. Scale-space and edge detection using anisotropic diffusion. IEEE Trans. Pattern Anal. Mach. Intell.,1990,12(7):629-639.
[39]钱维贤.复杂背景下红外小目标跟踪若干关键技术研究.南京：南京理工大学博士学位论文,2010.
[40]许彬,郑链,王永学,宋承天.红外序列图像小目标检测与跟踪技术综述.红外与激光工程,2004,33(5)：482-487.
[41]Ostu N A. Threshold Selection Method from Gray-Level Histograms. IEEE Trans, on System, Man, and Cybernetics.1979,9(1):62-66.
[42]何斌,马天予等.Visual C++数字图像处理.第二版.北京：人民邮电出版社,2002：459～471.
[43]李俊山,杨威,张雄美.红外图像处理、分析与融合.北京：科技出版社,2009.
[44]Cassasent D. Sub-pixel target detection and tracking. Proc of the SPIE.1986,726. 1206-1210.
[45]Iranni M, Rousso B, Peleg S. Detection and tracking multiple moving objects using temporal integration. Second European Conference on Computer Vision.1992.282-287.
[46]Horn B K, Schunck B G. The image flow constraint equation. CVGIP,1986,35(3): 26231.
[47]Wang Yang, Zhen Qinbo, Zhang Junping. Real-time detection of small target in IR grey image based on mathematical morphology. Infrared and Laser Engineering,2003, 32(1):28-31.
[48]Ye Bin, Peng Jiaxiong.Small target detection based on energy accumulation and order morphology filtering in infrared image. Journal of Image and Graphics,2002,7(A) (3): 251-255.
[49]Ye Bin, Peng Jiaxiong. Moving small target detection based on order morphology filtering in infrared image sequences. Journal of Data Acquisition &, Processing,2001, 16(3):315-319.
[50]Li Guokuan, Peng Jiaxiong, Li Hong. Small target detection based on multi-wavelet transform. J Huazhong Univ of Sci &, Tech,2000,28(1):72-75.
[51]Li Guokuan, Peng Jiaxiong, Li Hong. Infrared imaging dim target detection based on wavelet transform. J Huazhong Univ of Sci&, Tech,2000,28(5):69-71.
[52]Li Hong, Zheng Chengyong, Gao Jingli. Detecting small moving object based on wavelet multi-scale image registration. Infrared and Laser Engineering,2003,32(3): 268-270.
[53]Bruton L. T and Bartley N. The Enhancement and Tracking of Moving Object in Digital Images Using Adaptive 3D Rercursive Filters. IEEE Trans,1986,33(6).
[54]B. Porat and B. Friendlander. A Frequency Domain Approch to MultiFrame Detection and Estimation of Dim Targets. Proc. On ASSP,1987.