摘要
性能评估是通过向ATR系统输入指定图像序列,统计分析ATR系统产生的数据,并根据评估数据发现系统缺陷,从而调整系统参数,提升系统性能的有力手段。性能评估需要大量的各种条件下的图像作为ATR系统的输入,但是采集大量的图像需要耗费巨大的人力物力,这种情况促使图像合成方法被日益重视。同时必须对评估结果数据加以有效地分析,才能获得被评估系统的真实性能。本文针对已有性能评估中的图像合成方法不足,提出利用温度预测模型和纹理置换方法,模拟在指定条件下红外图像,并且对评估系统结果数据采用各种回归模型和插值方法进行了分析。本文包含的内容有:一种温度预测模型,两种随机场纹理模型及它们的参数估计方法、合成图像背景区域分布构造方法以及性能评估数据回归分析和插值方法。
每一种地表的表面红外辐射度随天气时间变化而变化,不同地表的这种变化是有差异的。根据采集的红外图像,模拟指定天气条件下的红外图像,必须考虑这种变化差异。本文利用已有的一维温度预测模型预测不同地表的温度,并转换成红外辐射度,然后转换成模拟图像各像素的灰度值,从而获得指定天气时间条件下的图像。
对合成图像背景某地表区域进行纹理置换,可能会遇到纹理图像尺寸不够大的问题。为了解决该问题,本文采用纹理模型对已有纹理图像进行模拟,生成所需尺寸的纹理图像。本文介绍了马尔科夫随机场模型和一般长纹理模型,并给出了这两种模型参数的估计方法。纹理置换的另一个问题是置换区域问题。要进行纹理置换,首先要获得合成图像背景区域分布,这就需要对背景图象进行区域编码。本文采用两种方式对背景图象编码,一是图像分割方法,而是人工几何方法。本文详细介绍了采用马尔科夫随机场图像分割算法和各方向灰度方差特征分类方法进行背景区域编码。
评估数据的分析采用了线性回归和非线性回归两种,其中非线性回归方法采用多项式回归,插值方法则采用分段多项式以及B-样条插值方法。本文采用多项式回归、分段多项式插值及B-样条插值对评估数据进行拟合实验,取得了一定的效果。
Performance evaluation is a strong method to improve ATR system performance by inputting special image data to ATR system, analyzing result data and finding disadvantages of it and adjusting system parameters. Performance evaluation needs lots of image data as the input of ATR system, but collecting image data is very costly, and this situation urge the research of the image forming technology. The result of performance evaluation must be effectively analyzed so that real performance of the system can be got. Aiming at the shortcoming of original image forming technology, this paper give the solutions by modeling temperature of the surface of different terrain regions and replacing texture to simulate IR image in given conditions and using several recursion methods to analyzing performance data. This paper include temperature of surface of different terrain regions ,texture models and parameters estimation method ,method of coding different terrain regions of background and recursion methods of result data.
Radiance of different terrain regions change with weather and time and the change of those regions is different .It must be considered in simulating image of given condition from real image. This paper use temperature model of terrains to predict temperature of given condition, then transforms it to radiance and then get the gray level of result image. Replacing the texture of a region in background image will meet a problem. It is that the scale of texture image is not big enough. To overcome this problem, this paper use texture models to simulate texture in needed scale. MRF models and GLC models are introduced here and methods of parameters estimation. Another problem is that we must get the distribution of background. Here are two methods. The first one is image segmentation technology; the second one is geometric method .We introduce two image segmentation methods. One is based on MRF; another is based on features of deviation of four orientations.
Linear regression and non-linear regression methods are used in performance evaluation data. Non-linear method used in this paper is multinomial. Subsection multinomial and b-spine methods are also introduced. The three methods have been used in evaluation system.
引文
[1] 王广银,吕进,吴振武.热红外图像的计算机模拟.工程兵工程学院学报, 1997,12(4):64-69
[2] A. Marshall,K. St, J. Murphy. Modeling and Simulation of IR Scenes and Sensors in the UK-Current and Future Capabilities. SPIE,1996, 2744:59l-598
[3] N. Ben-Yosef, B. Rahat, G. Feifin. Simulation of IR Images of Natural Background. Applied Optics,1983,22 (1):190-193
[4] N. Ben-Yosef, K.Wilner, M.Abitbol. Natural Terrain in the Infrared: Measurements and Modeling. SPIE,1987,819:66-71
[5] Uri Bernstein, Catherine E.Keller. A thermal model for real-time textured IR background simulation. SPIE,1991,1486:345-351
[6] D.M Meredith, C. p. Walters, Carl W. Hoover. Suitability of synthetic imagery for ATR evaluation. IEEE, 30th Applied Imagery Pattern Recognition Workshop 2001:51-56
[7] Clarence P.Walters, Carl W. Hoover,Jr., James A. Ratches. Performance of an automatic target recognizer algorithm against real and two versions of synthetic imagery. Optical Engineering, 2000,39(8):2279-2284
[8] 宣益民,吴轩,韩玉阁.坦克红外热像理论建模和计算机模拟.弹道学报,1997,9 (1):17-21
[9] 乔学勇,宣益民,韩玉阁.坦克三维瞬态温度场的边界元算法.兵工学报,1995, 20 (1):29-33
[10] 毛宏霞,高兰香,杨保成.海面舰船红外辐射场的灵敏因素分析.目标与环境特性研究, 2001,5 (2) :162-168
[11] 魏合理,宋正方.随机风场对自然地表红外辐射统计分布的影响.红外与毫米波学报, 1995, 14 (6) : 424- 428
[12] 张建奇,方小平,张海兴等.自然环境下地表红外辐射特性对比研究.红外与毫米波学报,1994,13 (6): 418- 424
[13] 毛宏霞,杨宝成,沈国土等.海面反射特性研究[J].华东师范大学学报(自然科学版), 2000, (3): 56-61
[14] 张建奇,方小平,张海兴等.雪面辐射温度预测模型.红外与毫米波学报,1997,16 (3):206-210
[15] 邵晓鹏,张建奇.基于GLC模型的红外纹理合成方法研究.红外与毫米波学报,2003,22(5):341-345
[16] 邵晓鹏,张建奇,徐军.一种自然红外纹理的生成方法.西安电子科技大学学报(自然科学版),2003,30(5):612-616
[17] 廖云涛. ATR性能评估方法的研究:[硕士学位论文].武汉:华中科技大学,2002
[18] Gavin Powell, Ralph Martin, David Marshall, Keith Markham. Simulation of FLIR Data Using Graphics Animation Software. Eighth Pacific Conference on Computer Graphics and Applications, 2000: 126-134
[19] 韩玉阁,宣益民.地面目标与背景红外热像模拟的现状、问题及对策.红外技术,2003,25(5):22-25
[20] Maximo Lorenzo, Bob Deaso, Young Lu, Jae Cha. DIS IR simulation models for fidelity, signature texture, and atmosphere-sensor effects. SPIE ,1995,2495:42-50
[21] 张建奇,方小平,张海兴,杨威.自然地表红外辐射特性统计分析.红外与毫米波学报,1997,16(4):266-272
[22] 杨德贵,黎湘,庄钊文.基于统一模型的典型地表红外辐射特性对比研究.红外与毫米波学报,2001,20(4):262-266
[23] 韩玉阁,宣益民.天然地形的随机生成及其红外辐射特性研究.红外与毫米波学报,2000,19(2):129-133
[24] 宣益民,李德沧,韩玉阁.复杂地面背景的红外热像合成.红外与毫米波学报,2002,21(2):133-136
[25] S. Geman, D. Geman. Stochastic Relaxation Gibbs Distributions and the BayesianRestoration of Images. IEEE Trans on PAMI,1984, 6:721-741
[26] H Elliott, H Derin, R Cristi, D Gemad. Application of the Gibbs Distribution to 1mage Segmentation. Proc. of IEEE 1984 Int Conf. on ASSP, San Diego, CA 1984:1-4
[27] H.Dein, H.Eliiott. Modeling and Segmentation of Noisy and Textured Images Using Gibbs Random Fields. IEEE Trans on PAMI, 1987,9(1):39-55
[28] 匡锦瑜,H.Derin.一种用于图象分割和恢复的MAP松弛算法.北京师范大学学报(自然科学报),1985,2:19-26
[29] 匡 锦 瑜 . 吉 布 斯 随 机 场 模 型 及 其 在 图 象 处 理 中 的 应 用 . 通 信 学报,1995,11(3):44-52
[30] Rupert Paget, I.D.Longsta.Texture Synthesis and Unsupervised Recognition with Nonparametric Multiscale Markov Random Field Models. IEEE Transactions on PAMI,1998,2:1068-1070
[31] John W. Woods. Two-Dimensional Discrete Markovian Fields. IEEE Transactions on Information Theory,1972,18(2):232-240
[32] G. R. Cross, A. K. Jain. Markov Random Filed Texture Models [J]. IEEE Trans on Pattern Analysis and Machine Intelligence,1983,5(1):25-39
[33] 于 晓 晗 , 袁 保 宗 . 复 合 MRFM 在 纹 理 合 成 及 压 缩 中 的 应 用 . 通 信 学报,1998,10(2):21-28
[34] J. Bennett, A. Khotanzad. Modeling Textured Images Using Generalized Long Correlation Models[J]. IEEE Trans on Pattern Analysis and Machine Intelligence,1998,20(12):365-370
[35] R.L. Kashyap, R. Chellappa. Estimation and Choice of Neighbors in Spatial-Interaction Models of Images. IEEE Trans. Information Theory, 1983,29(1):60-72
[36] Chee Sun Won, Haluk Derin. Maximum likelihood estimation of Gaussian Markov random field parameters. IEEE Acoustics Speech and Signal Processing, 1988,2(9):1040-1043
[37] 邵晓鹏,杨威,张建奇.自然地面背景红外图像生成方法研究.红外与激光工程,2000,29(3):11-14
[38] 刘伟强,陈鸿,夏德深.基于马尔可夫随机场的遥感图像分割和描述.东南大学学报,1999,29(增刊):11-15
[39] Moravec H. Robot rover visual navigation [M]. Ann Arbor, MI: University of Michigan Research Press,1981:123-172
[40] 马逢时,何良材,余明书,范金城. 应用概率统计(上册).第一版.武汉:华中理工大学出版社,1999:131-170
[41] T.帕夫利迪斯.计算机图形显示和图像处理算法.吴成柯.第一版.北京:科学出版社,1987:229-272