A Target Recognition Algorithm for Sequential Aircraft Based on DSmT and HMM
-
摘要: 针对姿态多变化的飞机自动目标识别中的低识别率问题, 提出了一种基于DSmT (Dezert-Smarandache theory)与隐马尔可夫模型(Hidden Markov model, HMM)的飞机多特征序列信息融合识别算法(Multiple features and sequential information fusion, MFSIF). 其创新性在于将单幅图像的多特征信息融合识别和序列图像信息融合识别进行有机结合.首先, 对图像进行二值化预处理, 并提取目标的Hu矩和轮廓局部奇异值特征; 然后, 利用概率神经网络(Probabilistic neural networks, PNN)构造基本信度赋值(Basic belief assignment, BBA); 接着, 利用DSmT对该图像的不同特征进行融合,从而获得HMM的观察值序列;再接着, 利用隐马尔可夫模型对飞机序列信息融合, 计算观察值序列与各隐马尔可夫模型之间的相似度, 从而实现姿态多变化的飞机目标自动识别;最后, 通过仿真实验, 验证了该算法在飞机姿态发生较大变化时, 依然可以获得较高的正确识别率,同时在实时性方面也可以满足飞机目标识别的要求. 另外, 在飞机序列发生连续遮挡帧数τ ≤ 6的情况下, 也具有较高的飞机目标正确识别率.Abstract: For the low rate problem of recognition of automatic target recognition of aircraft caused by the great change of posture, a target recognition algorithm based on DSmT (Dezert-Smarandache theory) and HMM (Hidden Markov model) is proposed by utilizing the fusion of multiple features and sequential information (i.e. MFSIF). The novelty of the algorithm is integrating the multiple feature fusion recognition with the sequential images fusion recognition. Firstly, the sequential images are preprocessed with the binarization method, then the features of Hu moments and partial singular values of outline are picked up. Secondly, the PNN (Probabilistic neural network) is used to construct the BBA (Basic belief assignment). After that, these different features from the same image are fused by DSmT in order to gain the observation sequence of HMM. Then, the sequential information is fused by utilizing HMM to finish the automatic recognition of aircraft with varied multiple postures by calculating the similarities between the observation sequence and each HMM. Finally, according to the simulation experiment results, this algorithm has a high exact recognition rate even for aircraft with greatly varied postures. Simultaneously, this algorithm can also satisfy the requirement of aircraft target recognition in real-time. In addition, this algorithm can also guarantee a high recognition rate of aircraft target when the number τ of consecutive occulted frames is less than or equal to six.
-
[1] Marsiglia G, Fortunato L, Ondini A, Balzarotti G. Template matching techniques for automatic IR target recognition in real and simulated scenarios: tests and evaluations. In: Proceedings of SPIE-The International Society for Optical Engineering: Automatic Target Recognition XIII. Orlando, FL, USA: SPIE, 2003, 5094: 159-169 [2] Kwon O K, Sim D G, Park R H. Robust Hausdorff distance matching algorithms using pyramidal structures. Pattern Recognition, 2001, 34(10): 2005-2013 [3] Zhou Yu, Liu Jun-Tao, Bai Xiang. Research and perspective on shape matching. Acta Automatica Sinica, 2012, 38(6): 889-910 (周瑜, 刘俊涛, 白翔. 形状匹配方法研究与展望. 自动化学报, 2012, 38(6): 889-910) [4] Marouani S, Huertas A, Medioni G. Model-based aircraft recognition in perspective aerial imagery. In: Proceeding of International Symposium on Computer Vision. Coral Gables, FL, USA: IEEE, 1995. 371-376 [5] Das S, Bhanu B. A system for model-based object recognition in perspective aerial images. Pattern Recognition, 1998, 31(4): 465-491 [6] Gilmore J F. Knowledge-based target recognition system evolution. Optical Engineering, 1991, 30(5): 557-570 [7] Dudani S A, Breeding K J, McGhee R B. Aircraft identification by moment invariants. IEEE Transactions on Computers, 1977, C-26(1): 39-46 [8] Du Ya-Juan, Zhang Hong-Cai, Pan Quan. Three-dimensional aircraft recognition using moments. Journal of Data Acquisition and Processing, 2000, 15(3): 390-394 (杜亚娟, 张洪才, 潘泉. 基于矩特征的三维飞机目标识别. 数据采集与处理, 2000, 15(3): 390-394) [9] Gu Li, Zhuang Zhen-Quan, Zheng Guang-Yong, Wang Zai-Jian. Algorithm for hand shape matching based on feature fusion. Computer Applications, 2005, 25(10): 2286-2288 (顾理, 庄镇泉, 郑光勇, 王再见. 基于特征融合的手形匹配算法. 计算机应用, 2005, 25(10): 2286-2288) [10] Deng Cheng-Qiang, Feng Gang. Content-based image retrieval using combination features. Computer Applications, 2003, 23(7): 100-102 (邓诚强, 冯刚. 基于内容的多特征综合图像检索. 计算机应用, 2003, 23(7): 100-102) [11] Chen Li, Chen Jing. Multi-feature fusion method based on support vector machine and k-nearest neighbor classifier. Computer Applications, 2009, 29(3): 833-835 (陈丽, 陈静. 基于支持向量机和 k-!近邻分类器的多特征融合方法. 计算机应用, 2009, 29(3): 833-835) [12] Yang Fu-Ping, Bai Zhen-Xing. Target recognition method based on combination of BP neural networks with D-S evidence theory. Fire Control and Command Control, 2006, 31(10): 88-90 (杨福平, 白振兴. BP 神经网络和 D-S 证据理论的目标识别. 火力与指挥控制, 2006, 31(10): 88-90) [13] Li Xin-De, Yang Wei-Dong, Dezert J. An airplane image target's multi-feature fusion recognition method. Acta Automatica Sinica, 2012, 38(8): 1298-1307 (李新德, 杨伟东, Dezert J. 一种飞机图像目标多特征信息融合识别方法. 自动化学报, 2012, 38(8): 1298-1307) [14] Huang Jin, Liang Yan, Cheng Yong-Mei, Pan Quan, Hu Jin-Wen. Automatic target recognition method based on sequential images. Acta Aeronautica et Astrontica Sinica, 2006, 27(1): 87-93 (黄金, 梁彦, 程咏梅, 潘泉, 胡劲文. 基于序列图像的自动目标识别算法. 航空学报, 2006, 27(1): 87-93) [15] Hou Jun, Miao Zhuang, Pan Quan. Intelligent target recognition method of sequential images based on DSmT. Computer Applications, 2006, 26(1): 120-122 (侯俊, 苗壮, 潘泉. 基于DSmT的序列图像智能融合目标识别方法. 计算机应用, 2006, 26(1): 120-122) [16] Huang Z H, Leng J S. Analysis of Hu's moment invariants on image scaling and rotation. In: Proceedings of the 2nd International Conference on Computer Engineering and Technology. Chengdu, China: IEEE Computer Society, 2010. 7: 476-480 [17] Chen S Y, Feng J. Research on detection of fabric defects based on singular value decomposition. In: Proceedings of 2010 IEEE International Conference on Information and Automation. Harbin, China: IEEE, 2010. 857-860 [18] Hu M K. Visual pattern recognition by moment invariants. IEEE Transactions on Information Theory, 1962, 8(2): 179-187 [19] Pan Jin-Dong, Li Xin-De. An airplane target recognition method based on singular value and PNN. Aero Weaponry, 2013, (1): 45-50 (潘锦东, 李新德. 一种奇异值与PNN结合的飞机目标识别方法. 航空兵器, 2013, (1): 45-50) [20] Dezert J. Foundations for a new theory of plausible and paradoxical reasoning. Information and Security, 2002, 9: 13-57 [21] Smarandache F, Dezert J. On the generation of hyper-powersets for the DSmT. In: Proceedings of the 6th International Conference on Information Fusion. Cairns, QLD, Australia: IEEE Computer Society, 2003. 2: 1118-1125 [22] Li X D, Dezert J, Smarandache F, Huang X H. Evidence supporting measure of similarity for reducing the complexity in information fusion. Information Sciences, 2011, 181(10): 1818-1835 [23] Li Xin-De, Dezert J, Huang Xin-Han, Meng Zheng-Da, Wu Xue-Jian. A fast approximate reasoning method in hierarchical DSmT(A). Acta Electronica Sinica, 2010, 38(11): 2566-2572 (李新德, Dezert J, 黄心汉, 孟正大, 吴雪建. 一种快速分层递阶DSmT近似推理融合方法(A). 电子学报, 2010, 38(11): 2566-2572) [24] Smarandache F, Dezert J. Advances and Applications of DSmT for Information Fusion (Collected Works). USA: Infinite Study, 2006, Vol. 2. [25] Specht D F. Probabilistic neural networks. Neural Networks, 1990, 3(1): 109-118 [26] Specht D F. Enhancements to probabilistic neural networks. In: Proceedings of 1992 International Joint Conference on Neural Networks. New York, USA: IEEE, 1992: 761-768 [27] Baum L E, Petrie T. Statistical inference for probabilistic functions of finite state Markov chains. The Annals of Mathematical Statistics, 1966, 37(6): 1554-1563 [28] Baum L E, Eagon J A. An inequality with applications to statistical estimation for probabilistic functions of Markov processes and to a model for ecology. Bulletin of the American Mathematical Society, 1967, 73(3): 360-363 [29] Baum L E, Petrie T, Soules G, Weiss N. A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains. The Annals of Mathematical Statistics, 1970, 41(1): 164-171 [30] Rabiner L R. A tutorial on hidden Markov models and selected applications in speech recognition. Proceedings of the IEEE, 1989, 77(2): 257-286 [31] Makhoul J, Roucos S, Gish H. Vector quantization in speech coding. Proceedings of the IEEE, 1985, 73(11): 1551-1588 [32] Kuo S S, Agazzi O E. Machine vision for keyword spotting using pseudo 2D hidden Markov models. In: Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing. Minneapolis, MN, USA: IEEE, 1993: 81-84 [33] Wang Xiang-Hai, Cong Zhi-Huan, Fang Ling-Ling, Qin Ju-Ao. Determination of real-time vehicle driving status using HMM. Acta Automatica Sinica, 2013, 39(12): 2131-2142 (王相海, 丛志环, 方玲玲, 秦钜鳌. 基于HMM 的车辆行驶状态实时判别方法研究. 自动化学报, 2013, 39(12): 2131-2142) [34] Dempster A P, Laird N M, Rubin D B. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society. Series B (Methodological), 1977, 39: 1551-1588 [35] Welch L R. Hidden Markov models and the baum-welch algorithm. IEEE Information Theory Society Newsletter, 2003, 53(4): 10-13
点击查看大图
计量
- 文章访问数: 1858
- HTML全文浏览量: 75
- PDF下载量: 1208
- 被引次数: 0