An Airplane Image Target's Multi-feature Fusion Recognition Method
-
摘要: 提出了一种基于概率神经网络(Probabilistic neural networks, PNN)和DSmT推理 (Dezert-Smarandache theory)的飞机图像目标多特征融合识别算法. 针对提取的多个图像特征量,利用数据融合的思想对来自图像目标各个特征量提供的信息进行融合处理.首先,对图像进行二值化预处理,并提取Hu矩、归一化转动惯量、 仿射不变矩、轮廓离散化参数和奇异值特征5个特征量;其次, 针对DSmT理论中信度赋值构造困难的问题,利用PNN网络,构造目标识别率矩阵,通过目标识别率矩阵对证据源进行信度赋值;然后,用DSmT组合规则在决策级层进行融合,从而完成对飞机目标的识别;最后,在目标图像小畸变情形下, 将本文提出的图像多特征信息融合方法和单一特征方法进行了对比测试实验,结果表明本文方法在同等条件下正确识别率得到了很大提高,同时达到实时性要求,而且具有有效拒判能力和目标图像尺寸不敏感性. 即使在大畸变情况下,识别率也能达到89.3%.
-
关键词:
- 信息融合 /
- 目标识别 /
- Dezert-Smarandache推理 /
- 飞机图像 /
- 概率神经网络
Abstract: This paper proposes an image target's multi-feature fusion recognition method based on probabilistic neural networks (PNN) and Dezert-Smarandache theory (DSmT). To aim at multiple features extracted from an image, the information from them is fused. Firstly, the image is preprocessed with binarization and then multiple features are extracted, such as Hu moments, normalized moment of inertia, affine invariant moments, discrete outline parameters and singular values. Secondly, due to the difficulty of the construction of the basic belief assignment in DSmT, in this paper the target recognition rate matrix is constructed by PNN, that is, the basic belief assignments can be assigned to the evidence sources by PNN. Finally, the procedure of airplane target recognition can be accomplished by the DSmT combination rule at the level of decision fusion. For small distortion of target image, the multi-feature fusion method proposed in this paper is compared with the single-feature one through a series of experiments. The experimental result in this paper proves that this method greatly improves the right recognition rate, satisfies real-time requirements, and has good ability of rejection of judgement and strong insensitivity to target image size. And even for big distortion, the right recognition rate can also reach 89.3%. -
[1] Dudani S A, Breeding K J, McGhee R B. Aircraft identification by moment invariants. IEEE Transactions on Computers, 1977, C-26(1): 39-46[2] Roberts G A. Aircraft recognition using a parts analysis technique. In: SPIE Proceedings on Applications of Digital Image Processing XI. Bellingham, WA: Society for Photo-Optical Instrumentation Engineers, 1988. 338-345[3] 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)[4] Flusser J, Suk T. Pattern recognition by affine moment invariants. Pattern Recognition, 1993, 26(1): 167-174[5] Mitchell R A, Westerkamp J J. Robust statistical feature based aircraft identification. IEEE Transactions on Aerospace and Electronic Systems, 1999, 35(3): 1077-1094[6] Fung G M, Mangasarian O L. A feature selection Newton method for support vector machine classification. Computational Optimization and Applications, 2004, 28(2): 185-202[7] Cheng Yong-Mei, Pan Quan, Zhang Hong-Cai, Wang Gang. Information fusion image recognition algorithm and applications in three dimension airplane image recognition. Acta Aeronautica et Astronautica Sinica, 2004, 25(2): 176-179 (程咏梅, 潘泉, 张洪才, 王刚. 信息融合图像识别算法及其在三维飞机图像识别中的应用研究. 航空学报, 2004, 25(2): 176-179)[8] Hou Jun, Miao Zhuang, Pan Quan. Intelligent target recognition method of sequential images based on DSmT. Journal of 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)[9] Smarandache F, Dezert J. Advances and Applications of DSmT for Information Fusion. USA: American Research Press, 2004/2006/2009. Vol.1-3[10] Li X D, Dezert J, Smarandache F, Dai X Z. Combination of qualitative information based on 2-tuple modelings in DSmT. Journal of Computer Science and Technology, 2009, 24(4): 786-798[11] Li X D, Dezert J, Smarandache F, Huang X H. Evidence supporting measure of similarity for reducing the complexity in information fusion. Information Sciences, 2010. 181(10): 1818-1835[12] Li Xin-De, Yang Wei-Dong, Wu Xue-Jian, Dezert Jean. A fast approximate reasoning method in hierarchical DSmT (B). Acta Electronica Sinica, 2011, 39(3A): 31-36 (李新德, 杨伟东, 吴雪建, Dezert Jean. 一种快速分层递阶DSmT近似推理融合方法(B). 电子学报, 2011, 39(3A): 31-36)[13] Li Xin-De, Dezert Jean, 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 Jean, 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)[14] 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, 庄镇犬, 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)[15] 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)[16] Chen Li, Chen Jing. Multi-feature fusion method based on support vector machine and K-nearest neighbor classifier. Journal of 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[17] 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, Sichuan: IEEE Computer Society CPS, 2010. 7: 476-480[18] Wei M, Chen H N, Yan T Y, Zhou J, Wu Q Z, Xu B. NMI-based small target detecting method. In: Proceedings of the 2nd International Conference on Electrical and Control Engineering. Wuhan, Hubei: IEEE Computer Society CPS, 2010. 1363-1366[19] Wei Jia-Jie, Liu Zhi-Gui, Tang Yi. Research of key recognition based on improved affine invariant moments. Microcomputer Information, 2010, 26(17): 196-197 (魏嘉杰, 刘知贵, 唐宜. 基于改进的仿射不变矩的钥匙辨识技术研究. 微计算机信息, 2010, 26(17): 196-197)[20] 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, Heilongjiang: IEEE, 2010. 857-860[21] Specht D F. Probabilistic neural networks. Neural Networks, 1990, 3(1): 110-118[22] Specht D F. Enhancements to probabilistic neural networks. In: Proceedings of IEEE International Joint Conference on Neural Networks. New York, USA: IEEE, 1992. 761-768[23] Yang Bo, Jing Zhong-Liang. Image fusion algorithm based on the quincunx-sampled discrete wavelet frame. Acta Automatica Sinica, 2010, 36(1): 12-22 (杨波, 敬忠良. 梅花形采样离散小波框架图像融合算法. 自动化学报, 2010, 36(1): 12-22)[24] Li Wei, Zhu Xue-Feng. An image fusion algorithm based on second generation wavelet transform and its performance evaluation. Acta Automatica Sinica, 2007, 33(8): 817-822 (李伟, 朱学峰. 基于第二代小波变换的图像融合方法及性能评价. 自动化学报, 2007, 33(8): 817-822)[25] Liu Gui-Xi, Yang Wan-Hai. A wavelet-decomposition-based image fusion scheme and its performance evaluation. Acta Automatica Sinica, 2002, 28(6): 927-934 (刘贵喜, 杨万海. 基于小波分解的图像融合方法及性能评价. 自动化学报, 2002, 28(6): 927-934)[26] Chen Rong-Yuan, Lin Li-Yu, Wang Si-Chun, Qin Qian-Qing. Remote sensing image fusion based on differential evolution algorithm under data assimilation framework. Acta Automatica Sinica, 2010, 36(3): 392-398 (陈荣元, 林立宇, 王四春, 秦前清. 数据同化框架下基于差分进化的遥感图像融合. 自动化学报, 2010, 36(3): 392-398)
点击查看大图
计量
- 文章访问数: 2111
- HTML全文浏览量: 76
- PDF下载量: 1373
- 被引次数: 0