基于免疫克隆聚类协同神经网络的图像识别
doi: 10.3724/SP.J.1146.2007.00405
Image Recognition Using Synergetic Neural Networks Based on Immune Clonal Clustering
-
摘要: 该文提出了基于免疫克隆聚类的协同神经网络原型向量求解算法，该算法充分利用免疫克隆的高效全局最优搜索能力构造数据聚类算法，将新聚类算法用于训练协同神经网络的原形向量，并对Brodatz纹理图像库以及合成孔径雷达图像目标进行识别。仿真实验结果表明，相比标准协同神经网络，该算法可以提高网络的识别性能，同经典的支撑向量机相比，该算法在识别率相当的情况下，样本的训练和测试时间都明显缩短。Abstract: A novel image recognition algorithm, Synergetic Neural Networks (SNN) based on immune clonal lgorithm, is proposed in this paper. The presented method introduces the global optimal searching ability of immune clonal select algorithm to construct data clustering algorithm, which used to solve the prototype vector in SNN. The simulation result of the Brodatz images and Synthetic Aperture Radar (SAR) images show the proposed algorithm can improve the performance of SNN as compared with the standard SNN and it can reduce greatly the training and test time leave the classification accuracy almost unchanged as compared with the traditional support vector machine.
-
Haken H. Synergetic Computers and Recognition--aTop-Down Approach to Neural Nets. Berlin: Springer-Verlag,1991: 45-80.[2]王海龙，戚飞虎. 基于聚类法的协同神经网络学习算法. 上海交通大学学报, 1998, 32(10): 39-41.Wang H L and Qi F H. Learing algorithm of synergetic neuralNetwork based on clustering algorithm. Journal of ShanghaiJiaotong University, 1998, 32(10): 39-41.[3]Hinton G E. Connectionist learning procedures. ArtificialIntelligence, 1989, 40(1-3): 185-234.[4]Shutton R S. Two problems with backpropagation and othersteepest-decent learning procedures for networks,Proceedings of 8th Annual Conference of the CognitiveScience Soiety, Erlbaum, Hillsdale, NJ, 1986: 823-831.[5]焦李成，杜海峰. 人工免疫系统进展与展望. 电子学报，2003,31(10): 1540-1548.Jiao L C and Du H F. Development and prospect of artificialimmunity system. Acta Electronica Sinica, 2003,31(10):1540-1548.[6]Liu R C.[J].Du H F, and Jiao L C. Immunity clonal strategies.Proceedings of Fifth International Conference onComputational Intelligence and Multimedia Applications,China: Xian, IEEE.2003,:-[7]Babu G P and Murty M M. Clustering with evolutionstrategies[J].Pattern Recognition.1994, 27(2):321-329[8]Bezdek J C and Hathaway R J. Optimization of fuzzyclustering criteria using genetic algorithm. Proceedings of theFirst IEEE Conference on Evolutionary Computation, 1994,2: 589-594.[9]Weston J and Watkins C. Multi-class support vectormachines. Technical Report CSD-TR-98-04, Royal HollowayUniversity of London, 1998.[10]张向荣. 遥感图像的特征提取与目标识别方法研究. [硕士论文]，西安电子科技大学，2003.Zhang Xian-grong. Research on feature extraction and targetrecognition in remote sensing images. [Master dissertation],Xidian University, January 2003.[11]Ma X L and Jiao L C. An effective learning algorithm ofsynergetic neural network[J].Lecture Notes in ComputerScience.2004, 3173:258-263 -
计量
- 文章访问数: 3564
- HTML全文浏览量: 92
- PDF下载量: 1132
- 被引次数: 0
下载:
