| 基于PCNN和SVM的图像识别方法研究 |
| |
| 引用本文： | 安国平,姜长生,吴庆宪.基于PCNN和SVM的图像识别方法研究[J].电光与控制,2008,15(10). |
| |
| 作者姓名： | 安国平 姜长生 吴庆宪 |
| |
| 作者单位： | 南京航空航天大学自动化学院,南京,210016;南京航空航天大学自动化学院,南京,210016;南京航空航天大学自动化学院,南京,210016 |
| |
| 基金项目： | 航空基金，国家自然科学基金 |
| |
| 摘 要： | 利用脉冲耦合神经网络(PCNN)具有良好的脉冲传播特性,研究了利用PCNN进行图像分割的技术.针对模型参数和最优分割结果难以确定的问题,改进了标准PENN模型,提出了基于互信患量的PCNN图像分割方法.计算原图像与PCNN分割图像序列问的互信息量,对应最大互信息量的分割图像就是最佳分割结果.然后选取不同类型飞机图像提取不变矩特征,用支持向量机(SVM)实现分类识别.实验结果表明,该方法分割效果好,适应性强,能够很好地识别飞机图像.
|
| 关 键 词： | 脉冲耦合神经网络 图像分割 互信息量 支持向量机 |
Study on an image recognition method based on PCNN and SVM |
| |
| AN Guo-ping,JIANG Chang-sheng,WO Qing-xian.Study on an image recognition method based on PCNN and SVM[J].Electronics Optics & Control,2008,15(10). |
| |
| Authors: | AN Guo-ping JIANG Chang-sheng WO Qing-xian |
| |
| Affiliation: | College of Automation Engineering;Nanjing University of Aeronautics & Astronautics;Nanjing 210016;China |
| |
| Abstract: | A new image segmentation method is studied based on Pulse-Coupled Neural Network(PCNN) by use of the good pulse propagation feature of PCNN.For its problems such as it is difficult to decide model parameters and the optimum segmentation result,the standard model of PCNN is improved,and an approach for image segmentation based on mutual information Maximized PCNN is proposed.Then the mutual information between original image and the sequence of segmentation images is calculated out,and the segmentation image... |
| |
| Keywords: | pulse-coupled neural network image segmentation mutual information SVM |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
