| 基于神经网络的CT脑血管图像边缘检测算法 |
| |
| 引用本文： | 秦然.基于神经网络的CT脑血管图像边缘检测算法[J].电子测量与仪器学报,2010,24(4):346-352. |
| |
| 作者姓名： | 秦然 |
| |
| 作者单位： | 山东省青年学院信息工程系,济南,250014 |
| |
| 基金项目： | 山东省教育厅科技发展计划项目 |
| |
| 摘 要： | CT脑血管医学图像的三维重构都是源自二维断层扫描,脑血管边缘特征向量的提取是图像处理的关键步骤。为提高边缘特征的提取和保证三维重建图像的质量,在分析了某些常用的边缘检测算法性能基础上,同时结合CT脑血管图像的像素结构特点,将SA_SOFM神经网络算法成功地用于对CT脑血管图像的边缘特征信息提取中。并对算法进行有效的改进,基于真实图像的实验表明该算法提高了边缘特征信息的精度和鲁棒性。
|
| 关 键 词： | SA_SOFM神经网络 CT脑血管图像 边缘特征 检测算法 |
Edge information extraction algorithm for CT cerebrovascular medical image based on neural network |
| |
| Qin Ran.Edge information extraction algorithm for CT cerebrovascular medical image based on neural network[J].Journal of Electronic Measurement and Instrument,2010,24(4):346-352. |
| |
| Authors: | Qin Ran |
| |
| Affiliation: | Qin Ran(Department of Information and Engineering,Shandong Youth College,Jinan 250014,China) |
| |
| Abstract: | The source data of 3-dimension CT cerebrovascular medical image reconstruction come from 2D to-mography slices,the key part of pattern recognition is edge information extraction.To insure the quality of 3D image reconstruction and detail of edge information extraction,the characters of algorithm usually used and internal tissue of cerebrovascular image are analyzed.The SA_SOFM neural network is introduced in the edge information extraction for CT cerebrovascular medical image successfully.The SA_SOFM neural network algorithm is improved under the certain condition,the experiments on real images show that the method proposed in this paper is more accurate and robust. |
| |
| Keywords: | self-adaptive self_organizing feature map neural network CT cerebrovascular image edge information extraction detection algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
