| 中药材粉末显微图像模式识别方法 |
| |
| 引用本文： | 王亚杰,徐心和.中药材粉末显微图像模式识别方法[J].农机化研究,2006(12):180-183. |
| |
| 作者姓名： | 王亚杰 徐心和 |
| |
| 作者单位： | 1. 东北大学,教育部暨辽宁省流程工业综合自动化重点实验室,沈阳,110004;沈阳航空工业学院,计算中心,沈阳,110034
2. 东北大学,教育部暨辽宁省流程工业综合自动化重点实验室,沈阳,110004 |
| |
| 摘 要： | 提出了一种适合中药材粉末状态下显微图像的模式识别方法。使用3种分析方法进行特征提取,即分形分析方法、小波分析方法、灰度/梯度共生矩阵法,经特征选择后,对图像分类贡献较大的8个特征量被组合成特征向量,用于图像识别。实验中采用了k-近邻分类方法,其较高的识别率表明,此方法是可行的和有效的,为今后进一步进行中药材识别提供了理论参考。
|
| 关 键 词： | 农业基础科学 中药材粉末显微图像模式识别 分析 特征计算 特征选择 分类识别 |
| 文章编号： | 1003-188X（2006）12-0180-04 |
| 收稿时间： | 2006-04-16 |
| 修稿时间： | 2006年4月16日 |
Pattern Recognition Method on Micro-images of Powdered Chinese Medical Herbs |
| |
| WANG Ya-jie,XU Xin-he.Pattern Recognition Method on Micro-images of Powdered Chinese Medical Herbs[J].Journal of Agricultural Mechanization Research,2006(12):180-183. |
| |
| Authors: | WANG Ya-jie XU Xin-he |
| |
| Abstract: | A new way to recognize the micro-image of powdered Chinese medical herbs is put forward. The characteristic quantities are calculated by three methods, ie., fractal analysis, wavelet analysis and gray level-gradient joint occurrence matrix. The characteristic vectors are formed by combining together eight characteristic quantities that are picked out according to how important they play in the classification of images. The characteristic vector is used to recognition of images. Using the K-nearest neighbor classification, the experimental results show that the way proposed is feasible and effective, and provide a theoretical reference for the farther recognition of Chinese medical herbs for the future. |
| |
| Keywords: | agricultural basic science micro-images pattern recognition of powdered Chinese medical herbs analysis characteristic calculation characteristic selection classification recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
