| 噪声环境下木材纹理分类的研究 |
| |
| 引用本文： | 王辉,王克奇,白雪冰.噪声环境下木材纹理分类的研究[J].林业机械与木工设备,2006,34(10):13-15. |
| |
| 作者姓名： | 王辉 王克奇 白雪冰 |
| |
| 作者单位： | 东北林业大学机电工程学院,黑龙江,哈尔滨,150040 |
| |
| 基金项目： | 黑龙江省自然科学基金项目(C2004-03)，哈尔滨市自然科学基金项目(2004AFXXJ020) |
| |
| 摘 要： | 为了实现木材分类识别的自动化,应用灰度共生矩阵建立了木材纹理的参数体系,并进行了分类研究。首先在无噪声的环境下提取了木材的共生矩阵纹理原始特征参数,并对其进行特征选择,进而建立了木材纹理参数体系。对该参数体系进行噪声适应性测试的实验结果表明,无噪声情况下样本识别率为87.50%;0.2%￣1.0%椒盐噪声环境下样本识别率范围为87.00%￣88.00%。表明该参数体系具有良好的抗击噪声能力和一定的工程实用价值。
|
| 关 键 词： | 木材纹理 灰度共生矩阵 特征选择 噪声 分类 |
| 文章编号： | 1001-4462（2006）10-0013-03 |
| 收稿时间： | 2006-06-19 |
| 修稿时间： | 2006年6月19日 |
The Research on Wood Texture Classification in Noise Environment |
| |
| WANG Hui,WANG Ke-qi,BAI Xue-bing.The Research on Wood Texture Classification in Noise Environment[J].Forestry Machinery & Woodworking Equipment,2006,34(10):13-15. |
| |
| Authors: | WANG Hui WANG Ke-qi BAI Xue-bing |
| |
| Institution: | College of Machinery Electricity of Northeast Forestry University, Heilongjiang Harbin 150040, China |
| |
| Abstract: | In order to realize the automation of wood classification and recognition,wood texture parameter system was constructed by GLCM,and the research of classification was carried on.First,original co-occurrence matrix feature parameters of wood were extracted in noise condition,carried on feature selection,and constructed parameter system of wood texture.Next,noise compatibility test of this parameter system was carried on.The simulation result was: the sample recognition rate was 87.50% in non-noise condition;in 0.2%~1.0% salt-pepper noise condition,the scale of sample recognition rate was 87.00%~88.00%,which indicated this parameter system has good ability of noise resistance and the project practical value. |
| |
| Keywords: | wood texture gray level co-occurrence matrix(GLCM) feature selection noise classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
