灰度共生矩阵在蔬菜种类识别中的应用研究

Application of gray level co-occurrence matrix in vegetable species recognition

针对我国蔬菜类食品安全追溯系统中有关种类信息采集存在的一些问题,提出了一种利用图像纹理分析技术对多种蔬菜进行识别分类,进而提高信息采集的准确性与高效性.实验中分别用灰度共生矩阵算法(GLCM)、高斯马尔可夫模型法(GMRF)和Gabor法分别对蔬菜的纹理进行特征提取,并通过支持向量机(SVM)分类对比.结果证明,由灰度共生矩阵对蔬菜种类识别可以得到较好的分类结果.

Aimed at the problem of the acquisition of vegetable information, we proposed a texture analysis technique to identify and classify the variety of vegetables so as to improve the efficiency and accuracy of information acquisition. Vegetable texture features were extracted with the methods of gray level co-occurrence matrix （GLCM）, Gauss-Markov random field （GMRF） and Gabor,respectively, and were compared with that of the support vector machine （SVM）. It is found that vegetable recognition with the method of gray level co-occurrence matrix is the optimum.