中文摘要：

在模式识别中，灰度共生矩阵（GLCM）能够很好的提取图片的纹理特征，流形学习中的局部线性嵌入（LLE）方法是一种有效的非线性降维方法。结合两者的优点，经过严格的推导，提出一种基于灰度共生矩阵与流形学习的金属断口图像识别方法GLCM—LLE。将提出的方法与传统的基于灰度共生矩阵的方法进行对比，实验结果表明，提出的方法在识别率方面优于GLCM方法，具有实用、有效的优点。

英文摘要：

In pattern recognition, gray level co-occurrence matrix （GLCM） can extract texture feature of image, and locally linear embedding （LLE） method in the manifold learning is an effective nonlinear dimensionality reduction method. Combined the advantages of GLCM and LLE and undergoing a rigorous derivation, a recognition method of metal fracture image, so-called GLCM - LLE, is proposed. In comparison with the traditional GLCM method,experiment results show that the proposed method is superior to the traditional GLCM method in recognition rate, and it has the practical and effective advantages.