LBP和GLCM融合的织物组织结构分类

Fabric structure classification based on LBP and GLCM fusion

为实现织物组织结构的自动分类,提出一种基于局部二进制模式(LBP)与灰度共生矩阵(GLCM)相结合的织物组织分类算法。首先,采用中值滤波、双峰高斯函数规定化等算法对织物图像进行预处理,滤除图像噪声并提高对比度。进而用局部二进制模式和灰度共生矩阵两种方法获取图像的局部及全局纹理特征信息。最后,利用基于Levenberg-Marquardt(L-M)算法的BP神经网络分类器对特征向量进行训练和测试,实现对3种基本组织(平纹、斜纹和缎纹组织)的自动分类。实验结果表明,基于L-M算法的BP神经网络具有较快的训练速度能够对织物组织结构进行准确有效的分类。此外,与灰度共生矩阵和局部二进制模式方法进行对比,两者融合的特征信息能得到最好的分类结果(99.33%)。

To realize automatic classification of fabric structure, a fabric structure classification algorithm based on local binary pattern （LBP） and gray level co-occurrence matrix （GLCM） is proposed. Firstly, fabric image is pre- processed by median filter and bimodal Gaussian function specification algorithm in order to filter noise and improve contrast. Then, the two approaches which are local binary pattern and gray level co-occurrence matrix are applied to extract the local and global texture features of fabric image. Finally, an appropriate BP neural network classifier based on Levenberg-Marquardt （L-M） algorithm is used to training and testing the feature vector in order to achieve the automatic classification of three basic woven fabrics （plain, twill and satin weave）. The experimental results indicate that BP neural network classifier based on L-M algorithm with faster training speed can classify woven fab- rics accurately and efficiently. Besides, compared with GLCM method and LBP method, the fusion of the two fea- ture vectors obtains the best classification result （99.33 % ）.