纺织品疵点自动识别方法研究

疵点检测是纺织品质量检验中的重要一环，因织物的组织结构多样，不同织物表面的纹理特征具有较大差异，很难建立一种通用的数学模型来描述。介绍了一种基于SVM和自适应LBP算子的织物疵点识别方法。利用自适应LBP算子的灰度不变性和自适应纹理分析性能提取图像的归一化直方图特征参数，并将其作为支持向量机的输入参数进行训练，获得支持向量。然后将支持向量机作为分类器最终实现疵点图像的识别。实验表明，该方法的准确性较高，速度快，适应性强，是一种比较理想的织物疵点识别方法。

Research on the textile defect auto-recognition method

The fabric fault detection is one of the most important processes in the textile testing. Owing to numerous kinds of fabric weaves and varied characteristics of fabric surface,it is very difficult to establish a universal fabric defect detection model. A method for identification of the defects on the fabric by using the SVM and self-reacting LBP is introduced. Using the gray-level invariance and self-reacting capability of the LBP to get the unitized histogram character parameters. The parameters are input...