论文部分内容阅读
根据轮胎X光纹理缺陷区域灰度及灰度分布异常的特点,研究了一种通过分析统计特征进行在线缺陷检测的方法。在轮胎X光纹理灰度分布模型基础上,采用正则化预处理去除背景噪声,然后进行图像分块,分别计算每块的灰度均值和方差,并采用双线性插值运算形成均值图像和方差图像,再通过二值化实现缺陷检测。实验表明,与人工检测方法进行对比,该方法误判率低,检测精度高,并且运算速度快,能满足在线检测要求。
According to the abnormality of grayscale and grayscale distribution in the defect area of X-ray textured tire, a method of on-line defect detection by analyzing statistical features was studied. Based on the gray distribution model of X-ray texture of tire, the background noise is removed by regularization preprocessing, then the image is divided into blocks, the gray mean and variance of each block are calculated respectively, and the bilinear interpolation operation is used to form the mean image and variance Image, and then through the binary to achieve defect detection. Experiments show that, compared with the manual detection method, this method has low false positive rate, high detection accuracy and fast calculation speed, which can meet the requirements of online detection.