基于Hough变换和FFT多投影精测的织物纬斜快速检测方法研究

织物纬斜角度检测是布料整纬的关键技术环节,如何快速、准确检测纬斜角度对提高整纬质量具有重要意义;针对现有布料图像整纬方法存在速度慢、精度不高的缺点,提出了基于Hough变换原理和快速傅里叶变换(FFT)多投影的织物纬斜图像快速检测方法;首先对采集到的织物图像通过傅里叶变换对图像进行频域滤波再逆变换,滤除图像中不表示纬斜方向的区域信息,其次使用Sobel边缘方向检测算子对图像进行卷积以得到边缘方向图,提取纬纱方向信息,利用形态学滤波得到纬纱骨架图,进一步精简纬纱区域以减少计算量,最后进行Hough变换和FFT多投影分析得到织物图像的纬斜角度;实验测试证明对于不同类型的织物图像,该算法的检测时间低于0.55 s,误差值低于0.2°,能够兼顾检测精度和检测速度,满足工程实际应用要求.

Fast detection algorithm of fabric weft skew based on FFT multi-projection and Hough transform

Fabric weft angle detection is a key technology link in fabric weft rectification. How to detect the weft angle quickly and accurately is of great importance to improve the quality of the weft rectification. In view of the shortcomings of the existing fabric image rectification methods, which are slow and not high accuracy, a fast detection method of fabric weft angle image based on Hough transform principle and fast Fourier transform multi-projection is proposed.The fabric image of the Sobel is convoluted to get the edge pattern by using the Sobel edge direction detection operator, the weft direction information is extracted, the weft skeleton is obtained by using the morphological filter, and the weft area is further simplified to reduce the amount of calculation.The Hough transformation and FFT multi-projection analysis are used to get the weft-skew angle of the fabric image. The experimental results show that for different types of fabric images, the detection time of the algorithm is less than 0.55s, and the error value is less than 0.2 degree, which can take into account both the detection accuracy and the detection speed, and meet the practical application requirements of engineering.