基于亚像素边缘检测的纱线条干均匀度测量

Yarn evenness measurement based on sub-pixel edge detection

为了精确地评价纱线的表观条干均匀性,提出一种基于亚像素边缘检测的纱线条干均匀度检测方法。首先通过图像采集装置获取纱线图像;然后对纱线图像进行亚像素边缘检测获取纱线边缘点集;利用开运算对获取的边缘点集进行处理,进而获得纱线条干的边缘;最后采用坐标直方图方法计算纱线平均直径和条干CV值。为验证方法的有效性和准确性,对多种不同线密度的纯棉纱线进行测试,并将测试结果与电容式条干仪的均匀度检测及基于模糊C均值分类和Otsu图像法的检测结果进行了对比。结果表明:基于亚像素边缘检测的纱线条干均匀度检测方法与电容式的测量结果有着较好的一致性,证明所提方法可得到准确的结果。

In order to evaluate accurately yarn evenness,a yarn evenness measurement method based on sub-pixel edge detection was proposed. In this process,a yarn image was acquired firstly by using an image acquisition device,after that yarn edge points were acquired by sub-pixel edge detection of yarn image. Then,the yarn edge points were processed by the mathematical opening operation,and yarn evenness edges were obtained. Consequently,the average yarn diameter and evenness CV value were calculated by using the coordinate histogram method. In order to verify the validity and accuracy of the method,cotton yarns with different linear densities were tested,and the test results were compared with the results of uniformity test using the capacitive evenness tester and with the test results obtained by using the fuzzy C-means algorithm and Otsu image method. The results of the comparisons show that the yarn evenness measurement method based on sub-pixel edge detection is in good agreement with the capacitance measurement results,which proves that the proposed method can obtain accurate results.