图像处理在衣服尺寸在线测量中的应用

在衣服尺寸在线测量过程中,提出了一种基于图像处理的衣服尺寸自动在线测量系统。针对于衣服尺寸自动测量困难等问题,运用了基于机器视觉的图像处理技术,采用自适应阈值Harris角点检测算法提取初步边缘角点,构成初步角点集,运用Susan角点检测算法去除初步角点集中的伪角点,使用Harris角点检测算法和Susan角点检测算法相结合的算法对衣服关键角点进行提取标定和检测分析。通过坐标值定位对关键角点的位置进行分析,并且利用比例计算进行尺寸测量。设计的系统用于对512×512的256级灰度图像进行检测,测量的相对误差小于5%,绝对误差小于10 mm。实验误差和尺寸测量精度能够达到服装尺寸测量的标准。

Application of image processing in on line clothes size measurement

In the process of clothes size measurement, a new automotive clothes size measurement system is presented. Aiming at thematter ofautomotive clothes size measurement, image processing techniques based on machine vision is used, an adaptive threshold Harris algorithm is utilized to extract edge corner andthe initial corner set is constituted, and then Susan algorithm is used to remove the false contour on the corner. The method combining Harris algorithm and Susan algorithm is apply to detect calibration and extract analysis of clothes diagonal point. Finally, the key corner position is analyzed by the coordinate values positioning, and the proportional calculation is used to measure the clothes size. The system is designed for testing with 256 gray scale image of 512×512, the relative error is less than 5% and the absolute error of measurement is less than 10 mm. The experimental error can be measured up the clothing standard.