面向珍珠分拣机器人的形状视觉检测方法  

Visual Inspection Method of Pearl Shape for Sorting Robot

作　　者：魏哲[1] 王盼 WEI Zhe;WANG Pan(Xi’an Polytechnic University,Xi’an 710048,China)

机构地区：[1]西安工程大学,陕西西安710048

出　　处：《机械与电子》2021年第8期68-71,76,共5页Machinery & Electronics

基　　金：国家自然科学基金(61701384);陕西省自然科学基础研究计划项目(2017JM5141);西安市科技局创新引导项目(201805030YD8CG14(5))。

摘　　要：针对人工进行珍珠形状分拣效率低、精度不稳定等问题,提出基于机器视觉的珍珠形状检测方法。采用背光成像方式消除珍珠表面纹理和光泽的影响,对获取的珍珠图像进行同态滤波等预处理,提高图像对比度。为了解决相互接触珍珠影响珍珠轮廓提取的问题,采用分水岭算法对珍珠图像进行分割,得到独立存在的珍珠个体,再通过连通域标记、质心算法对珍珠进行定位。根据国家标准对珍珠形状的规定,基于珍珠图像信息建立珍珠形状参数模型,对珍珠形状进行量化。实验结果表明,不同形状的珍珠样本的检测误差为0.63%,形状统计精度为100%,算法耗时24 ms。A pearl shape detection method based on machine vision is proposed to solve the problem of low efficiency and low precision in artificial pearl shape sorting.The backlight imaging method is used to eliminate the influence of the pearl surface texture and luster,and the obtained pearl image is subjected to preprocessing algorithms such as homomorphic filtering to improve the image contrast.In order to solve the problem of extracting the contours of pearls in contact with each other,the watershed algorithm is used to segment the pearl image,and the individual pearl individuals are obtained,and then the pearls are located through the connected domain mark and the centroid algorithm.According to national standards on pearl shape,a quantitative model of pearl shape parameters is established based on the information of the pearl image.Experimental results show that the detection error of pearls of different shapes is within 0.63%,the shape statistics accuracy is 100%,and the algorithm takes 24 ms.

关 键 词：机器视觉 珍珠形状 分水岭算法 轮廓特征 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]