Investigation on automated inspection of highly reflective automotive parts

高反射汽车零部件自动检测研究

• 批准号: 521317-2017
• 负责人: Mirhassani, Mitra
• 金额: $ 1.82万
• 依托单位: University of Windsor
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2017
• 资助国家: 加拿大
• 起止时间: 2017-01-01 至 2018-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=640058
• 关键词: Investigation; automated; inspection; highly; reflective

This project investigates the development of an automated non-contact inspection station for quality control ofhighly reflective plastic and chrome-based parts. The developed system is going to be based on machine visionsystems using Ultra Violet light. The inspected parts are used in various sections of the vehicle and are requiredto be clear of any defects.The defects are surface faults, rather small in size, and are hard to detect. The current procedure in place isbased on the manual inspection of each part, which is performed in specific lighting stations on themanufacturing floor. Inspected parts are packaged carefully and shipped to their destination. These parts gointo a second round of manual inspection, and there are always faulty parts within the packages., which arerejected. These rejects require the Canadian manufacturer to pay fines and increases their productions costs.The size of the faults, shape of the parts, and the fact that their surface is highly reflective, results in mistakesduring the inspection. Moreover, despite the operator's experience, since the method is based on the judgmentof the operator at that instance, a change of operator results in different decisions regarding the pass/fail of apart. Therefore, a reliable and automated method of inspection can not only increase the efficiency of thesystem, but also by reducing human error it can result in cost saving in shipping and secondary complaintsfrom the overseas assembly plants.The typical faults are scratch marks, cracks, and surface imperfections, which are highly irregular. Using anautomated, machine vision system, allows the manufacturer to quantify and classify the degree of the defects.

该项目研究了用于高反光塑料和铬基零件质量控制的自动化非接触检测站的开发。开发的系统将基于使用紫外光的机器视觉系统。被检测的零件用于车辆的各个部位，要求没有任何缺陷。缺陷是表面缺陷，体积较小，不易检测。目前的程序是基于对每个部件的人工检查，这是在制造楼层的特定照明站执行的。检验过的零件会被仔细包装，然后运往目的地。这些零件将进入第二轮人工检验，而且包装内总是有故障零件，因此被拒绝。这些不合格品要求加拿大制造商支付罚款并增加他们的生产成本。缺陷的大小、部件的形状以及它们的表面高度反光的事实导致在检查过程中出现错误。此外，尽管操作员有经验，但由于该方法是基于操作员在该情况下的判断，操作员的改变会导致关于分离的通过/失败的不同决定。因此，一种可靠和自动化的检测方法不仅可以提高系统的效率，还可以通过减少人为错误来节省运输成本和海外组装厂的二次投诉。典型的故障是划痕、裂纹和表面缺陷，这些缺陷非常不规则。使用自动化的机器视觉系统，允许制造商对缺陷的程度进行量化和分类。