Automated Inspection Method for Painted Surfaces.
Automated Inspection Method for Painted Surfaces.
复制标题
涂漆表面的自动检测方法。
DOI：
10.1299/kikaic.60.3201
复制
发表时间：
1994
期刊：
影响因子：
--
通讯作者：
Kazuto Sasanishi
中科院分区：
文献类型：
--
作者：
Tanaka Kazumoto;Shinbara Yoshimi;H. Ikeda;Naoki Yamada;Hiroshi Kiba;Kazuto Sasanishi
汽车外观的光洁度是直接影响消费者购买决策的重要因素。然而，虽然近年来油漆质量有所提高，但在油漆表面的检查和修复过程中需要高度发达的视觉和触觉，操作人员日益沉重的负担已成为一个主要问题。本文介绍了一种自动化检测方法在马自达开发。这种新方法可以感知油漆表面上凹凸缺陷的独特反射，以便识别修复。该系统由辐射强度渐变的近紫外辐射源、视频图像传感器和图像处理器组成。使用该系统，可以一致地检测缺陷并确定它们是凸还是凹。
The finish of an automobile exterior is an important factor directly affecting a customer's purchase decision. However, while paint quality has increased in recent years, highly developed visual and tactile senses have been required in the process for painted surfaces inspection and repair, and the increasingly heavy burden on operators has become a major problem. This paper describes an automated inspection method developed at Mazda. The new method senses the distinctive reflection of convex and concave defects on painted surfaces in order to identify for repair. The corresponding system consists of a near-ultraviolet radiation emitting source with gradated radiant intensity, a video image sensor, and an image processor. Using this system, it is possible to detect defects consistently and determine whether they are convex or concave.