Automatic Inspection of Aeronautical Mechanical Assemblies by Matching the 3D CAD Model and Real 2D Images.
Automatic Inspection of Aeronautical Mechanical Assemblies by Matching the 3D CAD Model and Real 2D Images.
复制标题
DOI：
10.3390/jimaging5100081
复制
发表时间：
2019-10-19
影响因子：
3.2
通讯作者：
Brèthes L
中科院分区：
文献类型：
--
作者：
Ben Abdallah H;Jovančević I;Orteu JJ;Brèthes L
在航空工业中，自动化检测对于确保生产质量至关重要。它可以加速零件或机械组件的质量控制程序。因此，对旨在确保生产线高质量的智能视觉检测系统的需求正在增加。在这项工作中，我们解决了质量控制中的一个非常常见的问题。问题是验证是否存在正确的零件及其位置。我们从两个部分解决这个问题：首先，在检查过程开始之前自动选择信息丰富的视点（检查的离线准备），第二，通过将所述视点获取的图像与3D CAD中的信息进行匹配来自动处理这些图像模型启动。我们应用该检测系统检测航空机械组件的缺陷，目的是检查所有子部件是否存在并正确安装。该系统可在制造或维护操作期间使用。在两种平台上对系统的精度进行了评估。一个是自主导航机器人，另一个是手持平板电脑。实验结果表明，我们提出的方法是准确的，具有实时检测的可能性，适合工业应用。
In the aviation industry, automated inspection is essential for ensuring quality of production. It allows acceleration of procedures for quality control of parts or mechanical assemblies. As a result, the demand of intelligent visual inspection systems aimed at ensuring high quality in production lines is increasing. In this work, we address a very common problem in quality control. The problem is verification of presence of the correct part and verification of its position. We address the problem in two parts: first, automatic selection of informative viewpoints before the inspection process is started (offline preparation of the inspection) and, second, automatic treatment of the acquired images from said viewpoints by matching them with information in 3D CAD models is launched. We apply this inspection system for detecting defects on aeronautical mechanical assemblies with the aim of checking whether all the subparts are present and correctly mounted. The system can be used during manufacturing or maintenance operations. The accuracy of the system is evaluated on two kinds of platform. One is an autonomous navigation robot, and the other one is a handheld tablet. The experimental results show that our proposed approach is accurate and promising for industrial applications with possibility for real-time inspection.