Defect inspection of coated automobile roofs using a single camera

Abstract

Automobile surface defects, such as scratches and dents, can largely affect the first impressions of consumers. These defects are likely to occur during the processes of manufacturing and painting. Most global automobile companies still rely on visual inspection to detect the defects, which results in instability and inefficiency in the inspection procedure. In this paper, a low‐cost detection system is proposed. The hardware part of the system consists of only light emitting diode (LED) tubes and a single camera. The software part of the system first obtains defect candidates greedily by thresholding binarization result. Then, candidates in each frame are filtered using a differential image. Finally, as the main contribution, our algorithm further filters the candidates using tracking trajectories of defect candidates over multiple frames. Experimental results demonstrate that our method is able to detect defects while suppressing false‐positive detections and is effective in real‐world situations. © 2020 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Dirt in Automotive Painting: An Approach Supported Bycyber-Physical Systems for Process Automation

“…The technology used is based on a machine vision system to capture an image combined with a classifier to identify defects. [7] was developed from a visual inspection system and the corresponding algorithm for the detection of paint defects in the automotive industry; [8] studied the defect inspection of coated automobile roofs using a single camera; [9] investigated a machine vision detection method for surface defects of automobile stamping parts; [6] implemented a system to detect defects at one paint shop with artificial vision. Other areas of industry have also developed quality monitoring systems based on a similar principle: [10] proposed an automatic section system based on vision machine for bearing surface and [11] developed a defect detection method for a rail surface based on line-structured light.…”

“…orange peel, overspray, crater, solvent, bubbles), while [17] identifies craters, dirt, stripes, humidity marks, hair, and drops of various types on non-flat surface. [8,10,17] These articles make it clear that it is very difficult to obtain adequate responses between dark and metallic colors, as well as the influence of lighting in obtaining the image and the generation of pseudo-defects.…”

“…The research has shown that the effects of illumination and different types of colors have to be taken into consideration during the definition of best method of image acquisition. Articles [8,10,17,18] also showed that there already is technology able to identify many types of paint defects using non-destructive processes using neural networks. However, these solutions focused on helping operators to improve the accuracy of inspection of vehicles at final stages of the paint process, operating online and with huge frames.…”

Intelligent Vehicle Inspection Tool Design Based on Freeman Chain Code for Automatic Annotation of 3D Models

“…Framework of Automobile Inspection Tools. The boost of the automotive manufacturing industry not only requires continuous improvement in product quality, structural design, and processing methods, but also puts forward stricter requirements for the design of inspection fixture structures [15]. At present, traditional inspection tools are difficult to meet the requirements of rapid automobile manufacturing, which has a serious impact on its production efficiency.…”

Sifting Method of Defect Candidate on Coated Automobile Roofs based on Binarization and Brightness Difference

“…A patent (22) uses only one camera and firstly binarizes the image to get defect candidates, then separates noise by using the time-axis information and moving distance of the defect candidates between two frames. Research (23) proposes defect motion trajectory detection with higher robustness to separate noise. Also, this research uses a sifting method, which combines binarization and using differential image to get defect candidates at first.…”