Research indexJournal article
Car Defect Detection Using Image Processing
Durai, Sujithra, Charan and 2 others
5 Jan 2023
NobleIDark:/48914/w1/20260525/2D0D1531DOI10.1109/iceconf57129.2023.10083542
Car Defect Detection Using Image Processing
Durai, Sujithra, Charan and 2 others · 2023 · Subscription
NobleIDark:/48914/w1/20260525/2D0D1531DOI10.1109/iceconf57129.2023.10083542
The full text is at the publisher
You can read the abstract here, or open it at the publisher.
Abstract
Defect Detection became one of the hardest problems in this world nowadays. As of now humans detected the defects with the help of their naked eye. This process is time consuming and this can be done by only experts and these decisions may be differ from expert to expert. To avoid this problem we are training the collected data set. This will consume very less time and provides the best solution. The detection of defects is essential in quality control in manufacturing to get high quality products in less time. Humans can identify large defects but in case of micro defects humans can't found those so, this implementation will helpful in finding large as well as micro and milli defects as well. In this project we are trying to detect the defects such as dents and ripples present on the surface of the car which are macro in size at the time of manufacturing the cars and in the case of accidents happened. We'll use a high-quality camera to simultaneously capture photographs of the vehicle's five sides. We will use Convolution Neural Network [CNN] for training (which highlights the location of the defects). Once trained, that model we can use it for different end user applications.