Vehicle underbody small parts detection using Computer vision

Manual installation of vehicle underbody small parts is tedious task which sometimes results in incomplete & inaccurate installation. Based on process quality guidelines, this comes under potential defect category at End of line inspection area for which few hours of manual efforts are required for identifying location of error & doing rework activity. Existing deep learning & image comparison method falls short in identifying error location. To overcome this challenge, deep learning algorithm with co-ordinate system developed which comprises of identifying plurality of entities present in test and reference image and indexing it as a small part & a hole. This further comprises of determining 2D position in terms of X and Y co-ordinate of each indexed hole & small parts between reference & test image. By seamless integration of deep learning-based part detection with coordinate constraints logic, our system enables real-time identification of incorrectly assembled parts or the absence of parts within the vehicle assembly, thereby enhancing the overall quality control process. YOLO V8 algorithm is employed as the primary tool for part detection, offering robust performance in real-time object detection tasks. Furthermore, we augment this algorithm with a logic framework designed to validate the presence and correct placement of the detected parts within the vehicle assembly. This logic leverages the coordinate constraints to determine if the identified part is appropriately positioned according to predefined specifications. This approach results in precise comparison & identification of error in terms of missing or misplacing of small parts which also offers significant saving in manual installation and rework efforts.