Optimizing Automotive Component Assembly Inspection via an Advanced Detection Model Based on Faster R-CNN

In the field of automotive manufacturing, the precision of assembly processes is critical for ensuring product quality and efficient cost management.To address the inefficiency and unreliability of traditional manual inspection techniques,this study proposes an advanced detection model based on the Faster R-CNN framework for evaluating the assembly status of eight critical components within a vehicle's engine bay.Firstly, deformable convolutions(DCN) are incorporated into the backbone network to enhance adaptability to objects of varying scales.Secondly, a Balanced Feature Pyramid Network (BFPN) is adopted to improve feature integration and minimize information loss across different feature levels.Lastly, a Dynamic Label Assignment (DLA)mechanism is introduced, which adjusts the Intersection over Union (IoU) threshold to improve the quality of positive sample classification.Experimental results demonstrate that the improved model achieves a mean Average Precision (mAP) of 87.1%, representing a 2.3% improvement over the baseline Faster R-CNN model.