ADNet-initialized YOLO11s for high-precision aircraft surface damage detection

This paper aims to improve the accuracy and efficiency of aircraft surface damage detection by enhancing the YOLO11s algorithm and conducting a comparative analysis of the performance before and after the improvement. The study proposes an improved YOLO11s model, utilizing ADNet as the initialization layer. This enhancement optimizes feature extraction, information fusion, and model expressiveness, significantly improving detection performance. Experimental results show that the improved model outperforms the baseline model in terms of accuracy, with specific metrics as follows: precision of 97.9%, recall of 94%, mAP50 of 97.4%, and mAP50-95 of 87.6%. The experiments demonstrate that, compared to the baseline model, the improved model exhibits higher accuracy and superior efficiency in aircraft surface damage detection, providing strong support for the application of intelligent aircraft surface damage detection technology.