Share
Export Citation
UAV-based Clove Maturity Detection using YOLOv11n with Adaptive Gamma Correction and Histogram Equalization
Armayasari D.
Proceedings of the 6th International Conference on Pervasive Computing and Social Networking Icpcsn 2026
Abstract
Determining clove fruit maturity is an important step in the harvesting process because it directly affects product quality and market value. However, UAV-based monitoring in plantation environments often faces non-uniform illumination caused by canopy shadows and varying outdoor lighting conditions, which may reduce detection reliability. This study investigates the use of Adaptive Gamma Correction and Histogram Equalization as image preprocessing techniques for YOLOv11n-based clove maturity detection using UAV imagery. Four experimental configurations were evaluated, namely original images, Adaptive Gamma Correction, Histogram Equalization, and the combined Adaptive Gamma Correction plus Histogram Equalization approach. Model performance was assessed using precision, recall, mAP@0.5, and mAP@0.5 to 0.95 for the ripe and unripe classes. The results show that the combined preprocessing strategy produced the most balanced overall detection performance, with particularly strong results for the unripe class, achieving a precision of 75.5 percent, an mAP@0.5 of 85.4 percent, and an mAP@0.5 to 0.95 of 53.4 percent. In addition, the same configuration achieved the highest precision for the ripe class at 81.9 percent. These findings indicate that combining brightness and contrast enhancement can improve clove maturity detection under uneven lighting conditions in UAV imagery.