Mingkang Peng

Papers

Total Citations

H-Index

About

Mingkang Peng is a researcher at the forefront of precision agriculture and computer vision, with a primary focus on deep learning-based object detection for weed management. His most influential work, "Weed detection with Improved YOLOv7" (2023), has already garnered 13 citations, demonstrating its early impact in the field. In this study, Peng addresses the critical challenge of detecting weeds in complex field backgrounds by enhancing the YOLOv7 architecture. His methodology involves robust online data augmentation to improve model generalization, followed by systematic improvements in feature extraction, feature fusion, and feature point judgment. This work represents a significant step toward automated, real-time weed identification, which is essential for reducing herbicide use and promoting sustainable farming. Peng’s contributions lie at the intersection of agricultural engineering and artificial intelligence, offering practical solutions for smart farming. His research not only advances the technical capabilities of deep learning models in unstructured environments but also provides a scalable framework for precision weed control. As a rising voice in applied computer vision, Peng’s work is poised to influence both academic research and agricultural technology development.