Jun Tie

Papers

Total Citations

H-Index

About

Jun Tie is a researcher whose work sits at the intersection of computer vision and precision agriculture, with a particular focus on image segmentation for fruit recognition. His most influential contribution to date is the development of a modified UNet3+ algorithm specifically designed to improve the segmentation of green walnuts in natural, complex environments. This work directly addresses critical challenges in automated agricultural systems, where traditional object detection algorithms often suffer from missed or false detections due to variable lighting, occlusions, and background clutter. By integrating channel and spatial attention mechanisms into the UNet3+ architecture, Tie’s method significantly enhances the model’s ability to distinguish target objects from their surroundings, achieving a notable 8 citations since its 2024 publication. This research holds practical promise for advancing robotic harvesting and yield estimation technologies. Tie’s work exemplifies how deep learning can be tailored to solve domain-specific problems in agriculture, making a tangible impact on the efficiency and accuracy of automated fruit recognition systems.