Changji Wen
Papers
1
Total Citations
22
H-Index
1
About
Changji Wen is a researcher whose work sits at the intersection of computer vision and agricultural automation, with a primary focus on developing efficient, lightweight deep learning models for precision fruit detection in complex orchard environments. Wen’s most impactful contribution to date is the 2023 paper “Improved YOLOv7-Tiny Complex Environment Citrus Detection Based on Lightweighting,” which has already garnered 22 citations. In this work, Wen addresses a critical real-world challenge: accurately detecting citrus fruits under difficult conditions such as variable lighting, branch occlusion, and fruit overlap. By proposing the YOLO-DCA model—an enhancement of the YOLOv7-tiny architecture—Wen introduces depth-separable convolution (DWConv) to replace standard convolutions, achieving a significant reduction in model complexity without sacrificing detection accuracy. This innovation makes the model suitable for deployment on resource-constrained edge devices, a key step toward practical, real-time agricultural monitoring. Wen’s research exemplifies how lightweight neural networks can bridge the gap between state-of-the-art AI and field-ready agricultural tools, offering a scalable solution for smart farming.
Research Focus
Key Achievements
Top Papers
- 1