Xiaohan Zhang

Title: Xiaohan Zhang and RGB-T Multispectral Pedestrian Detection

Introduction

Xiaohan Zhang is an inventor associated with Zhejiang University in Hangzhou, China. Her patent work focuses on multispectral pedestrian detection, combining visible-light and thermal infrared imagery to help identify pedestrians and determine their locations.

Latest Patents

Zhang is associated with 1 patent: “RGB-T Multispectral Pedestrian Detection Method Based on Target Aware Fusion Strategy.” The method extracts features from a pair of RGB and thermal infrared images, then fuses those features using a target-aware strategy. A detection head processes the fused features and outputs pedestrian bounding boxes and confidence scores. By strengthening features associated with pedestrians while suppressing irrelevant background information, the approach aims to improve detection accuracy.

Career Highlights

Zhang’s listed invention applies target-aware feature fusion to pedestrian detection using complementary visual and thermal data. This approach is designed to make better use of information from both image types.

Collaborations

The provided information lists Huiliang Shen and Xue Zhang as Zhang’s coworkers. It does not specify their roles in the patent or the nature of their collaboration.

Conclusion

Xiaohan Zhang’s patent work at Zhejiang University addresses multispectral pedestrian detection through a fusion strategy intended to emphasize pedestrian features and reduce background noise.