Liyan Zhang

Nanjing University of Aeronautics and Astronautics

Papers

Total Citations

H-Index

About

Liyan Zhang’s research lies at the intersection of computer vision, photogrammetry, and intelligent robotics, with a particular focus on feature point recognition and spatial estimation under challenging conditions. In their most cited work, "Design of Chinese Character Coded Targets for Feature Point Recognition Under Motion-Blur Effect" (2020, 9 citations), Zhang introduced a novel coded target system that maintains reliable identification even in degraded imagery—a critical advancement for applications in close-range photogrammetry, robot navigation, and augmented reality. This work directly addresses a persistent bottleneck in active visual feature tracking. Zhang’s earlier foundational contribution, "Constructing spherical curves by interpolation" (2006, 6 citations), demonstrates a sustained interest in geometric modeling. More recently, their 2024 paper on "Accurate Robot Arm Attitude Estimation Based on Multi-View Images and Super-Resolution Keypoint Detection Networks" (3 citations) proposes SRKDNet, a two-stage framework that leverages subpixel convolutions for high-precision pose estimation from multiple views. This work signals a forward-looking shift toward deep learning-enhanced spatial reasoning for industrial automation. With a career spanning geometric interpolation to deep learning for robotics, Zhang’s contributions are steadily building a bridge between classical photogrammetry and modern intelligent systems.