Peihong Zhu

GE Global Research (United States), Xi'an Jiaotong University

Papers

Total Citations

H-Index

About

Peihong Zhu is a researcher at the intersection of robotics, computer vision, and non-destructive testing (NDT), with key contributions in automated inspection systems and robust face recognition. Zhu’s most notable work, “Multi-Robot System for Automated Fluorescent Penetrant Indication Inspection with Deep Neural Nets” (2021, 12 citations), addresses a critical bottleneck in aerospace manufacturing by replacing manual fluorescent penetrant inspection (FPI) with a multi-robot framework powered by deep neural networks. This innovation automates the detection of defect-related indications, enhancing speed and reliability in quality control—a significant leap for industrial NDT. Earlier, Zhu’s research on “An Improved Robust Sparse Coding for Face Recognition with Disguise” (2012, 5 citations) tackled the challenging problem of recognizing faces under occlusion or disguise, advancing sparse representation-based classification (SRC) for robotic vision systems. By improving robustness in real-world conditions, this work laid groundwork for more reliable autonomous perception. With a focus on bridging deep learning and practical robotics, Zhu’s contributions demonstrate clear impact in both aerospace inspection and computer vision, offering scalable solutions that push the boundaries of automated visual inspection and human-robot interaction.