Papers
2
Total Citations
7
H-Index
2
About
Tan He’s research focuses on computer vision and image processing for underground infrastructure inspection, particularly addressing the challenging conditions of drainage pipeline environments. His major contributions center on developing specialized dehazing algorithms to restore clarity in pipeline inspection videos and images, which are often severely degraded by water fog, darkness, and haze. He pioneered a pipeline image haze removal system using a dark channel prior on a cloud processing platform (2020, 5 citations), enabling real-time enhancement for pipeline robots. Additionally, he proposed a novel dehazing algorithm based on the atmospheric scattering model combined with a multi-scale Retinex strategy (2019, 2 citations), which effectively mitigates blurring caused by water mist in detection videos. Although his citation counts are modest, his work addresses a critical real-world bottleneck in automated pipeline fault detection—a key application for urban drainage security. By improving image definition in complex, low-visibility conditions, He’s algorithms directly enhance the reliability of pipeline robots, reducing the risk of missed defects. His research represents an important step toward practical, deployable vision systems for underground infrastructure maintenance.
Research Focus
Key Achievements
Top Papers
- 1
- 2