Zhi Zheng

Zhi Zheng

Auto-generated profile, based on publicly available information

Zhi Zheng

Department of Geography and Resource Management, The Chinese University of Hong Kong, Hong Kong

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 January 2024 in International Journal of Applied Earth Observation and Geoinformation
Verified
Journal Article
Open access
YZYongjun Zhang
Yongjun Zhang
WZWenfei Zhang
Wenfei Zhang
YYYongxiang Yao
Yongxiang Yao
ZZZhi Zheng
Zhi Zheng
YWYi Wan
Yi Wan
MXMingtao Xiong
Mingtao Xiong

9 January 2024 in The Photogrammetric Record
Verified
Journal Article
Closed access
ZZZhi Zheng
Zhi Zheng
YWYi Wan
Yi Wan
YZYongjun Zhang
Yongjun Zhang
ZHZhonghua Hu
Zhonghua Hu
DWDong Wei
Dong Wei
RXRang Xiao
Rang Xiao

12 June 2024 in International Journal of Applied Earth Observation and Geoinformation
Verified
Journal Article
Open access
ZZZhi Zheng
Zhi Zheng
PMPeifeng Ma
Peifeng Ma
ZWZherong Wu
Zherong Wu
Identifying damaged areas after earthquakes is essential for emergency rescue and rebuilding efforts. Remote sensing image change detection can facilitate these tasks by analyzing features differences between bitemporal/multi-temporal optical images. However, the existing methods still face challenges due to the unpredictable timing, locations, and magnitudes of earthquakes. Firstly, the spectral features of optical images vary over time, making it difficult to differentiate actual damage from areas appearing damaged due to spectral noise. Secondly, variations in earthquake characteristics can cause different types and scales of damage, from individual buildings in urban areas to extensive landslides in mountain areas. Overcoming these variations to establish capabilities for detecting multiple types of damage is also critical. To tackle these challenges, this study introduces a change detection network based on the decoupling of context-structural features (referred to as CSDNet). CSDNet uses a CNN-transformer architecture and decouples image features into structural and context branches through a feature difference extractor. Subsequently, the structural branch extracts reliable features through multi-scale consistency feature fusion, ensuring robust damage detection across multiple scales. Context branch facilitates long-range supervision and reduces the impacts of pseudo-changes through partial feature exchange between bitemporal images. Furthermore, to enhance CSDNet’s adaptability across different scenarios and explore CSDNet’s potential for solving multiple tasks within a unified framework, an adaptive module has been designed to dynamically fuse structural and context features. Experiments were conducted on two representative tasks: individual building damage with partial rebuilding and large-scale landslides, to investigate CSDNet’s capabilities. As a result, CSDNet achieved F1-Scores of 92.2% and 90.2% on the two tasks respectively, indicating the effectiveness of CSDNet.

1 August 2025 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
YFYouran Fu
Youran Fu
ZWZherong Wu
Zherong Wu
ZZZhi Zheng
Zhi Zheng
QZQi Zhu
Qi Zhu
YGYating Gu
Yating Gu
Mei-Po KwanMei-Po Kwan
Mei-Po Kwan
Remote sensing-based urban change detection has seen substantial progress with the development of deep learning. Nevertheless, its performance under low-light imaging conditions, such as nighttime, twilight, and cloudy or rainy weather, remains limited due to image degradation, reduced contract, and weak spatial features. To address these challenges, we propose MambaLCD, a novel change detection method that integrates a Siamese visual state space encoder, a bi-temporal feature fusion module, and a mask decoder. The encoder captures spatial dependencies via multi-directional state-space scanning, while the fusion module enhances semantic consistency through illumination-aware recalibration. The decoder integrates multi-scale features using attention-inspired pooling for precise pixel-level change masks. Comprehensive experiments on three benchmark datasets including S2Looking, WHU-CD and LEVIR-CD demonstrate the effectiveness of Mamba-LCD. Specifically, on the low-lightdominated S2Looking dataset, Mamba-LCD outperforms 10 state-of-the-art CNN-, Transformer-, and Mamba-based methods, surpassing the previous best results by 0.73 and 1.58, respectively. Moreover, it maintains competitive performance on LEVIR-CD and WHU-CD under normal lighting. We further conducted ablation studies to evaluate the impact of different activation functions and fusion strategies. The results suggest the robustness and adaptability of Mamba-LCD, indicating its potential for urban monitoring across diverse illumination scenarios.

23 April 2026 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
ZHZhonghua Hu
Zhonghua Hu
XLXinyi Liu
Xinyi Liu
YWYi Wan
Yi Wan
ZZZhi Zheng
Zhi Zheng
XHXuejun Huang
Xuejun Huang
YZYongjun Zhang
Yongjun Zhang
Reconstructing 3D scenes from a single satellite image remains challenging because reliable metric depth is unavailable and many regions are occluded, which limits the geometric observability and appearance evidence required for faithful reconstruction. To address these issues, we propose SSRecon, a diffusion-driven framework for single-image satellite 3D reconstruction via iterative RGB-D completion. SSRecon fits the RPC model to a pinhole approximation, estimates depth with a language-conditioned diffusion model, and performs RGB and depth inpainting using Stable Diffusion and a step-segmented gradient-guidance scheme, respectively. The input image together with the estimated depth forms the first RGB-D frame to initialize a 3D Gaussian Splatting representation, which is then refined by repeatedly rendering novel views, completing missing regions, and updating the scene representation. Experiments on DFC2019 and SpaceNet4 show that SSRecon improves perceptual realism and cross-view consistency across diverse target poses without auxiliary semantics or height inputs. In the pure single-image setting, SSRecon increases CLIP-Score by 0.65 on DFC2019 and 0.76 on SpaceNet4 over the strongest baseline, with consistent gains in CLIP-IQA quality and sharpness.

14 October 2026 in ISPRS Journal of Photogrammetry and Remote Sensing
Verified
Journal Article
Closed access
XHXuejun Huang
Xuejun Huang
YWYi Wan
Yi Wan
LYLei Yu
Lei Yu
XLXinyi Liu
Xinyi Liu
ZZZhi Zheng
Zhi Zheng
YZYongjun Zhang
Yongjun Zhang