Wenwu Zheng

Dr. Wenwu Zheng


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
21 August 2019 in Sustainability
Verified
Journal Article
Open access
PDF file
LLLei Luo
Lei Luo
XWXinyuan Wang
Xinyuan Wang
JLJie Liu
Jie Liu
WZWenwu Zheng
Wenwu Zheng
JZJing Zhen
Jing Zhen
HWHong Wan
Hong Wan
This paper provides a brief history review of the use of ancient weirs in fishing on our planet, as well as a pilot study that involves investigating and mapping the coastal heritage of ancient stone tidal weirs (STWs) in the Penghu Archipelago which is located in the Taiwan Strait. The spatial distribution and morphological features of STWs across Penghu Archipelago were investigated and analyzed using very high-resolution (VHR) and freely available Google Earth (GE) imagery and geographic information system (GIS) analysis tools. A total of 539 ground-truthed STWs were identified from multiple temporal GE images, and these accounted for over 90% of the localized inventory databases. The proposed GE-based method was found to be more efficient, timely and effective compared to field and airborne surveys. This paper illustrates the utility of GE as a source of freely available VHR remote sensing imagery for archaeological surveys and heritage sustainability in coastal areas.

9 August 2026 in Sustainability
Verified
Journal Article
Open access
PDF file
YXYao Xiao
Yao Xiao
LYLiu shi Yu
Liu shi Yu
WZWenwu Zheng
Wenwu Zheng
Accurate mapping of terraces is important for farmland protection, food security, and the management of terrace abandonment. However, extracting terraces from high-resolution remote sensing imagery remains challenging because of complex backgrounds, fragmented ridge structures, weak boundary contrast, and high similarity between terraces and surrounding land-cover types. This study proposes AMF-Deep, an Attention Multi-scale Fusion DeepLabv3+ model, for binary terrace extraction across representative terrace regions and morphologies. AMF-Deep was evaluated using 0.61 m high-resolution remote-sensing imagery from five representative terrace regions in China, encompassing diverse terrace morphologies and background conditions. In the proposed model, MobileNetV2 is adopted as the backbone to reduce computational cost, a modified multi-scale ASPP module is used to capture terrace morphology at different spatial scales, CBAM enhances low-level boundary and texture features, ECA refines high-level channel responses before ASPP, and NAM recalibrates semantic features after ASPP. A residual decoder is further introduced to support low-level feature transfer and boundary recovery. Experimental results show that AMF-Deep achieved the highest regional performance in Huangling, with mIoU, mPA, Accuracy, Recall, and Precision of 89.47%, 95.17%, 96.70%, 94.99%, and 93.68%, respectively. Stepwise ablation experiments further indicated that the modified ASPP module, the differentiated attention arrangement, and the residual decoder contributed to the final performance. Qualitative seasonal visualization suggests that winter images and dryland terraces in Shexian remain more challenging because of weak texture contrast and high background similarity. These results indicate that AMF-Deep has potential for terrace extraction across representative regions, although further quantitative seasonal evaluation and cross-region validation are still needed.