Yan Tong

Mr. Yan Tong


Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

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

Representative Publications

View All Publications
15 June 2026 in Sustainability
Verified
Journal Article
Open access
PDF file
WWwen Wei
wen Wei
DLDan Liang
Dan Liang
Yan TongYan Tong
Yan Tong
TLTong Li
Tong Li
CLChenyu Lyu
Chenyu Lyu
WCWuxue Cheng
Wuxue Cheng
This study investigates the spatiotemporal evolution of ecological environment quality and its driving mechanisms in the Sichuan section of the Yellow River Basin using Landsat imagery from 2000 to 2023. The Remote Sensing Ecological Index (RSEI) was constructed on the Google Earth Engine platform, and a comprehensive evaluation model was developed using principal component analysis. Sen’s slope, the Mann–Kendall test, and the Hurst exponent were applied to assess temporal trends and future persistence, while the optimal parameter-based Geodetector model was used to identify the driving factors of spatial differentiation. Results show that: (1) ecological environment quality exhibits a fluctuating but overall increasing trend, with a multi-year mean RSEI of 0.58, indicating a transition from “moderate” to “good–excellent” conditions; (2) spatially, ecological quality demonstrates significant heterogeneity and clear altitudinal gradients, with better conditions in the northwest than in the southeast, where low- and mid-altitude areas show higher ecological quality and stronger improvement, whereas high-altitude areas remain relatively poor due to strong natural constraints; (3) the spatial differentiation is jointly driven by multiple factors, among which precipitation and temperature are dominant, elevation exerts a fundamental constraint, and human activity plays a relatively minor role, while the interaction between climate and topographic factors shows the strongest explanatory power. These findings provide insights into the evolution and drivers of ecological environment quality in high-altitude regions and support ecological protection and regional management in the upper Yellow River Basin.

23 April 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
Yan TongYan Tong
Yan Tong
Kongwen Frank ZhangKongwen Frank Zhang
Kongwen Frank Zhang
WCWuxue Cheng
Wuxue Cheng
Jane LiuJane Liu
Jane Liu
Individual tree classification has a long history of diverse development, with recent trends focusing on the adoption of machine learning and deep learning approaches. A simple and powerful approach that lets the model auto-pilot, but weakens the need for physical characteristic understanding. Over more than a decade of our research, we have focused on establishing a direct representation of individual trees that bridges 2D top-down imagery and true 3D models. In this study, we investigated the fundamental question of the influence of the input data on these ML/DL models. In 2024, we introduced a novel data transformation method, the Pseudo Tree Crown (PTC), which provides a pseudo-3D pixel-value perspective that enhances the informational richness of images and significantly improves classification performance. Our original implementation was successfully tested on urban and deciduous trees in 2024 and was later extended to Canadian natural conifer species under snow conditions in 2025. However, the original PTC relied on the green band, limiting its applicability to green-leaf species. In this study, we analyzed and compared the performance of different data variations and transformations, such as the Green–Red Vegetation Index (GRVI) and Principal Component Analysis (PCA), as direct input and used their PTC forms. Classifications were conducted using Random Forest, ResNet50, and YOLOv10. The results confirmed the effectiveness of the PTC, which consistently improves classification accuracy by at least 7% without introducing additional computational time or complexity. Furthermore, PTC exhibits robust, consistent behaviour across all data forms, demonstrating its strong resilience and reliability.

4 June 2026 in Remote Sensing
Verified
Journal Article
Open access
PDF file
Yan TongYan Tong
Yan Tong
Kongwen Frank ZhangKongwen Frank Zhang
Kongwen Frank Zhang
WCWuxue Cheng
Wuxue Cheng
Jane LiuJane Liu
Jane Liu
Individual tree classification has a long history of diverse development, with recent trends focusing on the adoption of machine learning and deep learning approaches. It is a simple and powerful approach that allows the model to auto-pilot while reducing the need for physical characteristic understanding. Over more than a decade of research, we have focused on establishing a direct representation of individual trees that bridges 2D top-down imagery and true 3D models. In this study, we investigated the fundamental question of the influence of the input data on these ML/DL models. In 2024, we introduced a novel data transformation method, the Pseudo Tree Crown (PTC), which provides a pseudo-3D pixel-value perspective that enhances the informational richness of images and significantly improves classification performance. Our original implementation was successfully tested on urban and deciduous trees in 2024 and was later extended to Canadian natural conifer species under snow conditions in 2025. However, the original PTC relied on the green band, limiting its applicability to green-leaf species. In this study, we analyzed and compared the performance of different data variations and transformations, such as the Green–Red Vegetation Index (GRVI) and principal component analysis (PCA), as direct input and used their PTC forms. Classifications were conducted using Random Forest (RF), ResNet50, YOLOv10 and Segment Anything (SA). The results confirmed the effectiveness of the PTC, which consistently improves the classification accuracy by at least 5% without introducing additional computational time or complexity. Furthermore, PTC exhibits robust, consistent behavior across all data forms, demonstrating its strong resilience and reliability.