Fan Fan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Fan Fan has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
Chengzhi Fan
2 shared publications- Fusuo Zhang
China Agricultural University
2 shared publications Pengxiang Yan
2 shared publicationsQiang Xu
2 shared publicationsSiyu Jia
2 shared publicationsWeimin Guo
2 shared publications
Latest Publications
- Open Access
- 7 July 2026
- Published by Elsevier in Smart Agricultural Technology
- Vol. 15
Multi-metric clustering for feature abstraction in UAV hyperspectral estimation of tobacco leaf starch content
- 7 July 2026
- Published by Elsevier in Smart Agricultural Technology
- Vol. 15
Open AccessMulti-metric clustering for feature abstraction in UAV hyperspectral estimation of tobacco leaf starch content
- Open Access
- 13 February 2026
- Published by Elsevier in SSRN Electronic Journal
Multi-metric Clustering for Feature Abstraction in UAV Hyperspectral Estimation of Tobacco Leaf Starch Content
Tobacco leaf starch content is an important biochemical indicator for assessing leaf maturity and industrial processability, exerting a crucial influence on tobacco quality formation and curing characteristics. This study proposes a physiology-aware spectral feature abstraction framework based on Multi-metric Clustering Analysis (MMCA), integrating multi-metric bands evaluation with clustering analysis and combining UAV hyperspectral data with machine learning to achieve accurate and robust estimation of tobacco leaf starch content. Five feature importance metrics, namely correlation coefficient, grey relational analysis, variable importance in projection, mutual information, and Shapley Additive Explanations (SHAP), were applied to identify tobacco starch sensitive bands (695–705 nm). Unlike approaches that rely on a single sensitivity criterion predominantly confined to the red edge region, MMCA adopts a multi metric guided clustering strategy to construct a joint evaluation framework over the full spectral domain and extract cluster center bands (487, 564, 701, 735, and 792 nm) using cosine distance with average linkage, thereby reducing spectral redundancy while preserving biochemical sensitivity. Among four machine learning algorithms, the MMCA-KRR model exhibited the best performance (test R² = 0.86, RMSE = 4.24 %). The model effectively captured starch spatial–temporal variation across nitrogen treatments and growth stages, aligning with physiological trends of starch accumulation and conversion. With advancing growth, starch content generally increases from the early to the mid stages, then gradually stabilizes toward the late stage and may even decline slightly. This research demonstrates that integrating multi-metric evaluation clustering with nonlinear machine learning provides a robust and interpretable framework for high-accuracy estimation of tobacco leaf starch content using UAV hyperspectral data.
- 13 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMulti-metric Clustering for Feature Abstraction in UAV Hyperspectral Estimation of Tobacco Leaf Starch Content
Tobacco leaf starch content is an important biochemical indicator for assessing leaf maturity and industrial processability, exerting a crucial influence on tobacco quality formation and curing characteristics. This study proposes a physiology-aware spectral feature abstraction framework based on Multi-metric Clustering Analysis (MMCA), integrating multi-metric bands evaluation with clustering analysis and combining UAV hyperspectral data with machine learning to achieve accurate and robust estimation of tobacco leaf starch content. Five feature importance metrics, namely correlation coefficient, grey relational analysis, variable importance in projection, mutual information, and Shapley Additive Explanations (SHAP), were applied to identify tobacco starch sensitive bands (695–705 nm). Unlike approaches that rely on a single sensitivity criterion predominantly confined to the red edge region, MMCA adopts a multi metric guided clustering strategy to construct a joint evaluation framework over the full spectral domain and extract cluster center bands (487, 564, 701, 735, and 792 nm) using cosine distance with average linkage, thereby reducing spectral redundancy while preserving biochemical sensitivity. Among four machine learning algorithms, the MMCA-KRR model exhibited the best performance (test R² = 0.86, RMSE = 4.24 %). The model effectively captured starch spatial–temporal variation across nitrogen treatments and growth stages, aligning with physiological trends of starch accumulation and conversion. With advancing growth, starch content generally increases from the early to the mid stages, then gradually stabilizes toward the late stage and may even decline slightly. This research demonstrates that integrating multi-metric evaluation clustering with nonlinear machine learning provides a robust and interpretable framework for high-accuracy estimation of tobacco leaf starch content using UAV hyperspectral data.
- Open Access
- 1 January 2021
- Published by MDPI AG in Sustainability
- Vol. 13 (1) , 342
Evaluation of Sustainability of Irrigated Crops in Arid Regions, China
Ensuring national food security amidst ongoing economic development whilst both protecting the environment and reducing the dependence on fossil fuels are significant challenges for Chinese sustainable development. The main objectives of this study were to reveal irrigated crop (wheat, maize, and sunflower) performance in terms of energy, economic, and environmental aspects in China’s largest designed irrigation area, Hetao irrigation district (HID), and to evaluate agricultural suitability based on plant structure. An integrated indicator and comprehensively assessment method were used to evaluate the above objectives based on the results from in-person surveys. The results show that maize exhibits the best overall performance compared to two other major crops (wheat and sunflower), which supports the government policy of adjust and optimize the planting structure program (AOPST), an effective way to achieve the multiple-objectives for sustainable agricultural development. However, reducing fertilizer remains a more critical factor than AOPST. These study results provide useful guidance for policy-makers and relevant stake-holders both in a regional context for the HID and at the global agricultural governance and management level.
- 1 January 2021
- Published by MDPI AG in Sustainability
- Vol. 13 (1) , 342
Open AccessEvaluation of Sustainability of Irrigated Crops in Arid Regions, China
Ensuring national food security amidst ongoing economic development whilst both protecting the environment and reducing the dependence on fossil fuels are significant challenges for Chinese sustainable development. The main objectives of this study were to reveal irrigated crop (wheat, maize, and sunflower) performance in terms of energy, economic, and environmental aspects in China’s largest designed irrigation area, Hetao irrigation district (HID), and to evaluate agricultural suitability based on plant structure. An integrated indicator and comprehensively assessment method were used to evaluate the above objectives based on the results from in-person surveys. The results show that maize exhibits the best overall performance compared to two other major crops (wheat and sunflower), which supports the government policy of adjust and optimize the planting structure program (AOPST), an effective way to achieve the multiple-objectives for sustainable agricultural development. However, reducing fertilizer remains a more critical factor than AOPST. These study results provide useful guidance for policy-makers and relevant stake-holders both in a regional context for the HID and at the global agricultural governance and management level.
- Open Access
Long-Term Effect of Manure and Mineral Fertilizer Application Rate on Maize Yield and Accumulated Nutrients Use Efficiencies in North China Plain
Overuse of mineral fertilizer has become common at the North China Plain. Simultaneously, more organic manure resources are available for smallholder farmers. In order to increase the use of organic manure and reduce mineral fertilizer applications, a 10-year fertilization experiment with maize took place between 2008 and 2017. We assessed the long-term effects of cattle manure (CM) application and a complete nutrient substitution with mineral fertilizer (MF) at four application levels (3, 6, 9 and 12 t ha−1 CM) on yield, macronutrients (N, P and K) use efficiencies and soil conditions. Results showed that maize yields from CM and MF treatments differed across time and were significantly different in the first year of the experiment to no significant differences with increasing experimental time. In addition, increased MF levels did not result in increased maize yields; this response was different with CM applications. The highest 10-year maize average yield was 7.7 t ha−1 obtained with 9 t ha−1 of CM. Our results also showed that at the lowest application level (3 t ha−1 CM), the partial factor productivity (PFP) and the agronomic efficiency (AE) of all macronutrients were significantly higher with MF than with CM applications. Nevertheless, these differences narrowed with increased fertilizer input levels. The MF and CM recovery efficiency (RE) of N, P and K performed differently. Generally, MF exhibited significantly higher N-RE than CM treatments. CM treatments had significantly higher P-RE, but no K-RE differences were found between CM and MF. Soil available N, P and K significantly increased when fertilizer levels raised. MF treatments exhibited similar levels of soil available N, but lower soil available P and K compared with CM treatments.
Open AccessLong-Term Effect of Manure and Mineral Fertilizer Application Rate on Maize Yield and Accumulated Nutrients Use Efficiencies in North China Plain
Overuse of mineral fertilizer has become common at the North China Plain. Simultaneously, more organic manure resources are available for smallholder farmers. In order to increase the use of organic manure and reduce mineral fertilizer applications, a 10-year fertilization experiment with maize took place between 2008 and 2017. We assessed the long-term effects of cattle manure (CM) application and a complete nutrient substitution with mineral fertilizer (MF) at four application levels (3, 6, 9 and 12 t ha−1 CM) on yield, macronutrients (N, P and K) use efficiencies and soil conditions. Results showed that maize yields from CM and MF treatments differed across time and were significantly different in the first year of the experiment to no significant differences with increasing experimental time. In addition, increased MF levels did not result in increased maize yields; this response was different with CM applications. The highest 10-year maize average yield was 7.7 t ha−1 obtained with 9 t ha−1 of CM. Our results also showed that at the lowest application level (3 t ha−1 CM), the partial factor productivity (PFP) and the agronomic efficiency (AE) of all macronutrients were significantly higher with MF than with CM applications. Nevertheless, these differences narrowed with increased fertilizer input levels. The MF and CM recovery efficiency (RE) of N, P and K performed differently. Generally, MF exhibited significantly higher N-RE than CM treatments. CM treatments had significantly higher P-RE, but no K-RE differences were found between CM and MF. Soil available N, P and K significantly increased when fertilizer levels raised. MF treatments exhibited similar levels of soil available N, but lower soil available P and K compared with CM treatments.
- 1 July 2019
- Published by Elsevier in Science of The Total Environment
- Vol. 675, 642-650
Effects and relationships of grazing intensity on multiple ecosystem services in the Inner Mongolian steppe
- 1 July 2019
- Published by Elsevier in Science of The Total Environment
- Vol. 675, 642-650
Effects and relationships of grazing intensity on multiple ecosystem services in the Inner Mongolian steppe