Search

Your search keyword '"Shao, Guofan"' showing total 452 results
Start Over Author "Shao, Guofan"

Refine your results

more »

more »

more »

more »

more »
452 results on '"Shao, Guofan"'

9. Mapping Planted Forests in the Korean Peninsula Using Artificial Intelligence.

Author

Mitra, Ankita, Alvarez, Cesar Ivan, Abbasi, Akane O., Harris, Nancy L., Shao, Guofan, Pijanowski, Bryan C., Jahanshahi, Mohammad Reza, Gamarra, Javier G. P., Kim, Hyun-Seok, Kim, Tae-Kyung, Ryu, Daun, and Liang, Jingjing

Subjects
MACHINE learning, ARTIFICIAL intelligence, CONVOLUTIONAL neural networks, DATA augmentation, FOREST plants, DEEP learning
Abstract

Forests are essential for maintaining the ecological balance of the planet and providing critical ecosystem services. Amidst an increasing rate of global forest loss due to various natural and anthropogenic factors, many countries are committed to battling forest loss by planting new forests. Despite the reported national statistics on the land area in plantations, accurately delineating boundaries of planted forests with remotely sensed data remains a great challenge. In this study, we explored several deep learning approaches based on Convolutional Neural Networks (CNNs) for mapping the extent of planted forests in the Korean Peninsula. Our methodology involved data preprocessing, the application of data augmentation techniques, and rigorous model training, with performance assessed using various evaluation metrics. To ensure robust performance and accuracy, we validated the model's predictions across the Korean Peninsula. Our analysis showed that the integration of the Near Infrared band from 10 m Sentinel-2 remote sensing images with the UNet deep learning model, incorporated with unfrozen ResNet-34 backbone architecture, produced the best model performance. With a recall of 64% and precision of 76.8%, the UNet model surpassed the other pixel-based deep learning models, including DeepLab and Pyramid Sense Parsing, in terms of classification accuracy. When compared to the ensemble-based Random Forest (RF) machine learning model, the RF approach demonstrates a significantly lower recall rate of 55.2% and greater precision of 92%. These findings highlight the unique strength of deep learning and machine learning approaches for mapping planted forests in diverse geographical regions on Earth. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

13. Integrating forest structural diversity measurement into ecological research

Author

Atkins, Jeff W., primary, Bhatt, Parth, additional, Carrasco, Luis, additional, Francis, Emily, additional, Garabedian, James E., additional, Hakkenberg, Christopher R., additional, Hardiman, Brady S., additional, Jung, Jinha, additional, Koirala, Anil, additional, LaRue, Elizabeth A., additional, Oh, Sungchan, additional, Shao, Gang, additional, Shao, Guofan, additional, Shugart, H. H., additional, Spiers, Anna, additional, Stovall, Atticus E. L., additional, Surasinghe, Thilina D., additional, Tai, Xiaonan, additional, Zhai, Lu, additional, Zhang, Tao, additional, and Krause, Keith, additional

Published
2023
Full Text
View/download PDF

20. Zhan Wang (1911-2000)

Author

Shao, Guofan, Liu, Qijing, Qian, Hong, Chen, Jiquan, Ma, Jinshuang, and Tan, Zhengxiang

Published
2000

21. Ecosystem Management

Author

Dai, Limin, Shao, Guofan, Ren, Jizhou, Wang, Tao, Chen, Guangting, Lv, Xianguo, Xu, Qinzeng, Yang, Hongsheng, Xu, Qiang, Luo, Shiming, Ma, Lin, Qian, Liu, Ma, Wenqi, Zhang, Fusuo, Huang, Jikun, Jia, Xiangping, Xiang, Cheng, and Li, Wenhua, editor

Published
2015
Full Text
View/download PDF

26. A Robust Stepwise Clustering Approach to Detect Individual Trees in Temperate Hardwood Plantations Using Airborne LiDAR Data

Author

Shao, Gang, Fei, Songlin, Shao, Guofan, Shao, Gang, Fei, Songlin, and Shao, Guofan

Abstract

Precise tree inventory plays a critical role in sustainable forest planting, restoration, and management. LiDAR-based individual tree detection algorithms often focus on finding individual treetops to discern tree positions. However, deliquescent tree forms (broad, flattened crowns) in deciduous forests can make these algorithms ineffective. In this study, we propose a stepwise tree detection approach, by first identifying individual trees using horizontal point density and then analyzing their vertical structure profiles. We first project LiDAR data onto a 2D horizontal plane and apply mean shift clustering to generate candidate tree clusters. Next, we apply a series of structure analyses on the vertical phase, to overcome local variations in crown size and tree density. This study demonstrates that the horizontal point density of LiDAR data provides critical information to locate and isolate individual trees in temperate hardwood plantations with varied densities, while vertical structure profiles can identify spreading branches and reconstruct deliquescent crowns. One challenge of applying mean shift clustering is training a dynamic search kernel to identify trees of different sizes, which usually requires a large number of field measurements. The stepwise approach proposed in this study demonstrated robustness when using a constant kernel in clustering, making it an efficient tool for large-scale analysis. This stepwise approach was designed for quantifying temperate hardwood plantation inventories using relatively low-density airborne LiDAR, and it has potential applications for monitoring large-scale plantation forests. Further research is needed to adapt this method to natural stands with diverse tree ages and structures.

Published
2023

27. Bolstering Performance Evaluation of Image Segmentation Models With Efficacy Metrics in the Absence of a Gold Standard

Author

Tang, Lina, Shao, Jinyuan, Pang, Shiyan, Wang, Yameng, Maxwell, Aaron, Hu, Xiangyun, Gao, Zhi, Lan, Ting, and Shao, Guofan

Abstract

Image segmentation using deep learning has become overwhelmingly widespread. However, routine model testing methods can encounter evaluation inconsistencies or bias, largely due to how accuracy metrics respond to variations in class share distribution. Here, we address the effects of class imbalance on model performance evaluation and demonstrate a refined approach that incorporates image classification efficacy (ICE) metrics within the context of semantic segmentation in remote sensing. This evaluation approach was applied in six segmentation experiments that involved multispectral and LiDAR data, single or multiple models tested with the same or different datasets, and binary and multiclass schemes. ICE metrics revealed unique aspects of model’s segmentation capabilities compared to precision, recall, F-score, and overall accuracy. By mitigating the class imbalance effect, per-class efficacy enables precise class-level optimization of segmentation models, while whole-class efficacy facilitates evaluating a model’s potential performance when adapted to new datasets. The suitability of the kappa coefficient, ROC-AUC, and PR-AUC for model evaluation under class imbalance was discussed in comparison with ICE metrics. This efficacy-enhanced model evaluation protocol can be implemented for deep learning model training and testing. The routine use of this evaluation approach will strengthen the dependability and applicability of segmentation tools in various fields.

Published
2024
Full Text
View/download PDF

30. Introduction

Author

Shao, Guofan, Reynolds, Keith M., Shao, Guofan, editor, and Reynolds, Keith M., editor

Published
2006
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results