Peilin Liu
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
Peilin Liu has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Albert Fekete
Institute of Landscape Architecture, Urban Planning and Garden Art, Hungarian University of Agriculture and Life Sciences (MATE), 1118 Budapest, Hungary
1 shared publication - Chao Yin
Guangzhou Institute of Geography, Guangdong Academy of Sciences, Guangzhou 510070, China
1 shared publication - Jinbiao Yan
College of Geography and Tourism, Hengyang Normal University
1 shared publication - Martin Van Den Toorn
St Istvan Univ, Fac Landscape Architecture & Urbanism, Budapest, Hungary
1 shared publication - Ning DongGe
Institute of Landscape Architecture, Urban Planning and Garden Art, Hungarian University of Agriculture and Life Science, 1118 Budapest, Hungary
1 shared publication - Qing Han
National-Local Joint Engineering Laboratory on Digital Preservation and Innovative Technologies for the Culture of Traditional Villages and Towns, Hengyang Normal University, Hengyang 421002, China
1 shared publication
Latest Publications
- Open Access
Historical Water Management Strategies—Case Study of Traditional Villages in Southern China, Hunan Province
Based on the landscape architecture of traditional settlements in southern China, this study takes water as a vital element through field investigation and model analysis to explore the water management strategies of two traditional villages in Xiangjiang River Basin, Hunan Province. We have found that traditional settlements are located between rivers and mountains. The community of the settlement has a strong interaction with the water environment. The water management system consists of two parts: the rainwater collection and storage system of a single building and the settlement’ s water collection and drainage system. Through calculation, we found that the amounts of water collected (per year) between the two villages are different: ZhangGuYing (Z village) = 5.73 million L, ShangGanTang (S village) = 1.784 million L, in spite of the fact that water management strategies of the two settlements are similar. Further analysis shows that the difference is related to the adaption of the precipitation and topography of the surrounding areas. The above-mentioned systematic management strategy of water resources has been used until currently, with adaptability, low cost, and sustainability. It has outstanding significance for the current demand for sustainable development from both resource management and cultural aspects.
Open AccessHistorical Water Management Strategies—Case Study of Traditional Villages in Southern China, Hunan Province
Based on the landscape architecture of traditional settlements in southern China, this study takes water as a vital element through field investigation and model analysis to explore the water management strategies of two traditional villages in Xiangjiang River Basin, Hunan Province. We have found that traditional settlements are located between rivers and mountains. The community of the settlement has a strong interaction with the water environment. The water management system consists of two parts: the rainwater collection and storage system of a single building and the settlement’ s water collection and drainage system. Through calculation, we found that the amounts of water collected (per year) between the two villages are different: ZhangGuYing (Z village) = 5.73 million L, ShangGanTang (S village) = 1.784 million L, in spite of the fact that water management strategies of the two settlements are similar. Further analysis shows that the difference is related to the adaption of the precipitation and topography of the surrounding areas. The above-mentioned systematic management strategy of water resources has been used until currently, with adaptability, low cost, and sustainability. It has outstanding significance for the current demand for sustainable development from both resource management and cultural aspects.
- Open Access
- 20 October 2022
- Published by MDPI AG in Remote Sensing
- Vol. 14 (20) , 5250
Towards Classification of Architectural Styles of Chinese Traditional Settlements Using Deep Learning: A Dataset, a New Framework, and Its Interpretability
The classification of architectural style for Chinese traditional settlements (CTSs) has become a crucial task for developing and preserving settlements. Traditionally, the classification of CTSs primarily relies on manual work, which is inefficient and time consuming. Inspired by the tremendous success of deep learning (DL), some recent studies attempted to apply DL networks such as convolution neural networks (CNNs) to achieve automated classification of the architecture styles. However, these studies suffer overfitting problems of the CNNs, leading to inferior classification performance. Moreover, most of the studies apply the CNNs as a black box providing limited interpretability. To address these limitations, a new DL classification framework is proposed in this study to overcome the overfitting problem by transfer learning and learning-based data augmentation technique (i.e., AutoAugment). Furthermore, we also employ class activation map (CAM) visualization technique to help understand how the CNN classifiers work to abstract patterns from the input. Specifically, due to a lack of architectural style datasets for the CTSs, a new annotated dataset is first established with six representative classes. Second, several representative CNNs are leveraged to benchmark the new dataset. Third, to address the overfitting problem of the CNNs, a new DL framework is proposed which combines transfer learning and AutoAugment to improve the classification performance. Extensive experiments are conducted on the new dataset to demonstrate the effectiveness of our framework. The proposed framework achieves much better performance than baselines, greatly mitigating the overfitting problem. Additionally, the CAM visualization technique is harnessed to explain what and how the CNN classifiers implicitly learn for recognizing a specified architectural style.
- 20 October 2022
- Published by MDPI AG in Remote Sensing
- Vol. 14 (20) , 5250
Open AccessTowards Classification of Architectural Styles of Chinese Traditional Settlements Using Deep Learning: A Dataset, a New Framework, and Its Interpretability
The classification of architectural style for Chinese traditional settlements (CTSs) has become a crucial task for developing and preserving settlements. Traditionally, the classification of CTSs primarily relies on manual work, which is inefficient and time consuming. Inspired by the tremendous success of deep learning (DL), some recent studies attempted to apply DL networks such as convolution neural networks (CNNs) to achieve automated classification of the architecture styles. However, these studies suffer overfitting problems of the CNNs, leading to inferior classification performance. Moreover, most of the studies apply the CNNs as a black box providing limited interpretability. To address these limitations, a new DL classification framework is proposed in this study to overcome the overfitting problem by transfer learning and learning-based data augmentation technique (i.e., AutoAugment). Furthermore, we also employ class activation map (CAM) visualization technique to help understand how the CNN classifiers work to abstract patterns from the input. Specifically, due to a lack of architectural style datasets for the CTSs, a new annotated dataset is first established with six representative classes. Second, several representative CNNs are leveraged to benchmark the new dataset. Third, to address the overfitting problem of the CNNs, a new DL framework is proposed which combines transfer learning and AutoAugment to improve the classification performance. Extensive experiments are conducted on the new dataset to demonstrate the effectiveness of our framework. The proposed framework achieves much better performance than baselines, greatly mitigating the overfitting problem. Additionally, the CAM visualization technique is harnessed to explain what and how the CNN classifiers implicitly learn for recognizing a specified architectural style.