Ruifang Deng

Ms. Ruifang Deng


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
7 December 2023 in Sustainability
Verified
Journal Article
Open access
PDF file
RDRuifang Deng
Ruifang Deng
XDXue Ding
Xue Ding
Jinliang WangJinliang Wang
Jinliang Wang
The central Yunnan urban agglomeration represents a typical urban cluster in the southwestern region of China. The swift urbanization and land use changes in this region pose a severe threat to the ecosystem. A thorough assessment of the landscape ecological risk in the central Yunnan urban agglomeration holds paramount importance for devising effective risk management strategies and sustainable, high-quality development plans. This study utilizes long-term land-use raster data for six time periods (1995, 2000, 2005, 2010, 2015, and 2020) in the Central Yunnan urban agglomeration. Using GIS technology, a landscape risk index model is constructed, and a comprehensive assessment of landscape ecological risks in the Central Yunnan urban agglomeration is conducted using the 5 km × 5 km grid analysis method and Kriging interpolation. The results indicate that, between 1995 and 2020, the Central Yunnan urban agglomeration was dominated by forest land, grassland, and cultivated land as the primary land-use types. Forest land covered over 48% of the total area, while grassland and cultivated land accounted for more than 26% and 18%, respectively. Notably, construction land underwent a significant increase, mainly due to conversions from cultivated land, forest land, and grassland. Over a span of 25 years, the study area has experienced a continual rise in landscape ecological risk. The landscape ecological risk was mainly characterized by medium, higher, and high ecological risk. Grassland predominated in areas with medium levels of ecological risk, while cultivated land and construction land were predominant in regions with higher and high levels of ecological risk. Spatially, regions with lower ecological risk were primarily distributed in the Chuxiong Yi Autonomous Prefecture, whereas areas with higher and high levels of ecological risk were concentrated in Qujing City and Kunming City. The spatial aggregation patterns of landscape ecological risk in the Central Yunnan urban agglomeration featured “high–high” (H–H) and “low–low” (L–L) clusters, both displaying an initial increase followed by a decrease. The primary factors contributing to the rise in the landscape ecological risk index were identified as urban expansion, population growth, ecological fragmentation, and vegetation destruction. The study’s outcomes can offer valuable insights for optimizing land resources and promoting sustainable development in the Central Yunnan urban agglomeration.

Verified
Journal Article
Open access
PDF file
XDXue Ding
Xue Ding
ZWZhaoqian Wang
Zhaoqian Wang
SPShuangyun Peng
Shuangyun Peng
XSXin Shao
Xin Shao
RDRuifang Deng
Ruifang Deng
To address the challenges that convolutional neural networks (CNNs) face in extracting small objects and handling class imbalance in remote sensing imagery, this paper proposes a novel spatial contextual information and multiscale feature fusion encoding–decoding network, SCIMF-Net. Firstly, SCIMF-Net employs an improved ResNeXt-101 deep backbone network, significantly enhancing the extraction capability of small object features. Next, a novel PMFF module is designed to effectively promote the fusion of features at different scales, deepening the model’s understanding of global and local spatial contextual information. Finally, introducing a weighted joint loss function improves the SCIMF-Net model’s performance in extracting LULC information under class imbalance conditions. Experimental results show that compared to other CNNs such as Res-FCN, U-Net, SE-U-Net, and U-Net++, SCIMF-Net improves PA by 0.68%, 0.54%, 1.61%, and 3.39%, respectively; MPA by 2.96%, 4.51%, 2.37%, and 3.45%, respectively; and MIOU by 3.27%, 4.89%, 4.2%, and 5.68%, respectively. Detailed comparisons of locally visualized LULC information extraction results indicate that SCIMF-Net can accurately extract information from imbalanced classes and small objects.