Query：
学者姓名：陈崇成
Refining：
Year
Type
Indexed by
Source
Complex
Former Name
Co-
Language
Clean All
Abstract ：
The accurate and efficient 3D reconstruction of trees is beneficial for urban forest resource assessment and management. Close-range photogrammetry (CRP) is widely used in the 3D model reconstruction of forest scenes. However, in practical forestry applications, challenges such as low reconstruction efficiency and poor reconstruction quality persist. Recently, novel view synthesis (NVS) technology, such as neural radiance fields (NeRF) and 3D Gaussian splatting (3DGS), has shown great potential in the 3D reconstruction of plants using some limited number of images. However, existing research typically focuses on small plants in orchards or individual trees. It remains uncertain whether this technology can be effectively applied in larger, more complex stands or forest scenes. In this study, we collected sequential images of urban forest plots with varying levels of complexity using imaging devices with different resolutions (cameras on smartphones and UAV). These plots included one with sparse, leafless trees and another with dense foliage and more occlusions. We then performed dense reconstruction of forest stands using NeRF and 3DGS methods. The resulting point cloud models were compared with those obtained through photogrammetric reconstruction and laser scanning methods. The results show that compared to photogrammetric method, NVS methods have a significant advantage in reconstruction efficiency. The photogrammetric method is suitable for relatively simple forest stands, as it is less adaptable to complex ones. This results in tree point cloud models with issues such as excessive canopy noise and wrongfully reconstructed trees with duplicated trunks and canopies. In contrast, NeRF is better adapted to more complex forest stands, yielding tree point clouds of the highest quality that offer more detailed trunk and canopy information. However, it can lead to reconstruction errors in the ground area when the input views are limited. The 3DGS method has a relatively poor capability to generate dense point clouds, resulting in models with low point density, particularly with sparse points in the trunk areas, which affects the accuracy of the diameter at breast height (DBH) estimation. Tree height and crown diameter information can be extracted from the point clouds reconstructed by all three methods, with NeRF achieving the highest accuracy in tree height. However, the accuracy of DBH extracted from photogrammetric point clouds is still higher than that from NeRF point clouds. Meanwhile, compared to ground-level smartphone images, tree parameters extracted from reconstruction results of higher-resolution and varied perspectives of drone images are more accurate. These findings confirm that NVS methods have significant application potential for 3D reconstruction of urban forests.
Keyword ：
3D Gaussian splatting (3DGS) 3D Gaussian splatting (3DGS) 3D reconstruction 3D reconstruction computer vision computer vision deep learning deep learning neural radiance fields (NeRF) neural radiance fields (NeRF) photogrammetry photogrammetry urban forest urban forest
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Tian, Guoji , Chen, Chongcheng , Huang, Hongyu . Comparative Analysis of Novel View Synthesis and Photogrammetry for 3D Forest Stand Reconstruction and Extraction of Individual Tree Parameters [J]. | REMOTE SENSING , 2025 , 17 (9) . |
| MLA | Tian, Guoji 等. "Comparative Analysis of Novel View Synthesis and Photogrammetry for 3D Forest Stand Reconstruction and Extraction of Individual Tree Parameters" . | REMOTE SENSING 17 . 9 (2025) . |
| APA | Tian, Guoji , Chen, Chongcheng , Huang, Hongyu . Comparative Analysis of Novel View Synthesis and Photogrammetry for 3D Forest Stand Reconstruction and Extraction of Individual Tree Parameters . | REMOTE SENSING , 2025 , 17 (9) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Information extraction is crucial for building and updating the knowledge base of expert systems. Large language models face challenges with prompt sensitivity and model hallucinations during information extraction. This study introduces the TIME (Tourism, Individuals, Moments, Events) model, which organizes figure-related information into four main dimensions: attributes, relationships, events, and their linkage to tourism resources. Then present a unified information extraction framework for figures, termed TIME-UIE. This framework integrates a unified task definition, a format output constraint, carefully selected demonstrations, and knowledge injection to verify consistency across different inference chains. Experimental results show that TIME-UIE outperforms baseline models in deciphering complex relationships between historical figures by 26.2% and in extracting event triplets by 11.1%. The study also proposes a loose matching metric for model performance evaluation, which holds significant implications for the practical application of the research methods.
Keyword ：
ChatGPT ChatGPT Cultural tourism Cultural tourism Historical figure Historical figure Information model Information model Large language models Large language models Unified information extraction Unified information extraction
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Fan, Zhanling , Chen, Chongcheng , Luo, Haifeng . TIME-UIE: Tourism-oriented figure information model and unified information extraction via large language models [J]. | EXPERT SYSTEMS WITH APPLICATIONS , 2025 , 278 . |
| MLA | Fan, Zhanling 等. "TIME-UIE: Tourism-oriented figure information model and unified information extraction via large language models" . | EXPERT SYSTEMS WITH APPLICATIONS 278 (2025) . |
| APA | Fan, Zhanling , Chen, Chongcheng , Luo, Haifeng . TIME-UIE: Tourism-oriented figure information model and unified information extraction via large language models . | EXPERT SYSTEMS WITH APPLICATIONS , 2025 , 278 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
In recent years, large-scale point cloud semantic segmentation has been widely applied in various fields, such as remote sensing and autonomous driving. Most existing point cloud networks use local aggregation to abstract unordered point clouds layer by layer. Among these, position embedding serves as a crucial step. However, current methods of position embedding have limitations in modeling spatial relationships, especially in deeper encoders where richer spatial positional relationships are needed. To address these issues, this paper summarizes the advantages and disadvantages of mainstream position embedding methods and proposes a novel Hybrid Offset Position Encoding (HOPE) module. This module comprises two branches that compute relative positional encoding (RPE) and offset positional encoding (OPE). RPE combines explicit encoding to enhance position features through attention, learning position bias implicitly, while OPE calculates absolute position offset encoding by considering differences with grouping embeddings. These two encodings are adaptively mixed in the final output. The experiment conducted on multiple datasets demonstrates that our module helps the deep encoders of the network capture more robust features, thereby improving model performance on various baseline models. For instance, PointNet++ and PointMetaBase enhanced with HOPE achieved mIoU gains of 2.1% and 1.3% on the large-scale indoor dataset S3DIS area-5, 2.5% and 1.1% on S3DIS 6-fold, and 1.5% and 0.6% on ScanNet, respectively. RandLA-Net with HOPE achieved a 1.4% improvement on the large-scale outdoor dataset Toronto3D, all with minimal additional computational cost. PointNet++ and PointMetaBase had approximately only a 0.1 M parameter increase. This module can serve as an alternative for position embedding, and is suitable for point-based networks requiring local aggregation.
Keyword ：
attention mechanism attention mechanism large-scale point cloud large-scale point cloud local aggregation local aggregation positional encoding positional encoding position embedding position embedding semantic segmentation semantic segmentation
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Xiao, Yu , Wu, Hui , Chen, Yisheng et al. Hybrid Offset Position Encoding for Large-Scale Point Cloud Semantic Segmentation [J]. | REMOTE SENSING , 2025 , 17 (2) . |
| MLA | Xiao, Yu et al. "Hybrid Offset Position Encoding for Large-Scale Point Cloud Semantic Segmentation" . | REMOTE SENSING 17 . 2 (2025) . |
| APA | Xiao, Yu , Wu, Hui , Chen, Yisheng , Chen, Chongcheng , Dong, Ruihai , Lin, Ding . Hybrid Offset Position Encoding for Large-Scale Point Cloud Semantic Segmentation . | REMOTE SENSING , 2025 , 17 (2) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Amenity refers to health and well-being benefits that attract recreationists to protected areas. However, few studies have explored the specific components and characteristics of the recreational amenity experience from the perspective of Chinese recreationists. This concept is crucial for promoting protected areas as contributors to health and well-being and for advancing the well-planned development of harmonious coexistence between humans and nature. The study used semi-structured interviews and grounded theory to establish the recreational amenity experience model, through analysis of 25 qualitative interviews. The recreational amenity experience model was constructed, including four dimensions of physiological, psychological, spiritual, and social amenity. The recreational amenity experience model developed in this study is based not only on the recreationists' perceived health and well-being but also on the leisure and aesthetic aspects of traditional Chinese culture. Moreover, this study expands on the theory of cultural ecosystem services and promotes the harmonious coexistence of humans and nature. Management implications: center dot The Recreational Amenity Experience Model (RAE) provides a framework for understanding how amenity elements within protected areas contribute to visitors' health and well-being. center dot By recognizing the multifaceted dimensions of recreational amenity experience, managers of protected areas can design and implement programs that cater to a wide range of visitor needs. center dot Managers of protected areas can refer to the RAE to comprehend the full spectrum of visitors' expectations and identify intervention points to enhance the quality of recreation. center dot Understanding the critical role of recreational amenity experience in delivering cultural ecosystem services can inform resource allocation decisions.
Keyword ：
Cultural ecosystem services Cultural ecosystem services Health and well-being Health and well-being Leisure in Chinese culture Leisure in Chinese culture Protected areas Protected areas Recreational amenity experience Recreational amenity experience
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Lin, Kaimiao , Zhang, Qing , Lu, Qiong et al. Linking protected areas with health and well-being: Reconstructing the recreational amenity experience model in the Chinese context [J]. | JOURNAL OF OUTDOOR RECREATION AND TOURISM-RESEARCH PLANNING AND MANAGEMENT , 2025 , 50 . |
| MLA | Lin, Kaimiao et al. "Linking protected areas with health and well-being: Reconstructing the recreational amenity experience model in the Chinese context" . | JOURNAL OF OUTDOOR RECREATION AND TOURISM-RESEARCH PLANNING AND MANAGEMENT 50 (2025) . |
| APA | Lin, Kaimiao , Zhang, Qing , Lu, Qiong , Meng, Fang , Chen, Chongcheng . Linking protected areas with health and well-being: Reconstructing the recreational amenity experience model in the Chinese context . | JOURNAL OF OUTDOOR RECREATION AND TOURISM-RESEARCH PLANNING AND MANAGEMENT , 2025 , 50 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
在公网受限的应急环境中,利用无人机辅助物联网能促进传感数据的及时传递.当考虑无线通信距离时,无人机作为移动收集器在有限续航时间内收集尽可能多的传感数据的路径规划可建模为足够近定向问题(CEOP).现有求解CEOP的算法是逐个计算目标节点的访问顺序及其邻域内的采集点,这在节点邻域较大并覆盖周围多个节点时效率低下,这些方法也没有考虑数据传输时间和无人机遥控距离等约束.为此,建立了大邻域多约束无人机数据收集路径规划的数学模型,提出了基于贪婪随机自适应搜索过程(GRASP)的GRASP-LN算法进行求解.该算法不重复计算重合的采集点,而是维护路径每个航点采集的节点集合,无人机在每个航点悬停一段时间以收集集合内节点的数据.公开的CEOP数据集的实验结果表明,GRASP-LN比GSOA、VNS和GRASPopt具有更好的求解质量和更短的计算时间.与基线算法GRASPopt相比,GRASP-LN的路径奖励平均提高了5.86%,最大提高了14.91%,执行时间平均减少了69%,特别在节点邻域平均覆盖4.67个以上节点时,GRASP-LN的路径奖励和稳定性均优于GRASPopt.考虑数据传输时间和无人机遥控距离约束的实验验证了GRASP-LN算法对考虑这些约束的无人机数据收集路径规划问题的有效性.
Keyword ：
数据收集 数据收集 无人机 无人机 物联网 物联网 贪婪随机自适应搜索过程 贪婪随机自适应搜索过程 足够近定向问题 足够近定向问题 路径规划 路径规划
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 潘淼鑫 , 陈崇成 . 大邻域多约束无人机数据收集路径规划 [J]. | 计算机科学与探索 , 2025 , 19 (1) : 158-168 . |
| MLA | 潘淼鑫 et al. "大邻域多约束无人机数据收集路径规划" . | 计算机科学与探索 19 . 1 (2025) : 158-168 . |
| APA | 潘淼鑫 , 陈崇成 . 大邻域多约束无人机数据收集路径规划 . | 计算机科学与探索 , 2025 , 19 (1) , 158-168 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Tourism knowledge graphs lack cultural content, limiting their usefulness for cultural tourists. This paper presents the development of a cultural perspective-based knowledge graph (CuPe-KG). We evaluated fine-tuning ERNIE 3.0 (FT-ERNIE) and ChatGPT for cultural type recognition to strengthen the relationship between tourism resources and cultures. Our investigation used an annotated cultural tourism resource dataset containing 2,745 items across 16 cultural types. The results showed accuracy scores for FT-ERNIE and ChatGPT of 0.81 and 0.12, respectively, with FT-ERNIE achieving a micro-F1 score of 0.93, a 26 percentage point lead over ChatGPT's score of 0.67. These underscore FT-ERNIE's superior performance (the shortcoming is the need to annotate data) while highlighting ChatGPT's limitations because of insufficient Chinese training data and lower identification accuracy in professional knowledge. A novel ontology was designed to facilitate the construction of CuPe-KG, including elements such as cultural types, historical figures, events, and intangible cultural heritage. CuPe-KG effectively addresses cultural tourism visitors' information retrieval needs.
Keyword ：
ChatGPT ChatGPT Cultural tourism Cultural tourism Cultural type Cultural type Knowledge graph Knowledge graph Pretrained language models Pretrained language models Travel intelligence Travel intelligence
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Fan, Zhanling , Chen, Chongcheng . CuPe-KG: Cultural perspective-based knowledge graph construction of tourism resources via pretrained language models [J]. | INFORMATION PROCESSING & MANAGEMENT , 2024 , 61 (3) . |
| MLA | Fan, Zhanling et al. "CuPe-KG: Cultural perspective-based knowledge graph construction of tourism resources via pretrained language models" . | INFORMATION PROCESSING & MANAGEMENT 61 . 3 (2024) . |
| APA | Fan, Zhanling , Chen, Chongcheng . CuPe-KG: Cultural perspective-based knowledge graph construction of tourism resources via pretrained language models . | INFORMATION PROCESSING & MANAGEMENT , 2024 , 61 (3) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Three-dimensional (3D) reconstruction of trees has always been a key task in precision forestry management and research. Due to the complex branch morphological structure of trees themselves and the occlusions from tree stems, branches and foliage, it is difficult to recreate a complete three-dimensional tree model from a two-dimensional image by conventional photogrammetric methods. In this study, based on tree images collected by various cameras in different ways, the Neural Radiance Fields (NeRF) method was used for individual tree dense reconstruction and the exported point cloud models are compared with point clouds derived from photogrammetric reconstruction and laser scanning methods. The results show that the NeRF method performs well in individual tree 3D reconstruction, as it has a higher successful reconstruction rate, better reconstruction in the canopy area and requires less images as input. Compared with the photogrammetric dense reconstruction method, NeRF has significant advantages in reconstruction efficiency and is adaptable to complex scenes, but the generated point cloud tend to be noisy and of low resolution. The accuracy of tree structural parameters (tree height and diameter at breast height) extracted from the photogrammetric point cloud is still higher than those derived from the NeRF point cloud. The results of this study illustrate the great potential of the NeRF method for individual tree reconstruction, and it provides new ideas and research directions for 3D reconstruction and visualization of complex forest scenes.
Keyword ：
3D reconstruction 3D reconstruction 3D tree modeling 3D tree modeling deep learning deep learning individual tree individual tree lidar lidar neural radiance field (NeRF) neural radiance field (NeRF) photogrammetry photogrammetry terrestrial laser scanning terrestrial laser scanning
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Huang, Hongyu , Tian, Guoji , Chen, Chongcheng . Evaluating the Point Cloud of Individual Trees Generated from Images Based on Neural Radiance Fields (NeRF) Method [J]. | REMOTE SENSING , 2024 , 16 (6) . |
| MLA | Huang, Hongyu et al. "Evaluating the Point Cloud of Individual Trees Generated from Images Based on Neural Radiance Fields (NeRF) Method" . | REMOTE SENSING 16 . 6 (2024) . |
| APA | Huang, Hongyu , Tian, Guoji , Chen, Chongcheng . Evaluating the Point Cloud of Individual Trees Generated from Images Based on Neural Radiance Fields (NeRF) Method . | REMOTE SENSING , 2024 , 16 (6) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Indoor point clouds often present significant challenges due to the complexity and variety of structures and high object similarity. The local geometric structure helps the model learn the shape features of objects at the detail level, while the global context provides overall scene semantics and spatial relationship information between objects. To address these challenges, we propose a novel network architecture, PointMSGT, which includes a multi-scale geometric feature extraction (MSGFE) module and a global Transformer (GT) module. The MSGFE module consists of a geometric feature extraction (GFE) module and a multi-scale attention (MSA) module. The GFE module reconstructs triangles through each point's two neighbors and extracts detailed local geometric relationships by the triangle's centroid, normal vector, and plane constant. The MSA module extracts features through multi-scale convolutions and adaptively aggregates features, focusing on both local geometric details and global semantic information at different scale levels, enhancing the understanding of complex scenes. The global Transformer employs a self-attention mechanism to capture long-range dependencies across the entire point cloud. The proposed method demonstrates competitive performance in real-world indoor scenarios, with a mIoU of 68.6% in semantic segmentation on S3DIS and OA of 86.4% in classification on ScanObjectNN.
Keyword ：
geometric feature geometric feature multi-scale attention multi-scale attention point cloud analysis point cloud analysis real-world indoor scenario real-world indoor scenario transformer transformer
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Chen, Yisheng , Xiao, Yu , Wu, Hui et al. Multi-Scale Geometric Feature Extraction and Global Transformer for Real-World Indoor Point Cloud Analysis [J]. | MATHEMATICS , 2024 , 12 (23) . |
| MLA | Chen, Yisheng et al. "Multi-Scale Geometric Feature Extraction and Global Transformer for Real-World Indoor Point Cloud Analysis" . | MATHEMATICS 12 . 23 (2024) . |
| APA | Chen, Yisheng , Xiao, Yu , Wu, Hui , Chen, Chongcheng , Lin, Ding . Multi-Scale Geometric Feature Extraction and Global Transformer for Real-World Indoor Point Cloud Analysis . | MATHEMATICS , 2024 , 12 (23) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Accurate and continuous maps of maize distribution are essential for food security and sustainable agricultural development. However, there are no continuous national-scale and fine-resolution maize maps and explicit updated information on the spatiotemporal dynamics of maize for most countries. Maize mapping at the national scale is challenging due to the spectral heterogeneity caused by crop growth conditions, cropping patterns, and inter-annual variations. To this end, this study developed a novel crop index-based algorithm for national-scale maize mapping. Compared to other crops, maize is characterized by large-leaf-dominated canopies and high photosynthetic efficiency. Maize shows significant changes in chlorophyll and anthocyanin content. Therefore, a robust maize index was established by exploring the temporal Variation of the Vegetation-Pigment index (VVP) during the growing period. A simple decision rule was coded on the Google Earth Engine (GEE) platform, which was used for maize mapping based on the Sentinel-2 time series in China and the contiguous United States (US) from 2018 to 2022. The national-scale 10 m annual maize maps for China and the contiguous US were developed and in good agreement with the corresponding agricultural statistics data for many years (R-2 > 0.94) and 9,412 reference points (overall accuracy of 90.09 %). Compared with simply applying the vegetation index, the VVP index took account of spectral heterogeneity caused by variations in crop growth conditions, cropping patterns, and inter-annual, and the omission error of maize was reduced by over 20 %. Moreover, the VVP index can significantly improve the spatial transferability of the Random Forest (RF) classifier. The first 10 m annual maize maps for China revealed that the planted area trend decreased and then increased from 2018 to 2022. The year 2020 was the turning point. The maize planted area consisted of 68 % single maize and 32 % double cropping with maize in 2020, with the northern boundary for double cropping with maize in the Yanshan Mountains. The maize planted area in China consistently decreased by 39,352 km(2) (about 9 %) from 2018 to 2020. This is mainly due to the adjustment of the maize-planted structure in the "Sickle Bend" region of China (the "Sickle Bend" policy). However, the maize planted area gradually recovered from 2020 to 2022, primarily concentrated in regions with ever-planted. This study will provide essential information for cropping structure adjustment and related agricultural policy formulation and contribute to sustainable agricultural development by mapping maize from a national to a global scale.
Keyword ：
Crop mapping Crop mapping Cross -region Cross -region Maize index Maize index National -scale National -scale Spatiotemporal variations Spatiotemporal variations
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Huang, Yingze , Qiu, Bingwen , Yang, Peng et al. National-scale 10 m annual maize maps for China and the contiguous United States using a robust index from Sentinel-2 time series [J]. | COMPUTERS AND ELECTRONICS IN AGRICULTURE , 2024 , 221 . |
| MLA | Huang, Yingze et al. "National-scale 10 m annual maize maps for China and the contiguous United States using a robust index from Sentinel-2 time series" . | COMPUTERS AND ELECTRONICS IN AGRICULTURE 221 (2024) . |
| APA | Huang, Yingze , Qiu, Bingwen , Yang, Peng , Wu, Wenbin , Chen, Xuehong , Zhu, Xiaolin et al. National-scale 10 m annual maize maps for China and the contiguous United States using a robust index from Sentinel-2 time series . | COMPUTERS AND ELECTRONICS IN AGRICULTURE , 2024 , 221 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Point clouds are essential 3D data representations utilized across various disciplines, often requiring point cloud completion methods to address inherent incompleteness. Existing completion methods like SnowflakeNet only consider local attention, lacking global information of the complete shape, and tend to suffer from overfitting as the model depth increases. To address these issues, we introduced self-positioning point-based attention to better capture complete global contextual features and designed a Channel Attention module for adaptive feature adjustment within the global vector. Additionally, we implemented a vector attention grouping strategy in both the skip-transformer and self-positioning point-based attention to mitigate overfitting, improving parameter efficiency and generalization. We evaluated our method on the PCN dataset as well as the ShapeNet55/34 datasets. The experimental results show that our method achieved an average CD-L1 of 7.09 and average CD-L2 scores of 8.0, 7.8, and 14.4 on the PCN, ShapeNet55, ShapeNet34, and ShapeNet-unseen21 benchmarks, respectively. Compared to SnowflakeNet, we improved the average CD by 1.6%, 3.6%, 3.7%, and 4.6% on the corresponding benchmarks, while also reducing complexity and computational costs and accelerating training and inference speeds. Compared to other existing point cloud completion networks, our method also achieves competitive results.
Keyword ：
3D point cloud 3D point cloud attention mechanism attention mechanism deep learning deep learning point cloud completion point cloud completion
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Xiao, Yu , Chen, Yisheng , Chen, Chongcheng et al. GSSnowflake: Point Cloud Completion by Snowflake with Grouped Vector and Self-Positioning Point Attention [J]. | REMOTE SENSING , 2024 , 16 (17) . |
| MLA | Xiao, Yu et al. "GSSnowflake: Point Cloud Completion by Snowflake with Grouped Vector and Self-Positioning Point Attention" . | REMOTE SENSING 16 . 17 (2024) . |
| APA | Xiao, Yu , Chen, Yisheng , Chen, Chongcheng , Lin, Ding . GSSnowflake: Point Cloud Completion by Snowflake with Grouped Vector and Self-Positioning Point Attention . | REMOTE SENSING , 2024 , 16 (17) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
