A5055476501- Remote Sensing and Land Use
- Automated Road and Building Extraction
- Remote Sensing and LiDAR Applications
- Land Use and Ecosystem Services
- 3D Surveying and Cultural Heritage
- Geographic Information Systems Studies
- Medical Image Segmentation Techniques
- Simulation Techniques and Applications
- Time Series Analysis and Forecasting
- Advanced Vision and Imaging
- Remote-Sensing Image Classification
- Advanced Decision-Making Techniques
- Image Processing and 3D Reconstruction
- Video Surveillance and Tracking Methods
- Optical measurement and interference techniques
- Remote Sensing in Agriculture
- Image Retrieval and Classification Techniques
- Satellite Image Processing and Photogrammetry
Henan University
2019-2025
Zhengzhou University of Industrial Technology
2021-2022
The extraction of shape features from vector elements is essential in cartography and geographic information science, supporting a range intelligent processing tasks. Traditional methods rely on different machine learning algorithms tailored to specific types line polygon elements, limiting their general applicability. This study introduces novel approach called “Pre-Trained Shape Feature Representations Transformers (PSRT)”, which utilizes transformer encoders designed with three...
Simplifying building contours involves reducing data volume while preserving the continuity, accuracy, and essential characteristics of shapes. This presents significant challenges for sequence representation generation. Traditional methods often rely on complex rule design, feature engineering, iterative optimization. To overcome these limitations, this study proposes a Transformer-based Polygon Simplification Model (TPSM) end-to-end vector simplification contours. TPSM processes ordered...
Buildings are important entity objects of cities, and the classification building shapes plays an indispensable role in cognition planning urban structure. In recent years, some deep learning methods have been proposed for recognizing footprints modern electronic maps. Furthermore, their performance depends on enough labeled samples each class footprints. However, it is impractical to label type footprint shapes. Therefore, using few more preferable recognize classify this paper, we propose...
The classification and recognition of the shapes buildings in map space play an important role spatial cognition, cartographic generalization, updating. As are often represented as vector data, research was conducted to learn feature representations recognize their based on graph neural networks. Due principles networks, it is necessary construct a represent adjacency relationships between points (i.e., vertices polygons shaping buildings), extract list geometric features for each point....
Vegetation cover in the Loess Plateau region is an important component of ecological protection Yellow River Basin, and this study provides a scientific reference for further vegetation restoration. Based on Landsat images related data, we utilized dimidiate pixel model Geodetector method to Wuding Basin from 2000 2022. The results indicated spatial temporal distribution its changes over period. Additionally, driving factors influencing were also uncovered. We propose land use shift...
Shape is one of the core features buildings which are main elements map. The building shape recognition widely used in many spatial applications. Due to irregularity contour, it still challenging for recognition. Inspired by graph signal processing theory, we propose a deep filter neural network (DGFN) maps. Firstly, regard as combination subjective and objective filtering process. Secondly, construct extraction framework from perspective details, structure local information. Thirdly, DGFN...
The dynamic error correction of simulation turntable is an importance way to improve the turntable's accuracy and reliability in process simulation. In this paper, modelling prediction method combining ARMA (autoregressive moving average) model NN (neural network) proposed. method, AIC criterion used determine order model, which can solve problem confirm input output variables BP (back propagation) network. predicted corrected online by ARMA-NN get better performance. validity illustrated MATLAB