A5000354437- Advancements in PLL and VCO Technologies
- Remote-Sensing Image Classification
- Remote Sensing and LiDAR Applications
- Radio Frequency Integrated Circuit Design
- Advanced Image Processing Techniques
- Image Enhancement Techniques
- Photonic and Optical Devices
- Semiconductor Lasers and Optical Devices
- Oil Palm Production and Sustainability
- Image Processing Techniques and Applications
- Automated Road and Building Extraction
- Advanced Vision and Imaging
- Advanced Image Fusion Techniques
- Remote Sensing in Agriculture
- Image and Signal Denoising Methods
- Analog and Mixed-Signal Circuit Design
- Advanced Image and Video Retrieval Techniques
- VLSI and Analog Circuit Testing
- Advanced Neural Network Applications
- Domain Adaptation and Few-Shot Learning
- Remote Sensing and Land Use
- Advanced Fluorescence Microscopy Techniques
- Probabilistic and Robust Engineering Design
- Wood and Agarwood Research
- Ship Hydrodynamics and Maneuverability
Tsinghua University
2014-2024
Chinese Academy of Surveying and Mapping
2023-2024
University of Hong Kong
2024
Nanjing Normal University
2024
Yantai University
2024
Tianjin University of Science and Technology
2024
Shanghai Jiao Tong University
2020-2024
City University of Hong Kong
2022-2023
Nanjing Surveying and Mapping Research Institute (China)
2023
Hubei University of Education
2023
Accurate and efficient mapping of road networks is crucial for evaluating urban development, transportation accessibility, environmental impact. However, existing extraction methods utilizing remote sensing images suffer from limited generalization ability object occlusion, resulting in fragmented discontinuous segmentation. Consequently, these limitations impede the practical applicability multi-city diverse-scenario applications. To address challenges, we propose SWCARE, a method with...
Choosing clothes with complex patterns and colors is a challenging task for visually impaired people. Automatic clothing pattern recognition also research problem due to rotation, scaling, illumination, especially large intraclass variations. We have developed camera-based prototype system that recognizes in four categories (plaid, striped, patternless, irregular) identifies 11 colors. The integrates camera, microphone, computer, Bluetooth earpiece audio description of A camera mounted upon...
Ecological civilization has emerged as an important component within the 14th Five-Year Plan (2021–2025) in China. As a pivotal direction of ecological development, total number, scope, and complexity China’s ecotourism are steadily increasing. However, systematic reviews relevant literature have been lacking over past few years. This study employs CiteSpace for visual analysis research from 2014 to 2024 Web Science Core Collections (WoSCC) database. The main findings follows: (1) Ecotourism...
Oil palm is of great importance in agricultural productivity for many tropic developing countries and accordingly investigating as well counting oil palms a meaningful valuable research. In this paper, we firstly apply Faster-RCNN, one the most popular object detection algorithms, to detect tree crowns from satellite images. Although Faster-RCNN has an excellent performance well-known datasets general detection, it does not have obvious advantages study compared with other classical machine...
In high-resolution remote sensing images (RSIs), complex composite object detection (e.g., coal-fired power plant and harbor detection) is challenging due to multiple discrete parts with variable layouts leading weak inter-relationship blurred boundaries, instead of a clearly defined single object. To address this issue, article proposes an end-to-end framework, i.e., relational part-aware network (REPAN), explore the semantic correlation extract discriminative features among parts....
Existing methods for building extraction from remotely sensed images strongly rely on aerial or satellite-based with very high resolution, which are usually limited by spatiotemporally accessibility and cost. In contrast, relatively low-resolution have better spatial temporal availability but cannot directly contribute to fine- and/or high-resolution extraction. this paper, based image super-resolution segmentation techniques, we propose a two-stage framework (SRBuildingSeg) achieving (SR)...
Over the past decade, domain adaptation (DA) algorithms have been proposed to address gap problems as they do not need any interpretation in target domain. However, most existing efforts focus on scenarios with only one source and In this article, we explore scenario mixed multiple domains for remote sensing applications propose a new algorithm, named two-stage network (TSAN). First, utilize adversarial learning approach confuse classifier discriminate between whole mixed-multiple-target...
Providing accurate and timely oil palm information on a large scale is essential for both economic development ecological significance. However, owing to different sensors, photograph acquisition conditions, environmental heterogeneity, the volume variety of data make it extremely challenging large-scale cross-regional tree detection. It computationally expensive train model from images covering heterogeneous regions all conditions continuously accumulated multisource remote sensing data. In...
Matching clothes is a challenging task for many blind people. In this paper, we present proof of concept system to solve problem. The consists 1) camera connected computer perform pattern and color matching process; 2) spee
A 25 Gb/s transmitter (TX) and receiver (RX) chipset designed in a 65 nm CMOS technology is presented. The proposed quarter-rate TX architecture with divider-less clock generation can not only guarantee the timing constraint for highest-speed serialization, but also save power compared conventional designs. source-series terminated (SST) driver 2-tap feed-forward equalizer (FFE) far-end crosstalk canceller (XTC) implemented chip. RX chip employs an adaptive decision-feedback (DFE) baud-rate...
Nighttime light (NTL) remote sensing observation serves as a unique proxy for quantitatively assessing progress toward meeting series of Sustainable Development Goals (SDGs), such poverty estimation, urban sustainable development, and carbon emission. However, existing NTL observations often suffer from pervasive degradation inconsistency, limiting their utility computing the indicators defined by SDGs. In this study, we propose novel approach to reconstruct high-resolution images using...
Clothes pattern recognition is a challenging task for blind or visually impaired people. Automatic clothes also problem in computer vision due to the large variations. In this paper, we present new method classify patterns into 4 categories: stripe, lattice, special, and patternless. While existing texture analysis methods mainly focused on textures varying with distinctive changes, they cannot achieve same level of accuracy because intra-class variations each category. To solve problem,...
Sentinel-2 imagery has garnered significant attention in many earth system studies due to free access and high revisit frequency. Since its spatial resolution is insufficient for applications, e.g., fine-grained land cover mapping, some employ fusion technique that combines high-resolution RGB images with multispectral improve the of latter. However, there are two issues existing image methods. First, these methods usually assume time intervals between short (within several days), which a...
Semi-supervised learning has attracted increasing attention in the large-scale land cover mapping task. However, existing methods overlook potential to alleviate class imbalance problem by selecting a suitable set of unlabeled data. Besides, class-imbalanced scenarios, pseudo-labeling mostly only pick confident samples, failing exploit hard samples during training. To tackle these issues, we propose unified Class-Aware Semi-Supervised Semantic Segmentation framework. The proposed framework...