A5101820249- Advanced Image and Video Retrieval Techniques
- Remote-Sensing Image Classification
- Advanced Neural Network Applications
- Remote Sensing and Land Use
- Image and Object Detection Techniques
- Advanced Computational Techniques and Applications
- Optical Systems and Laser Technology
- Infrastructure Maintenance and Monitoring
- Simulation and Modeling Applications
- Robotics and Sensor-Based Localization
- Machine Learning and Data Classification
- Advanced Image Fusion Techniques
- Medical Image Segmentation Techniques
- Infrared Target Detection Methodologies
- Geological Modeling and Analysis
- Industrial Vision Systems and Defect Detection
- Image Retrieval and Classification Techniques
- Brain Tumor Detection and Classification
- 3D Surveying and Cultural Heritage
- Advanced Measurement and Detection Methods
- Handwritten Text Recognition Techniques
Wuhan University
2024
Huaneng Clean Energy Research Institute
2022
PLA Information Engineering University
2016-2022
Shanghai Jiao Tong University
2020
Ministry of Education of the People's Republic of China
2020
Nankai University
2016
Semi-supervised learning improves semantic segmentation performance by leveraging unlabeled data, thereby significantly reducing labeling costs. Previous semi-supervised (S4) methods explored perturbations at the image level but neglected to adequately utilize multi-scale information. When labeled information is insufficient, scale variation between different objects makes instances with extreme scales even more difficult. To address this issue, we propose ScaleMatch, which aims learn...
The damage of wind turbine blades is one the main problems restricting power development. Object detection can identify damaged regions and diagnose types. To handle high-resolution blade images, this article presents a novel efficient, accurate detector (EADD) for images. proposed method adopts Single Shot MultiBox Detector (SSD) as framework offers an improved ResNet backbone. Firstly, backbone uses dense connection blocks consisting factorized depth-wise separable bottleneck (FDSB)...
Semantic segmentation labels each pixel in high-resolution remote sensing (HRRS) images with a category. To tackle the large size and complexity of HRRS images, this letter presents novel multiscale feature aggregation lightweight network (MFALNet) for semantic segmentation. Unlike standard convolution, asymmetric depth-wise separable convolution residual (ADCR) unit is used to reduce parameter makes optimized structure deeper but less complex. The proposed an encoder–decoder structure,...
Since target points could be generated quickly and accurately by off-line programming in the industrial robot machining, it is used widely robots. Due to installation error mismachining tolerance which causes offset of processing paths actual machining process, registration algorithms for path calibration would involved. While traditional approach converge local optimal solution, an improved algorithm based on geometric properties point clouds proposed paper. Firstly, feature sets are...
Scene classification is an important tool for remote sensing image interpretation, and it has fundamental applications in research industry. However, given complex backgrounds scale variations, images have large intraclass diversity interclass similarity, which bring challenges to accurate of images. We proposed a scene method using joint learning multiscale attention alleviate the aforementioned problems. To fully utilize information improve adaptability objects with various sizes,...
With the rapid development of aerospace and remote sensing technology, various high-resolution Earth Observation Systems (EOS) are widely used in economic, social, military other fields playing an increasingly prominent role construction Digital national strategic planning. The normal operation system is premise high quality data acquisition. Compared with ground observation mode, EOS itself surrounding environment more complex, its control mainly depends on all kinds Space Situational...
The purpose of this paper is to solve the problems traditional single system for interpretation and draughting such as inconsistent standards, function, dependence on plug-ins, closed low integration level. On basis comprehensive analysis target elements composition, map representation similar features, a 3D integrated service platform multi-source, multi-scale multi-resolution geospatial objects established based HTML5 WebGL, which not only integrates object recognition, access, retrieval,...
Bi-temporal high resolution (BHR) remote sensing images assemble abundant information and ample ground details. Recently, scene parsing based on bi-temporal has played an important role in many fields. Nevertheless, for the purpose of understanding, conventional semantic segmentation can neither utilized feature relevance adequately nor tackle problem low-speed inference. In this paper, we propose a confused matrix unit (CMU) BHR images. The designed is capable representing transformation...