A5101568170- Geophysical Methods and Applications
- Geophysical and Geoelectrical Methods
- Seismic Waves and Analysis
- Geoscience and Mining Technology
- Remote-Sensing Image Classification
- Automated Road and Building Extraction
- Geomechanics and Mining Engineering
- Rock Mechanics and Modeling
- Earthquake Detection and Analysis
- Landslides and related hazards
- Mining and Gasification Technologies
- Advanced Image Processing Techniques
- Remote Sensing and LiDAR Applications
- Civil and Geotechnical Engineering Research
- Remote Sensing and Land Use
- Geotechnical and Geomechanical Engineering
- Coal Properties and Utilization
- Grouting, Rheology, and Soil Mechanics
- Advanced Sensor Technologies Research
- Homotopy and Cohomology in Algebraic Topology
- Nonlinear Optical Materials Research
- Image and Signal Denoising Methods
- Advanced Image Fusion Techniques
- Solar Thermal and Photovoltaic Systems
- Heat Transfer and Optimization
China University of Mining and Technology
2017-2024
Zhejiang University of Technology
2019-2022
Southeast University
2016
Zhejiang Tongji Vocational College of Science and Technology
2013
Shandong University
2011
Chonnam National University Hospital
2011
Taiyuan University of Technology
2011
Central South University
2006
A lensless blood cell counting system integrating microfluidic channel and a complementary metal oxide semiconductor (CMOS) image sensor is promising technique to miniaturize the conventional optical lens based imaging for point-of-care testing (POCT). However, such has limited resolution, making it imperative improve resolution from system-level using super-resolution (SR) processing. Yet, how towards better detection recognition with low cost of processing resources without degrading...
Availability of reliable delineation urban lands is fundamental to applications such as infrastructure management and planning. An accurate semantic segmentation approach can assign each pixel remotely sensed imagery a ground object class. In this paper, we propose an end-to-end deep learning architecture perform the pixel-level understanding high spatial resolution remote sensing images. Both local global contextual information are considered. The contexts learned by residual net,...
Building extraction is a basic task in the field of remote sensing, and it has also been popular research topic past decade. However, shape semantic polygon generated by segmentation irregular does not match actual building boundary. The boundary buildings edge detection difficulty ensuring continuity integrity. Due to aforementioned problems, we cannot directly apply results many drawing tasks engineering applications. In this paper, propose novel convolutional neural network (CNN) model...
This work aims to remotely monitor the grouting process of double-layered wellbores in deep coal mining area by proposing a new geophysical electrical potential inversion imaging method. The proposed method is able evaluate quality between inner and outer walls inverting geologic potential. Th challenge lies here that thickness space extremely narrow when comparing with wellbore diameter. To solve this problem, constrained model developed precisely generate three-dimensional (3D) resistivity...
Deep convolutional neural network (DCNN)-based methods have shown great improvements in building extraction from high spatial resolution remote sensing images. In this paper, we propose a postprocessing method based on DCNNs for extraction. Specifically, regions and boundaries are learned simultaneously or separately by DCNNs. The predicted then combined the to produce final regions. addition, introduce manually labeled dataset images detection, XIHU dataset. This is used experiments...
With the increase in coal mining depth, water hazards deep, strong karst aquifers have seriously threatened safety of mine production. Dynamic monitoring floor damage during seam under complex conditions is great significance for prediction inrush at working face and improving methods. Based on parallel electrical technology, a constrained time-lapse resistivity inversion method was adopted to process data floor. The background results were used as an priori model act subsequent data, fed...
Pansharpening, which fuses the panchromatic (PAN) band with multispectral (MS) bands to obtain an MS image spatial resolution of PAN images, has been a popular topic in remote sensing applications recent years. Although deep-learning-based pansharpening algorithm achieved better performance than traditional methods, fusion extracts insufficient information from image, producing low-quality pansharpened images. To address this problem, paper proposes novel progressive PAN-injected method...
Abstract Deformation and rupture of rock mass under loading cause the variation electric potential. Response characteristics self-potential stress during complete stress-strain process red sandstones play an important role in evaluating state sandstone on basis self-potential. Experimental results demonstrate that uniaxial compression is linearly correlated with difference before first inflection point initial stage loading. The average rate 0.1325 mV MPa−1. As pressure gradually increases...
With the continuous development of urban underground space, non-destructive methods for diaphragm wall leakage detection are becoming increasingly important. Recent studies have shown that apparent resistivity, stimulated current, and self-potential can be used as indicator parameters leak detection. The seepage evaluation a subway station was carried out using above parameters. response characteristics were obtained before after excavating wall. results by this study reveal an area is very...
Image segmentation is a basic task of Object-Based Analysis (OBIA) in remote sensing. Traditionally, the algorithms for this are mostly based on region merging procedure. But segmented objects may correlate poorly with actual object boundary. Inspired by recent remarkable improvements edge detection deep learning, we propose semantic edge-guided learning method to extract meaningful geographic from High-resolution Remote Sensing (HRS) images. The consists three stages: first stage,...
As the depth of coal mining increases, water hazards in under deep and complicated conditions become increasingly serious. The dynamic monitoring roof floor failure process complex is great significance to prediction inrush at face improvement methods. Based on parallel electrical technology, combining dual-mode electrode data collection method, we monitor research geoelectric field characteristics full space seam simultaneously. Additionally, obtain synchronous response resistivity...
Supply voltage compensation is important to keep the accuracy for hot wire/hot film gas flow sensors.A experiment device established, two steps curve fitting presented determine functional relations among flow、sensor output and supply voltage, coefficient of polynomial determined by Least Square method, thus eliminating influence variations.After compensation, error reduced 2%.The algorithm stabilized, easily be fullfilled in microcontroller unit.