A5019609937- Remote-Sensing Image Classification
- Advanced Image Fusion Techniques
- Remote Sensing in Agriculture
- Image and Signal Denoising Methods
- Millimeter-Wave Propagation and Modeling
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Remote Sensing and Land Use
- Remote Sensing and LiDAR Applications
- Image Enhancement Techniques
- Advanced SAR Imaging Techniques
- Advanced Wireless Network Optimization
- Advanced Image and Video Retrieval Techniques
- Advanced MIMO Systems Optimization
- Automated Road and Building Extraction
- Wireless Communication Networks Research
- Radio Wave Propagation Studies
- Infrared Target Detection Methodologies
- Advanced Measurement and Detection Methods
- Metal Forming Simulation Techniques
- Satellite Communication Systems
- Magnetic Field Sensors Techniques
- Video Coding and Compression Technologies
- Icing and De-icing Technologies
- Laser and Thermal Forming Techniques
- Advanced Optical Sensing Technologies
State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing
2019-2025
Wuhan University
2019-2025
North China Electric Power University
2020-2024
Harbin Engineering University
2018
University at Buffalo, State University of New York
2012
New York University
2012
University of Science and Technology of China
2012
Shanghai Jiao Tong University
2007
Beijing University of Technology
2006
Deep convolution neural network (DCNN) is among the most effective ways of performing land use classification high-resolution remote sensing images. Land by fusing optical and synthetic aperture radar (SAR) images has broad application prospects, but related research studies are few. In this study, we developed first largest joint SAR dataset, WHU-OPT-SAR, covering an area approximately 50,000 km2, designed a multimodal-cross attention (MCANet). MCANet comprises three core modules:...
Abstract The urban scaling theory (UST) strives for a universal taxonomy that depicts relationships among indicators (e.g. energy consumption, economic output) with city size. However, the lack of international agreement on definitions and statistics complicates cross-country comparisons performance. Remote sensing provides uniform standard measuring cities around world. To scrutinize consistency UST, we quantified changes in remotely sensed built-up areas (UBA) nighttime lights (NTL)...
Surface deformation data can be used to provide early warnings of geohazards and are useful in a variety research fields. The Small BAseline Subset InSAR (SBAS-InSAR) boosts the sampling rate improves accuracy extraction by restricting temporal spatial baselines. However, various factors, such as types ground objects seasons, affect coherence between SAR images. Traditional SBAS-InSAR employs fixed baseline, which does not guarantee good might lead decorrelation. In this paper, we propose...
Hyperspectral remote sensing images (HSIs) have been applied in urban planning, environmental monitoring, and other fields. However, they are susceptible to noise interference, such as Gaussian noise, stripe, mixed noises, from various factors the imaging process, which greatly limits their applications. Although previous efforts improve HSI quality achieved remarkable results, there still many challenges be solved. To avoid poor generalization ability stripe removal performance of network...
GaoFen-3 (GF-3) is a C-band multipolarization synthetic aperture radar (SAR) satellite with 12 imaging modes. However, its initial positioning accuracy remains unsatisfactory, thereby hindering use for large-area surveying and mapping. This study proposes block orthorectification method without ground control points (GCPs) using the GF-3 Fine strip II (FSII) mode. To address challenges efficiency of this method, an integrated was developed to conduct processing large-scale images GCPs....
The ice, cloud, and land elevation satellite-2 (ICESat-2) is equipped with a photon-counting laser altimeter system demonstrates outstanding ability to measure elevations in the ever-changing earth. However, ICESat-2 data contain several noise photons affected by solar returns, there are no reference of signal or for evaluating performance denoising algorithms. In this letter, we propose self-adaptive algorithm based on genetic (SADA-GA) data, which uses real-coded adaptively search global...
Sensor instability, dark currents, and other factors often cause stripe noise corruption in hyperspectral remote sensing images severely limit their application practical purposes. Previous studies have proposed numerous destriping algorithms that yielded impressive results. Although most are based on the premise of additive noise, a few focused directly multiplicative noise. This article fully analyzes characteristics OHS-01 proposes removal method. Specifically, is tackled by performing...
Optical satellites are affected by factors such as seasonal and atmospheric variation, illumination, sensor distortion. Thus, satellite images covering large-scale area often show conspicuous color differences, resulting in poor continuity of the mosaicked image. This study proposes a novel combined model correction (CMCC) method for high-resolution optical images, which constructively combines defogging with radiation model. First, this analyzed feasibility using easily available...
Remote sensing images, especially hyperspectral images (HSIs), are extremely vulnerable to random noise and stripe noise. As a key aspect of HSI data quality improvement, removal has always been pervasive issue in remote image processing. Convolutional neural networks have applied for destriping. However, the existing methods lose stripe-free component original certain extent. These models also ignore global spatial context correlation between information spectral information. Therefore, we...
In hyperspectral imaging (HSI), stripe noise is one of the most common types that adversely affects its application. Convolutional neural networks (CNNs) have contributed to state-of-the-art performance in HSI destriping given their powerful feature extraction and learning capabilities. However, it difficult obtain paired training samples for real data. Most CNN methods construct a dataset with simulated network training. when data complex, model constrained. To solve this problem, study...
Clustered delay line (CDL) model is a link-level evaluation with the characteristics of low complexity and relatively accuracy. In this paper, modeling method proposed to improve accuracy applicability existing CDL models in Third Generation Partnership Project (3GPP). To validate model, 28 GHz millimeter wave channel measurement was conducted an outdoor substation scenarios, improved established by identifying different clusters. The are validated validating large scale parameters power...
Abstract. Affected by factors such as season, illumination, atmospheric and sensor distortion, different satellite images often show obvious color difference, resulting in “stitching seams” at the edge of adjacent images, which seriously affects application images. This study proposes a novel consistency method for optical utilizing external reference. Firstly, we improved dark channel defogging combining with distribution characteristics used it to perform correction on images; Secondly,...
Shot classification is helpful for video analysis, in this paper, an algorithm proposed. First, key frames are extracted equal interval a shot; then frame, dominant colors obtained by hue histogram, furthermore, the variance of and ratio clustered computed, which corresponding frame could be classified into playing or non-playing frame; finally, shot percentage frames. Experimental results confirm efficiency proposed
Based on the Bird’s Nest 28 GHz channel measurement data, this paper proposes a spatially consistent large-scale parameter generation algorithm based spatial recursion, and compares it with traditional WINNER II algorithm. This uses multi-dimensional normal distribution specific recursion generates parameters of each simulation point sequentially, to realize auto-correlation cross-correlation characteristics parameters. The results show that, premise achieving consistency, recursive are...