A5031296460- Infrared Target Detection Methodologies
- Remote Sensing in Agriculture
- Remote Sensing and LiDAR Applications
- Remote-Sensing Image Classification
- Advanced Neural Network Applications
- Calibration and Measurement Techniques
- Remote Sensing and Land Use
- Advanced Image Fusion Techniques
- 3D Surveying and Cultural Heritage
- Image Enhancement Techniques
- Advanced Measurement and Detection Methods
- Thermography and Photoacoustic Techniques
- Wood and Agarwood Research
- Fire Detection and Safety Systems
- Anomaly Detection Techniques and Applications
- Domain Adaptation and Few-Shot Learning
- Space Satellite Systems and Control
- Adaptive optics and wavefront sensing
- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Optical Systems and Laser Technology
- Automated Road and Building Extraction
- Advanced SAR Imaging Techniques
- Random lasers and scattering media
- Synthetic Aperture Radar (SAR) Applications and Techniques
Nanjing University of Aeronautics and Astronautics
2022-2025
Dalian Neusoft University of Information
2023
Forest stock volume is the main factor to evaluate forest carbon sink level. At present, combination of multi-source remote sensing and non-parametric models has been widely used in FSV estimation. However, biodiversity natural forests complex, response spatial information images significantly reduced, which seriously affects accuracy To address this challenge, paper takes China’s Baishanzu Park with representative characteristics as research object, integrates survey data, SRTM Landsat 8...
Optical remote sensing has emerged as a crucial technique for earth observation. However, interference of clouds and fog can adversely affect the spatial spectral information images, presenting significant challenges in interpreting data limiting its availability. Moreover, existing methods addressing issue cloud occlusion primarily rely on either physical models or neural networks, lacking comprehensive integration advantages offered by both approaches. So, we propose an end-to-end...
Multispectral remote sensing images are widely used for monitoring the globe. Although thin clouds can affect all optical bands, influences of differ with band wavelength. When processing multispectral bands at different resolutions, many methods only remove in visible/near-infrared or rescale multiresolution to same resolution and then process them together. The former cannot make full use information, latter, rescaling will introduce noise. In this article, a deep-learning-based cloud...
ABSTRACT Building damage assessment in the face of natural disasters is crucial for economic development, disaster relief, and post‐disaster reconstruction. However, existing algorithms often overlook impact class when extracting difference features from high‐resolution pre‐ image pairs obtained through satellite remote sensing, without considering influence type, that is, different ways which affect buildings. To address this limitation, we propose U2DDS‐Net, a two‐stage model based on...
Understanding atmospheric motions and projecting climate changes depends significantly on cloud types, i.e., different types correspond to conditions, accurate classification can help forecasts meteorology-related studies be more effectively directed. However, of clouds is challenging often requires certain manual involvement due the complex forms dispersion. To address this challenge, paper proposes an improved method based a densely connected hybrid convolutional network. A dense...
Infrared detection, known for its robust anti-interference capabilities, performs well in all weather conditions and various environments. Its applications include precision guidance, surveillance, early warning systems. However, detecting infrared dim small targets presents challenges, such as weak target features, blurred with area percentages, missed detections, false alarms. To address the issue of insufficient feature information, this paper proposes a high-precision method based on...
The unknowability of the inner workings limits magnitude performance improvement ship target detection networks in synthetic aperture radar (SAR) images under Gaussian noise. However, none existing interpretation methods explain phenomenon network changes feature map can visually reflect image delivery network, and some metrics quantitatively characterize degree degradation a noise environment. So this paper, we propose comprehensive analysis method that integrates texture brightness...
Ship detection in large-scene offshore synthetic aperture radar (SAR) images is crucial civil and military fields, such as maritime management wartime reconnaissance. However, the problems of low rates, high false alarm missed rates ship targets SAR are due to occlusion objects or mutual among targets, especially for small targets. To solve this problem, study proposes a target model (TAC_CSAC_Net) that incorporates multi-attention mechanism detecting marine vessels images. Experiments were...
In the realm of non-cooperative space security and on-orbit service, a significant challenge is accurately determining pose abandoned satellites using imaging sensors. Traditional methods for estimating position target encounter problems with stray light interference in space, leading to inaccurate results. Conversely, deep learning techniques require substantial amount training data, which especially difficult obtain satellites. To address these issues, this paper introduces an innovative...
The sight of the high-orbit thermal infrared staring camera is concentrated and sensitive to temperature changes, therefore its image point positioning key ensuring geometric quality eliminating errors imaging system in orbit. According characteristics camera, internal external compensation models are proposed this paper based on two-dimensional pointing angle change, four kinds experimental schemes designed. results show that method has a good effect error.