A5057327033- Advanced Image and Video Retrieval Techniques
- Robotics and Sensor-Based Localization
- Remote-Sensing Image Classification
- Image and Object Detection Techniques
- Advanced Neural Network Applications
- 3D Surveying and Cultural Heritage
- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Image Retrieval and Classification Techniques
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Advanced Vision and Imaging
- Remote Sensing and LiDAR Applications
- Advanced SAR Imaging Techniques
- Remote Sensing and Land Use
- Infrared Target Detection Methodologies
- COVID-19 diagnosis using AI
- Satellite Image Processing and Photogrammetry
- Radar Systems and Signal Processing
- Automated Road and Building Extraction
- Aerospace and Aviation Technology
- Wireless Signal Modulation Classification
- Aerospace Engineering and Control Systems
- Video Surveillance and Tracking Methods
- Lung Cancer Diagnosis and Treatment
- Advanced Measurement and Detection Methods
National University of Defense Technology
2014-2024
State Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System
2021
Hangzhou Dianzi University
2021
National Engineering Research Center of Electromagnetic Radiation Control Materials
2021
Institute of Navigation
2020
Automatic registration of optical and synthetic aperture radar (SAR) remote sensing images has been extensively researched for decades. However, fast robust multimodal image is still a challenge due to the significant region differences in many scenarios (e.g. city sea coexist single harbor imaging). To address this problem, paper proposes novel optical-to-SAR method called FED-HOPC, which integrates Fourier transform (FFT) weighted edge density (WED) map into histogram oriented phase...
<title>Abstract</title> Oriented object detection is one of the most fundamental and challenging tasks in remote sensing, aiming to locate classify objects with arbitrary orientations.Recent advancements deep learning have significantly enhanced capabilities oriented detection.Given rapid development this field, paper presents a comprehensive survey recent advances detection.To be specific, we begin by tracing technical evolution from horizontal higlighting specific challenges, including...
Synthetic aperture radar (SAR) is a kind of high-resolution imaging radar, which can work in any weather conditions and often used for target positioning. Airborne SAR has the advantages flexible scanning geometry high spatial resolution compared with Spaceborne SAR, but quality image positioning accuracy affected by air turbulence mechanical vibration. In practical applications, it vital to improve SAR's geometric without ground control points or other reference information. When single...
Arbitrary-oriented vehicle detection via aerial imagery is essential in remote sensing and computer vision, with various applications traffic management, disaster monitoring, smart cities, etc. In the last decade, we have seen notable progress object natural imagery; however, such development has been sluggish for airborne imagery, not only due to large-scale variations spins/appearances of instances but also scarcity high-quality datasets, which could reflect complexities challenges...
Multi-modal image matching is crucial in aerospace applications because it can fully exploit the complementary and valuable information contained amount diversity of remote sensing images. However, remains a challenging task due to significant non-linear radiometric, geometric differences, noise across different sensors. To improve performance heterologous matching, this paper proposes normalized self-similarity region descriptor extract consistent structural information. We first construct...
A robust and accurate aircraft pose estimation method is proposed in this paper. The reflects the flight status of measurement great importance many aerospace applications. This work aims to establish a universal framework estimate based on generic geometry structure features. In our method, line features are extracted describe an single images exploited form groups for recognition. Parallel clustering utilized detect fuselage reference bilateral symmetry property provides important...
Image matching between optical and SAR images remains challenging due to their different imaging principles. Despite recent developments in image methods for optical-to-SAR images, few have seen practical application success the significant geometric radiometric differences two modalities. In this letter, we propose a new method called Orientated Modality Independent Region Descriptor (OMIRD) overcome these challenges. Our template-based approach effectively extracts consistent structural...
Aircraft pose estimation is a necessary technology in aerospace applications, and accurate parameters are the foundation for many tasks. In this paper, we propose novel method straight wing aircraft without relying on 3D models or other datasets, two widely separated cameras used to acquire information. Because of large baseline long-distance imaging, feature point matching difficult inaccurate configuration. our method, line features extracted describe structure images, performed based...
Optical-SAR image matching is a fundamental task for fusion and visual navigation. However, all large-scale open SAR dataset methods development are collected from single platform, resulting in limited satellite types spatial resolutions. Since images captured by different sensors vary significantly both geometric radiometric appearance, existing may fail to match corresponding regions containing the same content. Besides, most of datasets have not been categorized based on characteristics...
Accurate segmentation of lung cancer in pathology slides is a critical step improving patient care. We proposed the ACDC@LungHP (Automatic Cancer Detection and Classification Whole-slide Lung Histopathology) challenge for evaluating different computer-aided diagnosis (CADs) methods on automatic cancer. The 2019 focused (pixel-wise detection) tissue whole slide imaging (WSI), using an annotated dataset 150 training images 50 test from 200 patients. This paper reviews this summarizes top 10...
With the rapid development of unmanned aerial vehicle technology, vehicles have been widely used to obtain ground remote sensing images. Due limited field view a single image, it is necessary stitch images large-scale scene image apply various practical applications. This paper introduces two-stage stitching algorithm combining geographic coordinate information and features. The sequence divided into intra-line inter-line stitching. Firstly, we use coordinates features calculate...
In this paper, we propose a new method to automatically extract the main structure (fuselage and wings) of straight wing aircraft in photographic images using line clustering. We exploit location, length orientation features (parallel constraint) segments group them into clusters. The different clusters correspond aircraft, can fuselage accordingly. adopt segment detector features, mean-shift k-means++ approaches are applied segments. experimental results demonstrate that our is effective...
In this paper, we exploit prior information from global positioning systems and inertial measurement units to speed up the process of large scene reconstruction images acquired by Unmanned Aerial Vehicles. We utilize weak pose intrinsic parameter obtain projection matrix for each view. As compared unmanned aerial vehicles' flight altitude, topographic relief can usually be ignored, assume that is flat use perspective camera get projective transformations between two views. Furthermore,...
In this paper, first an overview and explanation of the lateral single-axis Oblique Active Tilting (l-sOAT) control structure is presented. A complete dynamic model vertical take-off landing (VTOL) unmanned aerial vehicles (UAVs) having longitudinal rotor tilting mechanism developed using a Newton-Euler formulation for l-sOAT mechanism. Then, theoretical analysis vehicle response stability pitch motion described simulated derived model. The validity proposed its verified by simulation experiments.