A5035559321- Underwater Vehicles and Communication Systems
- Image Enhancement Techniques
- Advanced Image Processing Techniques
- Underwater Acoustics Research
- Medical Image Segmentation Techniques
- Robotics and Sensor-Based Localization
- Image Processing Techniques and Applications
- Cell Image Analysis Techniques
- Advanced Vision and Imaging
- Water Quality Monitoring Technologies
- Advanced Image Fusion Techniques
- Retinal Imaging and Analysis
- Oceanographic and Atmospheric Processes
- Fuel Cells and Related Materials
- Cerebrovascular and Carotid Artery Diseases
- Advanced Neural Network Applications
- Energy Efficient Wireless Sensor Networks
- Image and Video Quality Assessment
- Cardiovascular Health and Disease Prevention
- Advanced battery technologies research
- Hydrological Forecasting Using AI
- Image and Signal Denoising Methods
- Energy Harvesting in Wireless Networks
- Arctic and Antarctic ice dynamics
- IgG4-Related and Inflammatory Diseases
Jilin Province Science and Technology Department
2019-2024
Jilin University
2018-2024
Jilin Medical University
2023
ETH Zurich
2022
Lanzhou Jiaotong University
2019
Single-image super-resolution (SISR) is essential for improving the extraction of useful information from images captured in real world. Most existing methods generally assume that low-resolution (LR) are generated high-resolution (HR) through a known degradation model, such as bicubic downsampling. As result, these do not exhibit favorable performance on real-world with complex authentic degradations, significantly limiting their practicality, especially application scenarios where image...
Underwater acoustic and optical data fusion has been developed in recent decades. Matching of underwater images is a fundamental critical problem exploration because it usually acts as the key step many applications, such target detection, ocean observation, joint positioning. In this study, method matching same object was designed, consisting two steps. First, an enhancement used to enhance ensure accuracy results based on iterative processing estimate similarity. The are first...
Deployment of surface-level gateways holds potential as an effective method to alleviate high-propagation delays and high-error probability in underwater wireless sensor network (UWSN). This promise comes from reducing distances nodes using radio waves forward information a control station. In UWSN, dynamic energy efficient gateway deployment is required cope with the mobility while considering remote three-dimensional nature marine space. general, problems are usually modeled optimization...
With the cost reduction of underwater sensor network nodes and increasing demand for detection monitoring, near-land areas, shallow water lakes rivers have gradually tended to densely arranged nodes. In order achieve real-time most now visual sensors instead acoustic collect analyze optical images, mainly because cameras might be more advantageous when it comes dense networks. this article, image enhancement, saliency detection, calibration refraction model calculation are performed on video...
A new synthesis method for organic-inorganic hybrid Poly(vinylidene fluoride)-SiO2 cation-change membranes (CEMs) is proposed. This involves mixing tetraethyl orthosilicate (TEOS) and 3-mercapto-propyl-triethoxy-silane (MPTES) into a polyvinylidene fluoride (PVDF) sol-gel solution. The resulting slurry was used to prepare films, which were immersed in 0.01 M HCl, caused hydrolysis polycondensation between the MPTES TEOS. Si-O-Si polymers chains intertwined and/or penetrated PVDF skeleton,...
The seawater temperature has the characteristics of both spatial and temporal. To better predict temperature, we propose a convolutional long short-term memory recurrent neural network method based on thermohaline data. We use networks to extract features from data in near time domains obtain corresponding values these features. These are input into LSTM for temporal prediction temperature. was empirically evaluated compared RNN methods considering only single variable salinity. results show...
With the increasing demand for underwater resource exploration, three-dimensional (3D) reconstruction technology is important searching lost civilizations, shipwrecks, or seabed structures. However, faced with limitations of unmanned systems in terms energy, bandwidth, and transmission delay, 3D based on video streams as direct data will not work well. We propose a terminal image processing strategy to save time cost obtain scene information soon possible. Firstly, we an adaptive threshold...
Alignment between preoperative images (high-resolution magnetic resonance imaging, angiography) and intraoperative medical (digital subtraction is currently required in neurointerventional surgery. Treating a lesion usually guided by 2D DSA silhouette image. increase procedure time radiation exposure due to the lack of anatomical information, but information from MRA can be utilized compensate for this order improve efficiency. In paper, we abstract into problem relative pose correspondence...
Whole heart segmentation (WHS) supports cardiovascular disease (CVD) diagnosis, monitoring, treatment planning, and prognosis. Deep learning has become the most widely used method for WHS applications in recent years. However, of whole-heart structures faces numerous challenges including shape variability during cardiac cycle, clinical artifacts like motion poor contrast-to-noise ratio, domain shifts multi-center data, distinct modalities CT MRI. To address these limitations improve quality,...
Existing ocean visualization studies are usually conducted independently using either sonar or optical imaging. However, they have shortcomings in different engineering applications. Acoustic imaging is not comprehensive enough to show the details of target, and perspective extensive enough. Combining advantages optics acoustics, this paper proposes a joint method with acoustic communication-assisted decision-making image stitching, aiming improve information acquisition efficiency process....
Displaying the operation information of AUVs in real-time is an important subject Smart Ocean. The design monitoring systems often limited by application scenarios and technological developments. This paper develops underwater system based on acoustic sensor network for cooperative AUV. comprehensively considers transmission characteristics underwater, air, land uses long baseline positioning technology to a reasonable architecture workflow. Through experiments, we have verified that has...
Autonomous underwater vehicle (AUV), as an indispensable tool for exploring the ocean, has developed rapidly in recent years. With gradual deepening of people's exploration a single detection range and capacity are limited to meet large-scale long-term operational requirements. Therefore, coordinated operation multi-AUV become important direction development vehicles. The precise navigation result is premise AUV cluster execution task. A collaborative method based on flexible fusion ideas...
This paper proposes an underwater SLAM algorithm based on image sonar salient target detection to optimize the cumulative error of autonomous vehicle (AUV) inertial navigation system. It is helpful for long-range AUV in marine archaeology, rescue search, and hydrogeology. The application science other fields great significance. Due complexity environment unpredictability submerged targets, implementation information-assisted positioning has always been a challenging task. variability...
Segmentation of sidescan sonar image is a significant issue in underwater object detection and recognition. However, most prior methods only consider segmentation accuracy, ignoring false alarm rate, which plays vital role In this paper, robust accurate framework for proposed, balances preferred tradeoff between accuracy rate. The proposed method integrates an improved Smallest Of Constant False Alarm Rate (SO-CFAR) algorithm Maximum A Posteriori probability Markov Random Field model...
Wang, Ge Drs.; Guan, Wenxue; Peng, Xiaoyan; Yeh, Steven; Rishi, Pukhraj; Khosroshahi, ArezouEditor(s): McDonald, H. Richard Author Information