A5078116487- Advanced Image Fusion Techniques
- Image Enhancement Techniques
- Remote-Sensing Image Classification
- Advanced Image Processing Techniques
- Image and Signal Denoising Methods
- Remote Sensing and Land Use
- AI in cancer detection
- Visual Attention and Saliency Detection
- Remote Sensing in Agriculture
- Advanced Neural Network Applications
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Advanced Vision and Imaging
- Radiomics and Machine Learning in Medical Imaging
- Video Surveillance and Tracking Methods
- Medical Image Segmentation Techniques
- Fire Detection and Safety Systems
- Image Processing Techniques and Applications
- Infrared Target Detection Methodologies
- Advanced Computational Techniques and Applications
- Brucella: diagnosis, epidemiology, treatment
- Advanced Algorithms and Applications
- Remote Sensing and LiDAR Applications
- Colorectal Cancer Screening and Detection
- Industrial Technology and Control Systems
Shandong Institute of Business and Technology
2014-2025
Harbin University
2025
Wuhan University of Technology
2024
Chinese Academy of Medical Sciences & Peking Union Medical College
2022-2024
Nanjing Institute of Astronomical Optics & Technology
2024
Beijing University of Posts and Telecommunications
2024
Shandong First Medical University
2023
Jilin University
2023
Institute of Computing Technology
2022
Harbin Engineering University
2022
Low-light images have low brightness and contrast, which presents a huge obstacle to computer vision tasks. image enhancement is challenging because multiple factors (such as brightness, artifacts, noise) must be considered simultaneously. In this study, we propose neural network—a progressive-recursive network (PRIEN)—to enhance low-light images. The main idea use recursive unit, composed of layer residual block, repeatedly unfold the input for feature extraction. Unlike in previous...
The absorption and scattering caused by the underwater medium degrade quality of optical imaging, which limits further development tasks. Recently, Transformer-based methods have shown same excellent performance as Convolutional Neural Networks (CNNs) in various vision tasks, but huge parameters such networks hinder their application deployment. In this paper, we propose a novel adaptive group attention (AGA), can dynamically select visually complementary channels based on dependencies,...
The pansharpening entails obtaining images with uniform spectral distribution and rich spatial details by fusing multispectral panchromatic images, which has become a major image fusion problem in the field of remote sensing. Convolutional neural networks are widely used processing. We propose transformer-based regression network (DR-NET) architecture. first stage was feature extraction, entailed extracting information from images. second fusion, integrating extracted information. In third...
With the development of imaging systems and satellite technology, higher quality high-resolution RS images are being applied in building change detection (BCD) techniques. Methods based on convolutional neural network (CNN) have achieved excellent success BCD techniques due to their feature discrimination ability. However, CNN relies heavily geometry prior conditions is limited by size convolution kernel, making it easy ignore global information. This makes difficult capture long-range...
Change detection (CD) of high-resolution remote sensing (RS) images is a basic task in RS image processing tasks. In recent years, CD tasks have made many attempts pure convolutional networks, attention mechanism, and transformer, achieved good results. Based on the power transformers, we hope to find method that can handle details better has generalization ability. this article, propose dual-feature mixed attention-based transformer network (DMATNet). First, adopt extraction method, using...
Automatic extraction of water bodies from various satellite images containing complex targets is a very important and challenging task in remote sensing image interpretation. In recent years, convolutional neural networks (CNNs) have become an choice the field semantic segmentation images. However, generic CNN models present many problems when performing body segmentation, such as: (1) blurred boundaries; (2) difficulty accommodating different scales rivers, often losing information about...
Edge accuracy and positional are the two goals pursued by medical image segmentation. In clinical medicine diagnosis research, these enable segmentation techniques to help in effective determination of lesions lesion analysis. At present, U-Net has become most important network field segmentation, technologies used various achievements derived from its architecture, which also proves practice that structure proposed is effective. We have found a large number experiments classical networks...
Pansharpening methods play a crucial role for remote sensing image processing. The existing pansharpening methods, in general, have the problems of spectral distortion and lack spatial detail information. To mitigate these problems, we propose multiscale hybrid attention Transformer network (MHATP-Net). In proposed network, shallow feature (SF) is first acquired through an SF extraction module (SFEM), which contains convolutional block (CBAM) dynamic convolution blocks. CBAM this can filter...
Remote sensing images are characterized by high dimensionality, complex textures, and large scales. Traditional Convolutional Neural Network (CNN) methods may overlook spatial relationships contextual information among pixels when dealing with remote data. Therefore, Graph Networks (GCN) have emerged as a promising solution. In this paper, we propose Contextual Spatial Awareness Sensing Image Change Detection Based on Convolution interaction (CSAGC). We aim to enhance the handling of...
Currently, the task of remote sensing image segmentation still faces some challenges, such as variations in illumination, shadows, and occlusions present images. Additionally, there may be similarities confusions between different types terrain features. In this paper, we aim to explore how utilize information exchange multiple modalities reduce impact interfering factors. To fully exploit complementary modalities, establish an mechanism optical images (visible light + infrared) features...
Benefiting from continuous innovations in deep learning (DL) algorithms, the accuracy of building change detection(BCD) remote sensing (RS) has significantly improved. Numerous networks combining CNN and Transformer architectures have emerged, yet effectively balancing local detail global context features remains a topic ongoing discussion. Furthermore, accurately leveraging edge information within RS images to enhance recognition structural changes buildings is another critical challenge....
In the field of remote sensing change detection, accurately capturing temporal information and efficiently integrating multi-level is a major challenge. order to extend sensory domain optimize fusion, model able capture Temporal-Spatial features more improve accuracy detection. this paper, we propose Multi-scale Graph Attention Network (TSMGA), specifically, TSMGA employs pair pre-trained ResNet18 for effective multiscale feature extraction, in enhance disparity bi-temporal images, also...
We carried out limited enzymatic hydrolysis with trypsin on rice bran protein (RBP) pretreated by high hydrostatic pressure (HHP) in this study. The effects of the degree (DH) structural and emulsifying properties were investigated. results indicated that molecular structure RBP changed after hydrolysis. hydrolysate (RBPH, DH8) exhibited a better distribution, smaller particle size (200.4 nm), activity index (31.82 m2/g), an improved stability (24.69 min). RBPH emulsions different DH (0-12)...
With the remarkable success of change detection (CD) in remote sensing images context deep learning, many convolutional neural network (CNN) based methods have been proposed. In current research, to obtain a better modeling method for and capture more spatiotemporal characteristics, several attention-based transformer (TR)-based Recent research has also continued innovate on TR-based methods, new Most them require huge number calculation achieve good results. Therefore, using mehtod while...