A5011573110- Advanced Image Processing Techniques
- Advanced Image Fusion Techniques
- Image and Signal Denoising Methods
- Image Enhancement Techniques
- Image Processing Techniques and Applications
- Infrared Target Detection Methodologies
- Digital Holography and Microscopy
- Metallurgy and Material Forming
- Advanced Neural Network Applications
- Wireless Power Transfer Systems
- Innovative Energy Harvesting Technologies
- Spectroscopy Techniques in Biomedical and Chemical Research
- Advanced Image and Video Retrieval Techniques
- Advanced Vision and Imaging
- Microstructure and mechanical properties
- Microstructure and Mechanical Properties of Steels
- Remote-Sensing Image Classification
- Cell Image Analysis Techniques
Harbin Institute of Technology
2022-2025
Ningxia University
2022
Yanshan University
2021
Wuhan University
2018-2020
Wuhan University of Technology
2017
Visible images contain rich texture information, whereas infrared have significant contrast. It is advantageous to combine these two kinds of information into a single image so that it not only has good contrast but also contains details. In general, previous fusion methods cannot achieve this goal well, where the fused results are inclined either visible or an image. To address challenge, new framework called generative adversarial network with multiclassification constraints (GANMcC)...
Image deblurring continues to achieve impressive performance with the development of generative models. Nonetheless, there still remains a displeasing problem if one wants improve perceptual quality and quantitative scores recovered image at same time. In this study, drawing inspiration from research transformer properties, we introduce pretrained transformers address problem. particular, leverage deep features extracted vision (ViT) encourage images be sharp without sacrificing measured by...
In this paper, we propose a new end-to-end model, called dual-discriminator conditional generative adversarial network (DDcGAN), for fusing infrared and visible images of different resolutions. Unlike the pixel-level methods existing deep learning-based methods, fusion task is accomplished through process between generator two discriminators, in addition to specially designed content loss. The trained generate real-like fused fool discriminators. discriminators are calculate JS divergence...
In this paper, we introduce MaeFuse, a novel autoencoder model designed for Infrared and Visible Image Fusion (IVIF). The existing approaches image fusion often rely on training combined with downstream tasks to obtain high-level visual information, which is effective in emphasizing target objects delivering impressive results quality task-specific applications. Instead of being driven by tasks, our called MaeFuse utilizes pretrained encoder from Masked Autoencoders (MAE), facilities the...
We propose Pixel2Pixel, a novel zero-shot image denoising framework that leverages the non-local self-similarity of images to generate large number training samples using only input noisy image. This employs compact convolutional neural network architecture achieve high-quality denoising. Given single observed image, we first aim obtain multiple with different noise versions. ensure content remains as consistent possible true signal while keeping independent. Specifically, construct pixel...
This paper provides a review of the NTIRE 2021 challenge targeting defocus deblurring using dual-pixel (DP) data. The goal this single-track was to reduce spatially varying blur present in images captured with shallow depth field. used were obtained DP sensor that provided pair views per image. Submitted solutions evaluated conventional signal processing metrics, namely peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). Out 185 registered participants, nine...
Reducing the defocus blur that arises from finite aperture size and short exposure time is an essential problem in computational photography. It very challenging because kernel spatially varying difficult to estimate by traditional methods. Due its great breakthrough low-level tasks, convolutional neural networks (CNNs) have been introduced deblurring achieved significant progress. However, previous methods apply same learned for different regions of blurred images, thus it handle nonuniform...
Image fusion is famous as an alternative solution to generate one high-quality image from multiple images in addition restoration a single degraded image. The essence of integrate complementary information source images. Existing methods struggle with generalization across various tasks and often require labor-intensive designs, which it difficult identify extract useful due the diverse requirements each task. Additionally, these develop highly specialized features for different downstream...
In this research, we introduce MaeFuse, a novel autoencoder model designed for infrared and visible image fusion (IVIF). The existing approaches often rely on training combined with downstream tasks to obtain high-level visual information, which is effective in emphasizing target objects delivering impressive results quality task-specific applications. however, deviates from the norm. Instead of being driven by tasks, our utilizes pretrained encoder Masked Autoencoders (MAE), facilities omni...
In our daily life environment, there are a lot of micro energy such as vibration, low thermal energy. the past due to limitations technical capabilities, this has not been effectively collected and utilized. With people's increasing attention effective use environmental protection, clean, renewable have become focus research in related fields. paper, by means electromagnetic vibration acquisition system at home abroad, working principle collector is clarified. On basis, designed realized, we...
Image deblurring continues to achieve impressive performance with the development of generative models. Nonetheless, there still remains a displeasing problem if one wants improve perceptual quality and quantitative scores recovered image at same time. In this study, drawing inspiration from research transformer properties, we introduce pretrained transformers address problem. particular, leverage deep features extracted vision (ViT) encourage images be sharp without sacrificing measured by...
The defocus deblurring raised from the finite aperture size and exposure time is an essential problem in computational photography. It very challenging because blur kernel spatially varying difficult to estimate by traditional methods. Due its great breakthrough low-level tasks, convolutional neural networks (CNNs) have been introduced achieved significant progress. However, they apply same for different regions of blurred images, thus it handle these nonuniform images. To this end, study...
In the rolling process of heavy cylinders, deformation section is subjected to effects compression and shear. order analyze influences shear effect on microstructure evolution characteristics, a mathematical model was established simulated. Firstly, shear-compression specimen (SCS) ordinary cylinder specimens were designed, high-temperature experiments carried out considering established; then, program based finite element software developed simulate process, feasibility development verified...