检索结果-内蒙古大学图书馆

Enhanced DehazeFormer: an image and video haze removal method based on SwinTransformer for superior quality

Multimedia Tools and Applications 2025年

作者： Ayoub, Abeer El-Shafai, Walid Aouf, Mohamed Abd El-Samie, Fathi E. Hamad, Ehab K. I. El-Rabaie, S. Electrical Engineering Department Faculty of Engineering Aswan University Aswan81542 Egypt Prince Sultan University Riyadh11586 Saudi Arabia Department of Electronics and Electrical Communication Engineering Faculty of Electronic Engineering Menoufia University Menouf32952 Egypt Biomedical Engineering Department Higher Technological Institute P.O. Box 228 Tenth of Ramadan City Egypt Department of Information Technology College of Computer and Information Sciences Princess Nourah bint Abdulrahman University Riyadh84428 Saudi Arabia

The visual quality of images and videos is significantly degraded by atmospheric particles such as smoke and dust, leading to the haze problem, which is characterized by low contrast and a whitish veil obscuring the content. Dehazing techniques have been developed to mitigate these effects and restore sharp, visually appealing imagery. This paper presents an enhanced dehazing approach, referred to as the Enhanced DehazeFormer technique, which integrates both preprocessing and dehazing stages to produce high-quality dehazed images. The proposed method incorporates a three-step preprocessing phase aimed at reducing noise and enhancing dynamic range, issues commonly introduced by measurement device errors and other external factors. Prior to dehazing, each image or video frame undergoes homomorphic filtering, followed by Contrast Limited Adaptive Histogram Equalization (CLAHE) and a fast dehazing algorithm to further improve visual quality. The dehazing stage utilizes an extended and customized Swin Transformer architecture, known as DehazeFormer, which is tailored specifically for haze removal tasks. The preprocessed frames are input into the modified Swin Transformer to generate dehazed outputs of superior visual quality. The proposed technique is thoroughly evaluated on visible images, Near-Infrared (NIR) frames, and real-world hazy datasets to assess its effectiveness. Evaluation metrics include entropy, Peak Signal-to-Noise Ratio (PSNR), Feature Similarity Index (FSIM), Feature Similarity Index Chromatic (FSIMC), edge intensity, average gradient, and correlation, all of which are used to quantitatively measure dehazing performance. Furthermore, histograms and spectral entropy analyses are employed to compare the proposed method against other dehazing techniques. A comparative analysis is conducted using five frames from each type of visible and NIR videos to assess the performance of the baseline DehazeFormer and the enhanced DehazeFormer technique. Additional eva

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Improved ML Models on Feature Extraction in HSI

Improved ML Models on Feature Extraction in HSI

引用

2020 International Conference on Computing and Information Technology, ICCIT 2020

作者： Narmatha, C. Sivaram, M. Manimegalai, P. Samikannu, Ravi Goel, Pallavi Murghai Khan, Juveriya University of Tabuk Faculty of Computers and Information Technology Department of Computer Engineering Saudi Arabia Lebanese French University Department of Computer Networking Erbil Kurdistan Region Iraq Karunya Institute of Technology and Sciences Biomedical Engineering Department Karunya Nagar India Botswana International University of Science and Technology Faculty of Electrical Engineering Botswana Botswana Galgotias University School of Computing Science and Engineering Greater Noida India Emirates College for Management and Information Technology Department of Information Technology United Arab Emirates

ISBN: (纸本)9781728126807

Hyper Spectral Image (HSI) is widely used in remote sensing applications that take into account thousands of spectral channel compositions over a single scene. HSI requires accurate models of learning to extract the features of the HSI. Due to the presence of its spatial and spectral resolution, the image learning model presents a core challenge due to its complicated nature of image frames. In order to assist it during the learning process, several attempts have been made to address its complicated nature. These methods, however, failed to provide HSI understanding. Using ML techniques addresses the issues due to the presence of mixed pixels, large amounts of data and limited samples of training. The process of ML addresses the complex relationship between image data. We present various methods of ML used to learn HSI in this paper. Initially, for different image processing techniques, we present an overview of different ML methods. A system review is then performed on the basis of deep learning based on various HSI learning models. Finally, for future study, we are discussing the challenges and possible directions. © 2020 IEEE.

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Frequency-Aware Diffusion Model for Multi-Modal MRI Image Synthesis

引用

Journal of imaging 2025年第5期11卷 152页

作者： Mingfeng Jiang Peihang Jia Xin Huang Zihan Yuan Dongsheng Ruan Feng Liu Ling Xia School of Computer Science and Technology Zhejiang Sci-Tech University Hangzhou 310018 China. School of Electrical Engineering and Computer Science The University of Queensland St. Lucia Brisbane QLD 4072 Australia. Department of Biomedical Engineering Zhejiang University Hangzhou 310027 China.

Magnetic Resonance Imaging (MRI) is a widely used, non-invasive imaging technology that plays a critical role in clinical diagnostics. Multi-modal MRI, which combines images from different modalities, enhances diagnostic accuracy by offering comprehensive tissue characterization. Meanwhile, multi-modal MRI enhances downstream tasks, like brain tumor segmentation and image reconstruction, by providing richer features. While recent advances in diffusion models (DMs) show potential for high-quality image translation, existing methods still struggle to preserve fine structural details and ensure accurate image synthesis in medical imaging. To address these challenges, we propose a Frequency-Aware Diffusion Model (FADM) for generating high-quality target modality MRI images from source modality images. The FADM incorporates a discrete wavelet transform within the diffusion model framework to extract both low- and high-frequency information from MRI images, enhancing the capture of tissue structural and textural features. Additionally, a wavelet downsampling layer and supervision module are incorporated to improve frequency awareness and optimize high-frequency detail extraction. Experimental results on the BraTS 2021 dataset and a 1.5T-3T MRI dataset demonstrate that the FADM outperforms existing generative models, particularly in preserving intricate brain structures and tumor regions while generating high-quality MRI images.

关键词： MRI image synthesis diffusion model multi-modal MRI wavelet transform

来源：评论

学校读者我要写书评

暂无评论

Real-Time Monitoring of NIR-Triggered Drug Release from Phase-Changeable Nanodroplets by Photoacoustic/Ultrasound Imaging

SSRN

引用

SSRN 2022年

作者： Cao, Yang Dumani, Diego S. Hallam, Kristina A. Ran, Haitao Emelianov, Stanislav Y. Chongqing Key Laboratory of Ultrasound Molecular Imaging Institute of ultrasound imaging Second Affiliated Hospital State Key Laboratory of Ultrasound in Medicine and Engineering Chongqing Medical University Chongqing400010 China Wallace H. Coulter Department of Biomedical Engineering Georgia Institute of Technology Emory University School of Medicine AtlantaGA30332 United States School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA30332 United States

Optical-responsive nanodroplets have been studied as a new mode of remotely controlled drug delivery. As a class of new emerging drug carriers, NIR-absorber-loaded perfluorocarbon nanodroplets can be converted into gas bubbles through laser stimulation, called optical droplet vaporization (ODV), which provides a potential strategy to deliver therapeutic agents to solid tumors on demand. However, there is a lack of suitable technologies to monitor these drug-loaded nanodroplet behaviors in vivo. In this study, ultrasound/photoacoustic imaging technology were applied to directly monitor optical-responsive, drug-loaded nanodroplets within the tissue. We explored the effects of laser energy, repetition rate, and number of pulses on the release profiles of the delivered drug as well as ultrasound and photoacoustic imaging signal-intensity curves. The conducted studies demonstrated that this noninvasive technology helped determine the optimum time-point for laser activation on accumulated drug-loaded nanodroplets within tumors, allowing for the potential to effectively treat tumors while minimizing drug-related toxicities. © 2022, The Authors. All rights reserved.

关键词： Tumors

来源：评论

学校读者我要写书评

暂无评论

A review on MR based human brain parcellation methods

arXiv

引用

arXiv 2021年

作者： Moghimi, Pantea Dang, Anh The Netoff, Theoden Lim, Kelvin Atluri, Gowtham Department of Neurobiology University of Chicago ChicagoIL United States Department of Electrical Engineering and Computer Science University of Cincinnati CincinnatiOH United States Department of Biomedical Engineering University of Minnesota MinneapolisMN United States Department of Psychiatry University of Minnesota MinneapolisMN United States

Brain parcellations play a ubiquitous role in the analysis of magnetic resonance imaging (MRI) datasets. Over 100 years of research has been conducted in pursuit of an ideal brain parcellation. Different methods have been developed and studied for constructing brain parcellations using different imaging modalities. More recently, several data-driven parcellation methods have been adopted from data mining, machine learning, and statistics communities. With contributions from different scientific fields, there is a rich body of literature that needs to be examined to appreciate the breadth of existing research and the gaps that need to be investigated. In this work, we review the large body of in vivo brain parcellation research spanning different neuroimaging modalities and methods. A key contribution of this work is a semantic organization of this large body of work into different taxonomies, making it easy to understand the breadth and depth of the brain parcellation literature. Specifically, we categorized the existing parcellations into three groups: Anatomical parcellations, functional parcellations, and structural parcellations which are constructed using T1-weighted MRI, functional MRI (fMRI), and diffusion-weighted imaging (DWI) datasets, respectively. We provide a multi-level taxonomy of different methods studied in each of these categories, compare their relative strengths and weaknesses, and highlight the challenges currently faced for the development of brain parcellations. © 2021, CC BY-NC-SA.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Surveillance of Pathogenic Bacteria on a Food Matrix Using Machine-Learning-Enabled Paper Chromogenic Arrays

SSRN

引用

SSRN 2023年

作者： Jia, Zhen Luo, Yaguang Wang, Dayang Holliday, Emma Sharma, Arnav Green, Madison M. Roche, Michelle R. Thompson-Witrick, Katherine Flock, Genevieve Pearlstein, Arne J. Yu, Hengyong Zhang, Boce Food Science and Human Nutrition Department University of Florida GainesvilleFL32611 United States Environmental Microbial and Food Safety Lab and Food Quality Lab U.S. Department of Agriculture Agricultural Research Service BeltsvilleMD20705 United States Department of Electrical and Computer Engineering University of Massachusetts LowellMA01854 United States School of Medicine Duke University DurhamNC27710 United States Department of Biomedical and Nutritional Sciences University of Massachusetts LowellMA01854 United States US Army Natick Soldier Research Development and Engineering Center NatickMA01760 United States Department of Mechanical Science and Engineering University of Illinois Urbana-Champaign UrbanaIL61801 United States

Global food systems can benefit significantly from real-time monitoring of microbial food safety, a task for which tedious operations, destructive sampling, and inability to monitor multiple pathogens remain challenging. This study reports significant improvements to a paper chromogenic array sensor - machine learning (PCA-ML) methodology sensing concentrations of volatile organic compounds (VOCs) emitted on a species-specific basis by pathogens, by streamlining dye selection, sensor fabrication, database construction, and machine learning and validation. This approach enables noncontact, time-dependent, simultaneous monitoring of multiple pathogens (Listeria monocytogenes, Salmonella, and E. coli O157:H7) at levels as low as 1 log CFU/g with over 90% accuracy. The report provides theoretical and practical frameworks demonstrating that chromogenic response, including limits of detection, depends on time integrals of VOC concentrations. The paper also discusses the potential for implementing PCA-ML in the food supply chain for different food matrices and pathogens, with species- and strain-specific identification. © 2023, The Authors. All rights reserved.

关键词： Supply chains

来源：评论

学校读者我要写书评

暂无评论

Implementing ASLNet V1.0: Progress and plans 11

Implementing ASLNet V1.0: Progress and plans

引用

11th Global Wordnet Conference, GWC 2021

作者： Lualdi, Colin P. Wright, Elaine Hudson, Jack Caselli, Naomi K. Fellbaum, Christiane SignSchool Inc. MadisonWI United States Department of Physics University of Illinois Urbana-ChampaignIL United States Department of Electrical Engineering Princeton University PrincetonNJ United States Program in Linguistics Princeton University PrincetonNJ United States Programs in Deaf Studies Boston University BostonMA United States Department of Computer Science Princeton University PrincetonNJ United States

ISBN: (纸本)9789464027310

We report on the development of ASLNet, a wordnet for American Sign Language (ASL). ASLNet V1.0 is currently under construction by mapping easy-to-translate ASL lexical nouns to Princeton WordNet synsets. We describe our data model and mapping approach, which can be extended to any sign language. Analysis of the 390 synsets processed to date indicates the success of our procedure yet also highlights the need to supplement our mapping with the "merge" method. We outline our plans for upcoming work to remedy this, which include use of ASL free-association data. © GWC 2021. All rights reserved.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Photoacoustic silk scaffolds for neural stimulation and regeneration

arXiv

引用

arXiv 2021年

作者： Zheng, Nan Fitzpatrick, Vincent Cheng, Ran Shi, Linli Kaplan, David L. Yang, Chen Division of Materials Science & Engineering Boston University BostonMA02215 United States Department of Biomedical Engineering Tufts University MedfordMA02215 United States Department of Chemistry Boston University BostonMA02215 United States Department of Electrical and Computer Engineering Boston University BostonMA02215 United States

Neural interfaces using biocompatible scaffolds provide crucial properties for the functional repair of nerve injuries and neurodegenerative diseases, including cell adhesion, structural support, and mass transport. Neural stimulation has also been found to be effective in promoting neural regeneration. This work provides a new strategy to integrate photoacoustic (PA) neural stimulation into hydrogel scaffolds using a nanocomposite hydrogel approach. Specifically, polyethylene glycol (PEG)-functionalized carbon nanotubes (CNT), highly efficient photoacoustic agents, are embedded into silk fibroin to form biocompatible and soft photoacoustic materials. We show that these photoacoustic functional scaffolds enable non-genetic activation of neurons with a spatial precision defined by the area of light illumination, promoting neuron regeneration. These CNT/silk scaffolds offered reliable and repeatable photoacoustic neural stimulation. 94% of photoacoustic stimulated neurons exhibit a fluorescence change larger than 10% in calcium imaging in the light illuminated area. The on-demand photoacoustic stimulation increased neurite outgrowth by 1.74-fold in a dorsal root ganglion model, when compared to the unstimulated group. We also confirmed that photoacoustic neural stimulation promoted neurite outgrowth by impacting the brain-derived neurotrophic factor (BDNF) pathway. As a multifunctional neural scaffold, CNT/silk scaffolds demonstrated non-genetic PA neural stimulation functions and promoted neurite outgrowth, providing a new method for non-pharmacological neural regeneration. Copyright © 2021, The Authors. All rights reserved.

关键词： Carbon nanotubes

来源：评论

学校读者我要写书评

暂无评论

Mid-infrared Chemical Imaging of Intracellular Tau Fibrils using Fluorescence-guided Computational Photothermal Microscopy

arXiv

引用

arXiv 2023年

作者： Zhao, Jian Jiang, Lulu Matlock, Alex Xu, Yihong Zhu, Jiabei Zhu, Hongbo Tian, Lei Wolozin, Benjamin Cheng, Ji-Xin Department of Electrical and Computer Engineering Boston University BostonMA02215 United States The Picower Institute for Learning and Memory Massachusetts Institute of Technology CambridgeMA02142 United States Department of Pharmacology and Experimental Therapeutics Boston University School of Medicine BostonMA02118 United States Department of Mechanical Engineering Massachusetts Institute of Technology CambridgeMA02142 United States Department of Physics Boston University BostonMA02215 United States State Key Laboratory of Luminescence and Applications Changchun Institute of Optics Fine Mechanics and Physics Chinese Academy of Sciences Changchun130033 China Department of Biomedical Engineering Boston University BostonMA02215 United States Photonics Center Boston University BostonMA02215 United States

Amyloid proteins are associated with a broad spectrum of neurodegenerative diseases. However, it remains a grand challenge to extract molecular structure information from intracellular amyloid proteins in their native cellular environment. To address this challenge, we developed a computational chemical microscope integrating 3D mid-infrared photothermal imaging with fluorescence imaging, termed Fluorescence-guided Bond-Selective Intensity Diffraction Tomography (FBS-IDT). Based on a low-cost and simple optical design, FBS-IDT enables chemical-specific volumetric imaging and 3D site-specific mid-IR fingerprint spectroscopic analysis of tau fibrils, an important type of amyloid protein aggregates, in their intracellular environment. Label-free volumetric chemical imaging of human cells with/without seeded tau fibrils is demonstrated to show the potential correlation between lipid accumulation and tau aggregate formation. Depth-resolved mid-infrared fingerprint spectroscopy is performed to reveal the protein secondary structure of the intracellular tau fibrils. 3D visualization of the β-sheet for tau fibril structure is achieved. Copyright © 2023, The Authors. All rights reserved.

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Model with Radiomics Analysis Integration for Glioblastoma Post-Resection Survival Prediction

arXiv

引用

arXiv 2022年

作者： Hu, Zongsheng Yang, Zhenyu Zhang, Haozhao Vaios, Eugene Lafata, Kyle Yin, Fang-Fang Wang, Chunhao Medical Physics Graduate Program Duke Kunshan University Jiangsu Kunshan215316 China Department of Radiation Oncology Duke University DurhamNC27710 United States Department of Radiology Duke University DurhamNC27710 United States Department of Electrical and Computer Engineering Duke University DurhamNC27710 United States

Purpose: To develop a novel deep learning model that integrates radiomics analysis in a multi-dimensional feature fusion workflow for glioblastoma (GBM) post-resection survival prediction Methods: A cohort of 235 GBM patients with complete surgical resection was divided into short-term/long-term survival groups with 1-yr survival time threshold. Each patient received a pre-surgery multi-parametric MRI exam with 4 scans: T1, contrast-enhanced T1 (T1ce), T2, and FLAIR. Three tumor subregions were segmented by neuroradiologists, and the whole dataset was divided into training, validation, and test groups following a 7:1:2 ratio. The developed model comprises three data source branches: in the 1st radiomics branch, 456 radiomics features (RF) were calculated from the three tumor subregions of each patient’s MR images;in the 2nd deep learning branch, an encoding neural network architecture was trained for survival group prediction using each single MR modality, and high-dimensional parameters from the last two network layers were extracted as deep features (DF). The extracted radiomics features and deep features were processed by a feature selection procedure to reduce the dimension size of each feature space. In the 3rd branch, patient-specific clinical features (PSCF), including patient age and three tumor subregions volumes, were collected from the dataset. Finally, data sources from all three branches were fused as an integrated input for a supporting vector machine (SVM) execution for survival group prediction. Different strategies of model design were investigated in comparison studies, including 1) 2D/3D-based image analysis, 2) different radiomics feature space dimension reduction methods, and 3) different data source combinations in SVM input design. Results: The model achieved 0.638 prediction accuracy in the test set when using patient-specific clinical features only, which was higher than the results using radiomics features/deep features as sole input of SVM

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：