检索结果-内蒙古大学图书馆

Zero Watermarking algorithms for Medical images Based on Multi-module Fusion with VGG16 and Dual-Logistic Mapping Encryption

引用

CIRCUITS systems AND SIGNAL processing 2025年 1-36页

作者： Li, Jingyou Wei, Rongle xi, xiaotian Zhang, Guangda Yang, Zixin Zhang, Fengshan Qiqihar Univ Sch Comp & Control Engn Qiqihar Heilongjiang Peoples R China Qiqihar Univ Heilongjiang Key Lab Big Data Network Secur Detect Qiqihar Heilongjiang Peoples R China

The current research proposes a new zero watermarking approach that aims to protect medical images from tampering by utilizing the powerful learning capabilities of deep learning. In this paper, we propose a robust zero watermarking scheme based on multi-module fusion of VGG16, which enhances the efficiency and security of medical image transmission. The scheme leverages depthwise separable convolution to speed up feature extraction, enabling the focus on global image features. Furthermore, by incorporating the Convolutional Attention Module with channel and spatial attention, the extracted features are enhanced with stronger perceptual, discriminative, and generalization *** preserve multi-scale information and improve feature reusability, jump-joins are employed. The multi-module fusion of VGG16 facilitates the extraction of deep features with superior generalization, multi-scale, and high perceptual capabilities. Binary feature vectors are generated using a mean hash algorithm and encrypted using a novel dual chaos dual logistics encryption method designed in this paper, which significantly enhances both image watermark security and overall network transmission *** effectiveness of the scheme is verified through extensive experiments. Experimental results demonstrate that the proposed method achieves an NC value of over 0.92 in the presence of noise filtering and rotation attacks, indicating robust security and strong resistance to common image distortions. These results confirm the scheme's capability to provide high security for medical image transmission, ensuring both robustness against attacks and confidentiality in medical systems.

关键词： Zero watermark Multi-module fusion VGG16 Dual-chaos dual-logistics encryption (DcDlE)

来源：评论

学校读者我要写书评

暂无评论

Fλ/d as a predictor of image classification algorithm performance

引用

APPLIED OPTICS 2025年第4期64卷 845-854页

作者： Hixson, Jonathan g. Teaney, Brian Finch, Michael f. Nehmetallah, George Driggers, Ronald US Army Combat Capabil Dev Command DEVCOM C5ISR Ctr RTI Aberdeen Proving Ground NY 21005 USA Catholic Univ Amer EECS Dept 620 Michigan Ave NE Washington DC 20064 USA Univ Arizona Wyant Coll Opt Sci Infrared Syst Grp Tucson AZ 85721 USA

This research studied the effect of variations in a sensor's F lambda/d metric value (FLD) on the performance of machine learning algorithms such as the YOLO (You Only Look Once) algorithm for object classification. The YOLO_v3 and YOLO_v10 algorithms were trained using static imagery provided in the commonly available training dataset provided by Teledyne FLIR systems. image processing techniques were used to degrade image quality of the test dataset also provided by Teledyne FLIR systems, simulating detector-limited to optics-limited performance, which results in a variation of the FLD metric between 0.339 and 7.98. The degraded test set was used to evaluate the performance of YOLO_v3 and YOLO_v10 for object classification and relate the FLD metric to the probability of detection. Results of YOLO_v3 and YOLO_v10 are presented for the varying levels of image degradation. A summary of the results is discussed along with recommendations for evaluating an algorithm's performance using a sensor's FLD metric value. (c) 2025 Optica Publishing Group. All rights, including for text and data mining (TDM), Artificial Intelligence (AI) training, and similar technologies, are reserved.

关键词： image quality image sensors Imaging systems Imaging techniques Machine learning Sensor performance

来源：评论

学校读者我要写书评

暂无评论

A Novel Visual Communication Scenario Based on image processing algorithms 2

A Novel Visual Communication Scenario Based on Image Process...

引用

2nd International conference on Machine Learning and Autonomous systems, ICMLAS 2025

作者： Yang, Jinhua Hou, Li Xijing University Shaanxi Xi'an710123 China

ISBN: (纸本)9798331505745

In order to comply with the trend of intelligent visual communication, this study proposed an innovative visual communication scenario based on image processing algorithms. The framework aims to optimize traditional key technologies such as the image generation, editing, style transfer and image compression. First, as the foundation of visual communication, this study proposes a generative adversarial network model based on text semantic information for image generation and editing. The model achieves stable image generation and efficient editing from a theoretical level through paired training of text and image pairs. Secondly, for image style transfer, this study designed an improved VGG19 convolutional neural network. At the same time, the adaptive instance normalization technology was combined to optimize the effect of style transfer. Finally, in terms of image compression, the study proposed an improved generative adversarial network (REVISED-GAN) model. This model can dynamically adjust the compression error based on structured information to improve image compression efficiency. Through comparative tests, the proposed image style transfer and image compression algorithms have shown excellent performance in terms of structural similarity, image quality and compression ratio. © 2025 IEEE.

关键词： Visual communication

来源：评论

学校读者我要写书评

暂无评论

Quantum Variational Singular Value Decomposition: A new hybrid approach towards quantum image compression 17

Quantum Variational Singular Value Decomposition: A new hybr...

引用

17th International conference on COMmunication systems and NETworkS, COMSNETS 2025

作者： Britant Tiwari, Abhishek CDAC Uttar Pradesh Noida India

ISBN: (纸本)9798331531195

This research paper explores the application of singular value decomposition (SVD) in quantum image processing (QIP), specifically focusing on the computation of eigenvalues using variational quantum algorithms. SVD is a powerful mathematical tool in image processing, used for tasks such as image compression, noise reduction, and feature extraction. In this study, we propose quantum variational singular value decomposition (QVSVD) based on the variational quantum deflation (VQD) algorithm to determine the eigenvalues that contributes in calculating the singular values of the image matrix and eventually doing the SVD. This facilitates the extraction of eigenvalues with exponential speedup on real hardware compared to classical methods. We detail the implementation of this quantum algorithm within the framework of QIP, highlighting its advantages in terms of computational efficiency and accuracy on a Fault-Tolerant Quantum Computing (FTQC). Furthermore, we present a comparative analysis of the quantum and the classical method of SVD, demonstrating the accuracy of the image data. The simulation results validate the theoretical advantages of using quantum algorithms for SVD in image processing. This work not only underscores the potential of quantum computing in enhancing image processing techniques but also sets the stage for future research in the field, exploring more complex image processing tasks and other quantum algorithms. Our findings suggest that quantum image processing can offer unprecedented capabilities, paving the way for advancements in various applications such as medical imaging, remote sensing, and multimedia processing. © 2025 IEEE.

关键词： Noise abatement

来源：评论

学校读者我要写书评

暂无评论

Application of Various Novel Computer image processing (CIP) algorithms in Digital Media Art images

Application of Various Novel Computer Image Processing (CIP)...

引用

2025 International conference on Multi-Agent systems for Collaborative Intelligence, ICMSCI 2025

作者： Du, xian Shanghai Institute of Visual Arts Shanghai201620 China

ISBN: (纸本)9798331509828

With the continuous development of digital image processing algorithms, its application scenarios have been integrated from the simple research of a single image and a single algorithm to a multi-algorithm fusion analysis paradigm. Therefore, this study proposes a digital media art image analysis framework that combines multiple computer image processing (CIP) algorithms. This breakthrough covers three core architectures: image edge extraction, image style transfer, and image compression. In the image edge extraction module, this study designed an improved edge detection algorithm based on de-noising auto-encoder. This algorithm improves the accuracy of edge detection through multi-directional feature extraction and (I, O)-fuzzy rough set optimization, while maintaining global stability. In the image style transfer module, this study proposed a dual-module network. The network includes a texture translation network and a perceptual loss network, and achieves cross-domain style transfer through a fusion model. At the same time, the algorithm optimizes the perceptual loss function to enhance semantic matching capabilities. In the image compression module, this study uses chaotic systems to construct an optimized measurement matrix. Specifically, through a distributed data parallel training framework, a compression method based on the LPAC gradient sparse algorithm is constructed. By testing the three modules separately, the experimental results confirm the effectiveness of the proposed algorithm. © 2025 IEEE.

关键词： image compression

来源：评论

学校读者我要写书评

暂无评论

Ultrafast Ultrasound Imaging System With Beamforming on Single FPGA

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Zhang, xi Qi, Sheng Li, Yifan Xu, Lijun Ma, Jianguo Beihang Univ Sch Instrumentat & Optoelect Engn Beijing 100191 Peoples R China Shandong Jianzhu Univ Sch Sci Jinan 250100 Peoples R China

Ultrafast ultrasound imaging technologies, such as ultrasound localization microscopy (ULM) and functional ultrasound (FUS), are gaining significant attention in the ultrasound field, driving the demand for higher frame rates in open ultrasound systems. However, current open ultrasound systems face the challenge of balancing high imaging frame rates with the complexity of hardware architecture. This work reports the detailed implementation and hardware architecture of a 128-channel ultrafast ultrasound platform based on a single field-programmable gate array (FPGA). This platform connects the ultrasound card to a personal computer (PC) via peripheral component interconnect express (PCIe) 3.0, providing excellent compatibility. The platform uses custom instructions to implement flexible parameter configuration in the FPGA. It can control the transmission and reception of signals, the output of raw or beamformed RF data, and the sampling depth. Additionally, a 64 sub-aperture delay-and-sum (DAS) beamformer within the FPGA is implemented in the platform. This beamformer gives output as a 128x2048 image at 10000 frames/s (fps) using plane wave. The core components of the platform measure only 12x21 cm, achieving a compact design while delivering high-frame-rate (HFR) imaging.

关键词： Ultrasonic imaging Field programmable gate arrays Delays Array signal processing Imaging Hardware Computer architecture Software Radio frequency Software algorithms Beamforming field-programmable gate array (FPGA) ultrasound system ultrafast ultrasound

来源：评论

学校读者我要写书评

暂无评论

Evaluating multi-modal image matching algorithms for remote sensing applications 6

Evaluating multi-modal image matching algorithms for remote ...

引用

6th International conference on Geoscience and Remote Sensing Mapping, GRSM 2024

作者： Liu, Zhanquan Han, Yilong Fu, Mingsheng Song, Baiwan Gong, Danchao Ming, Yanfang Sun, Lin College of Geodesy and Geomatics Shandong University of Science and Technology Qingdao266590 China Shandong Province Research Center of Engineering and Technology for GPS RS and GIS Qingdao266590 China Star-Rising Technologies Co. Ltd. Qingdao266590 China Chinese Academy of Surveying and Mapping Beijing100036 China Shandong GEO-Surveying & Mapping Institute Jinan250000 China State Key Laboratory of Geo-Information Engineering Xi’an710054 China Xi’an Institute of Surveying and Mapping Xi’an710054 China

ISBN: (纸本)9781510688018

image matching is a fundamental task in image processing, particularly in photogrammetry and remote sensing. Conventional image matching methods often struggle with multi-modal images due to significant geometric distortions and nonlinear radiometric effects. To address these challenges, researchers have developed a variety of algorithms specifically tailored for multi-modal data. Despite this, there is a notable lack of comprehensive theoretical summaries of multi-modal image matching. This paper aims to bridge this gap by discussing the complexities of matching multi-modal images and providing a succinct overview of existing methods. We evaluate three Conventional algorithms—OS-SIFT, RIFT, and POS-GIFT—and two deep learning methods—SemLA and SuperPoint+SuperGlue. These methods are assessed using six types of multi-modal remote sensing datasets: Optical-Optical, Optical-Infrared, Optical-Depth, Optical-Map, Optical-Synthetic Aperture Radar (SAR), and Day-Night. The results indicate that the POS-GIFT algorithm outperforms others, achieving an average of 1217 correct matches with a success rate of 96.70%. This paper concludes by discussing the challenges of multi-modal image matching and suggesting future research directions, serving as a valuable resource for researchers and practitioners in remote sensing. © 2025 SPIE.

关键词： image matching

来源：评论

学校读者我要写书评

暂无评论

Subspace-based DOA Estimation Method for Coherent Signals Assisted by Transformer 16

Subspace-based DOA Estimation Method for Coherent Signals As...

引用

16th International conference on Signal processing systems, ICSPS 2024

作者： Mao, Wei xi, Feng School of Electronic and Optical Engineering Nanjing University of Science and Technology China

ISBN: (纸本)9781510689251

Direction of Arrival (DOA) estimation plays a crucial role in array signal processing, with broad applications across various fields. Traditional Model-Based (MB) DOA estimation algorithms, however, are constrained by their reliance on prior knowledge and tend to perform poorly in non-ideal conditions, such as when dealing with coherent sources or limited snapshot numbers. In contrast, Data-Driven (DD) approaches using deep learning offer novel insights and capabilities for enhancing DOA estimation. This paper introduces a novel hybrid MB-DD framework for estimating the DOAs of multiple coherent narrowband signals captured by Uniform Linear Arrays (ULA). Specifically, the proposed method, TransSubNet, employs an autoencoder module centered around a Transformer Encoder to generate a proxy covariance matrix, which is then fed into classical subspace algorithms, such as root-MUSIC, to estimate the DOAs of the received signals. By integrating the strengths of both MB and DD methods, TransSubNet utilizes the Transformer-based network to improve the covariance matrix estimation process, enhancing the robustness and accuracy of subspace-based DOA algorithms. Our architecture achieves high-precision DOA estimation even under challenging conditions, such as low snapshot numbers, low Signal-to-Noise Ratios (SNRs), and 1-bit quantized signals. Experimental results show that TransSubNet outperforms existing algorithms in the domain, delivering superior performance across a range of scenarios. © 2025 SPIE.

关键词： Covariance matrix

来源：评论

学校读者我要写书评

暂无评论

All-In-One image Restoration in Adverse Weather Conditions with Spatial-Frequency Domain Information 17th

All-In-One Image Restoration in Adverse Weather Conditions w...

引用

17th International conference on Intelligent Robotics and Applications

作者： Jing, Panfeng Liang, Yudong Wang, Siyu Gao, Pin An, Hongqiang Shanxi Univ Sch Comp & Informat Technol Taiyuan Shanxi Peoples R China Shanxi Univ Key Lab Computat Intelligence & Chinese Informat Minist Educ Taiyuan Shanxi Peoples R China Qingdao Htinvent Intelligent Technol Co Ltd Qingdao Shandong Peoples R China

ISBN: (纸本)9789819607730;9789819607747

Recently, the collaborative operations of multi-domain unmanned vehicle systems have garnered significant attention, particularly regarding their visual perception capabilities under various complex weather conditions. Adverse weather such as rain, snow, and fog can severely degrade image quality, thereby affecting the efficiency and safety of unmanned vehicle systems to a certain extent. Therefore, image enhancement algorithms capable of handling multiple complex weather conditions are crucial for improving the visual perception capabilities of unmanned vehicle systems. This paper presents a unified image enhancement algorithm designed to address various complex weather conditions, such as rain, snow, and fog. The algorithm restores images by integrating spatial and frequency domain information. Experimental results fully demonstrate that this method can significantly enhance images under different adverse weather conditions and greatly improve image quality.

关键词： Deep Learning image Enhancement Adverse Weather Conditions

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Survey of Methods for Identifying Counterfeit Banknotes Using image processing and Machine Learning 6th

A Comprehensive Survey of Methods for Identifying Counterfei...

引用

6th International conference on Intelligent Communication, Control and Devices, ICICCD 2024

作者： Sharmila, B. Sowmiya, A. Mary, Teena Sandeep, J. Sreeja, C.S. CHRIST University Dharmaram College Post Hosur Road Karnataka Bengaluru560029 India

ISBN: (纸本)9789819783281

The world economy is threatened by counterfeit currencies. Counterfeit currencies are often difficult, time-consuming and ineffective to identify manually. Automated methods based on image processing techniques and machine learning algorithms are helpful in detecting counterfeit notes. This survey paper reviews the current strategies on fake banknote detection using image processing techniques and machine learning algorithms. We discuss various stages of the detection process, including image acquisition, preprocessing, feature extraction and classification. Furthermore, we analyze the limitations and comparative performance of different algorithms and approaches mentioned in the literature. The survey aims to provide insights into the various methodologies, challenges and future directions in the field of fake banknote detection, facilitating the development of more robust and effective counterfeit detection systems. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Adversarial machine learning

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：