检索结果-内蒙古大学图书馆

A comparison of methods for generating synthetic training data for domain adaption of deep learning models in ultrasonic non-destructive evaluation

引用

NDT & E INTERNATIONAL 2024年 141卷

作者： Mcknight, Shaun Pierce, S. Gareth Mohseni, Ehsan Mackinnon, Christopher Macleod, Charles 'Hare, Tom Loukas, Charalampos Univ Strathclyde Elect & Elect Engn Glasgow Scotland Spirit Aerosyst Belfast North Ireland

This work provides a solution to the challenge of small amounts of training data in Non-Destructive Ultrasonic Testing for composite components. It was demonstrated that direct simulation alone is ineffective at producing training data that was representative of the experimental domain due to poor noise reconstruction. Therefore, four unique synthetic data generation methods were proposed which use semi-analytical simulated data as a foundation. Each method was evaluated for its performance in the classification of real experimental images when trained on a Convolutional neural Network which underwent hyperparameter optimization using a genetic algorithm. The first method introduced task specific modifications to CycleGAN, a generative network for image -to-image translation, to learn the mapping from physics-based simulations of defect indications to experimental indications in resulting ultrasound images. The second method was based on combining real experimental defect free images with simulated defect responses. The final two methods fully simulated the noise responses at an image and signal level respectively. The purely simulated data produced a mean classification F1 score of 0.394. However, when trained on the new synthetic datasets, a significant improvement in classification performance on experimental data was realized, with mean classification F1 scores of 0.843, 0.688, 0.629, and 0.738 for the respective approaches.

关键词： Deep Learning Ultrasonic Testing Synthetic Data Generation Generative Adversarial Networks image processing and Computer Vision Genetic Algorithms Non-Destructive Testing

来源：评论

学校读者我要写书评

暂无评论

Imaging With Equivariant Deep Learning: From unrolled network design to fully unsupervised learning

引用

IEEE signal processing MAGAZINE 2023年第1期40卷 134-147页

作者： Chen, Dongdong Davies, Mike Ehrhardt, Matthias J. Schonlieb, Carola-Bibiane Sherry, Ferdia Tachella, Julian Univ Edinburgh Computat Sensing & Machine Learning Edinburgh EH8 9YL Midlothian Scotland Univ Edinburgh Signal & Image Proc Edinburgh EH9 3JL Midlothian Scotland European Assoc Signal Proc Sesimbra Portugal Royal Acad Engn London England Univ Bath Bath BA2 7AY Avon England Univ Cambridge Cambridge CB3 0WA England Univ Cambridge Dept Appl Math & Theoret Phys Cambridge CB3 0WA England Ecole Normale Super Lyon CNRS F-69364 Lyon France

From early image processing to modern computational imaging, successful models and algorithms have relied on a fundamental property of natural signals: symmetry. Here symmetry refers to the invariance property of signal sets to transformations, such as translation, rotation, or scaling. Symmetry can also be incorporated into deep neural networks (DNNs) in the form of equivariance, allowing for more data-efficient learning. While there have been important advances in the design of end-to-end equivariant networks for image classification in recent years, computational imaging introduces unique challenges for equivariant network solutions since we typically only observe the image through some noisy ill-conditioned forward operator that itself may not be equivariant. We review the emerging field of equivariant imaging (EI) and show how it can provide improved generalization and new imaging opportunities. Along the way, we show the interplay between the acquisition physics and group actions and links to iterative reconstruction, blind compressed sensing, and self-supervised learning.

关键词： Deep learning neural networks Imaging signal processing algorithms Self-supervised learning Noise measurement Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Multi-user watermarking based on paralleled invertible neural networks

引用

signal image AND VIDEO processing 2025年第6期19卷 1-12页

作者： Shi, Jinghui Yang, Zhiguang Li, Li Ye, Jingyu Dong, Zhicheng Wu, Hanzhou Shanghai Univ Sch Commun & Informat Engn Shanghai 200444 Peoples R China Digitial Guangxi Grp Guilin 541001 Guangxi Peoples R China Guangxi Normal Univ Guilin 541001 Guangxi Peoples R China Tibet Univ Lhasa 850000 Xizang Peoples R China

With the rapid development of artificial intelligence technology, deep-faked audios have become increasingly prevalent in cyberspace, creating an urgent need for effective detection and tracing mechanisms. Watermarking offers an active defense strategy for identifying deep-faked audios, and the integration of multi-user watermarking further enhances its protection and tracking capabilities. By leveraging multi-bit watermarking technology, multiple independent watermarks can be embedded within a single audio file, providing unique identifiers for different stakeholders (e.g., lyricists, composers, platforms, etc.). This enables granular content protection and precise source tracing across diverse scenarios. For instance, users can upload audios embedded with both personal watermarks (e.g., lyricist and composer) and platform watermarks, ensuring better monitoring of audio dissemination and usage. Existing methods for embedding multi-user watermarks often require training multiple networks, which can lead to interference during the training process. To address this challenge, we propose an audio watermarking technique based on Paralleled Invertible neural Networks (PINN). Our approach enables the splitting of a single network into multiple sub-networks for multi-user watermark embedding while requiring only one network to be trained. Specifically, we design a parallel-structured invertible neural network that is highly flexible, allowing the network to be divided into multiple sub-networks for watermark rewriting or multi-user watermark embedding. These embedded watermarks remain independent of each other, ensuring robust and interference-free operations. Experiments demonstrate the framework's robustness, achieving an average BER of 1.8% and SNR of 26.5 dB when embedding four 32-bit watermarks, and also exhibits strong resistance to common distortions, such as noise addition, MP3 compression, and re-sampling. Furthermore, partial rewriting experiments highlight the f

关键词： Invertible neural networks Audio Deep learning Digital watermarking

来源：评论

学校读者我要写书评

暂无评论

Lightweight Plank Grading Based on a Multichannel Spatial Attention Mechanism

引用

APSIPA TRANSACTIONS ON signal AND INFORMATION processing 2024年第1期13卷

作者： Li, Ziqiang Wang, Huixuan Lv, Na Guo, Yihong Zhang, Yan Univ Jinan Shandong Prov Key Lab Network Based Intelligent Co Jinan 250022 Peoples R China Univ Jinan Sch Informat Sci & Engn Jinan 250022 Peoples R China China Mobile Grp Shandong Co Ltd Jinan 250013 Peoples R China Mstar Machinery Co Ltd Fei County 273400 Peoples R China Shandong Jianzhu Univ Sch Sci Jinan 250101 Peoples R China

The grading and sorting process of the wood planks is a critical stage within the production line. However, in the real world, many factories still rely on humans to perform this task manually. This method is not only inefficient, but also time-consuming and laborintensive. To solve this problem, a lightweight wood board image classification algorithm based on a multichannel spatial attention mechanism is proposed in this paper, which can be used for the real-time classification of wood planks on the production line. This method is used to classify freshly rotated cut wood planks based on defects such as damage and voids on the production line. Specifically, the received images of the wood planks were processed by a feature extraction module to effectively separate the interfering background from the foreground of the wood planks. After fusing the edge information map with the foreground image of the wood planks, a multichannel convolutional neural network with spatial and channel attention ability was used to learn the features for correctly grading the wood planks. Experimental results show that the proposed method is superior to traditional methods and some existing deep learning algorithms in terms of performance and benefits.

关键词： Wood plank image grading real-time classification feature fusion lightweight neural network convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Cadmamba: a differential feature fusion-based neural network for coronary artery disease screening from facial videos

引用

signal image AND VIDEO processing 2025年第6期19卷 1-9页

作者： Ou, Haohan Jia, Baoping Hu, Yishan Cai, Meiling Qiang, Yan Chen, Qi Pei, Bo Zhao, Juanjuan Taiyuan Univ Technol Coll Comp Sci & Technol Coll Data Sci Taiyuan Shanxi Peoples R China Taiyuan Univ Technol Sch Software Taiyuan Shanxi Peoples R China Jinzhong Coll Informat Coll Informat Jinzhong Shanxi Peoples R China Shanxi Cardiovasc Hosp Dept Cardiol Taiyuan Shanxi Peoples R China North Univ China Sch Software Taiyuan Shanxi Peoples R China Macau Univ Sci & Technol Macau Peoples R China Univ S Florida Tampa FL USA

Coronary artery disease (CAD) is a leading cause of mortality worldwide, with its insidious nature and high fatality rate emphasizing the critical importance of early screening. However, current facial video-based CAD screening methods struggle with feature extraction, making it hard to capture subtle long-term facial variations. To address this issue, this paper proposes CADMamba, a deep neural network model based on the Mamba module. CADMamba designed to assess CAD risk non-invasively by analyzing easily accessible facial videos from daily life. To enhance feature extraction and capture long-term temporal dependencies, CADMamba incorporates a differential frame feature fusion method based on self-attention (DFF-AFF) and the CADMamba Encoder. The model was evaluated using 2,250 video samples (374 subjects, of whom 54% had CAD), demonstrating a strong correlation between CADMamba's predictions and actual outcomes. The recall and precision rates were 84.70% and 80.37%, respectively, representing a notable improvement compared to baseline video models. This model shows promise in supporting early CAD screening efforts, especially in areas with scarce medical resources.

关键词： Coronary heart disease Deep learning Facial video

来源：评论

学校读者我要写书评

暂无评论

Developing an explainable Deep neural Network for stress detection using biosignals and human-engineered features

引用

BIOMEDICAL signal processing AND CONTROL 2025年 109卷

作者： Lee, Hyeokjong Kim, Jaewon Han, Bohyung Park, Sang Min Chang, Jooyoung Seoul Natl Univ Coll Med Dept Biomed Sci Seoul 03080 South Korea Seoul Natl Univ Dept Elect & Comp Engn Seoul 08826 South Korea Seoul Natl Univ Interdisciplinary Program Artificial Intelligence Seoul 08826 South Korea XAIMED Co Ltd Seoul 02738 South Korea

Detecting stress is crucial for preventing long-term health issues, and recent research, particularly using WESAD (WEarable Stress and Affect Detection) dataset, has made significant advances in this area. neural Networks (DNNs) demonstrate a key advantage by efficiently processing raw signals, unlike traditional machine learning methods that predominantly rely on human-engineered features. In our study, we enhance DNN capabilities by integrating both raw signals and human-engineered altogether. Our model, augmented with attention mechanisms, significantly improves its performance. approach yields state-of-the-art results, demonstrating high accuracies across various modalities in the classification task: 92.89% in Electrodermal Activity (EDA), 92.16% in Photoplethysmography (PPG), 96.57% in all modalities. Explaining model decisions in biosignals is challenging compared to the image domain, where are inherently more intuitive. Many Explainable AI (XAI) techniques, optimized for images, face difficulties when applied to biosignals. To address this, we employ human-engineered features for biosignals, grounded the 'Human Comprehensible Space' concept, making explanations intuitively understandable. Furthermore, survey conducted to statistically validate the comprehensibility of our XAI methods confirms their effectiveness in enhancing the accessibility of explanations.

关键词： Artificial intelligence Deep learning Explainable AI Biosignal analysis Stress detection

来源：评论

学校读者我要写书评

暂无评论

UPNet: Uncertainty-Based Picking Deep Learning Network for Robust First Break Picking

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2024年 62卷

作者： Wang, Hongtao Zhang, Jiangshe Wei, Xiaoli Long, Li Zhang, Chunxia Guo, Zhenbo Xi An Jiao Tong Univ Sch Math & Stat Xian 710049 Shaanxi Peoples R China China Natl Petr Corp Bur Geophys Prospecting Inc Geophys Technol Res Ctr Zhuozhou Beijing 072750 Hebei Peoples R China

In seismic exploration, first break (FB) picking is a crucial aspect in determining subsurface velocity models, significantly influencing the placement of wells. Many deep neural networks (DNNs)-based automatic picking methods have been proposed to accelerate this process. Significantly, the segmentation-based DNN methods provide a segmentation map and then estimate FB from the map using a thresholding technique. However, these automatic methods applied in field datasets cannot ensure robustness, especially in the case of a low signal-to-noise ratio (SNR). In this article, we introduce uncertainty quantification (UQ) into FB picking and propose a novel uncertainty-based picking deep learning network (UPNet). UPNet specifically consists of two DNNs. A Bayesian network infers a posterior distribution by sampling the segmentation map of FB. Subsequently, a regression network integrates the segmentation map, the original trace, and the low-frequency (LF) trace to infer the FB trace by trace. Finally, a decision-making method provides the final FB based on uncertainty analysis, offering robust FB. Importantly, UPNet avoids post-processing to obtain FB using the threshold method, as in the segmentation-based picking methods, and instead provides the FB of each trace end-to-end. Moreover, UPNet not only estimates the uncertainty of the network output but can also filter out predictions with low confidence. Many experiments have shown that UPNet demonstrates higher accuracy and robustness than the deterministic DNN-based model, achieving state-of-the-art (SOTA) performance in field surveys. In addition, we verify that the calculated uncertainty is significant, which can serve as a reference for human decision-making.

关键词： Uncertainty image segmentation Task analysis Accuracy Deep learning Convolutional neural networks Clustering methods Bayesian deep learning (DL) first break (FB) picking uncertainty quantification (UQ)

来源：评论

学校读者我要写书评

暂无评论

Adaptive resizer-based transfer learning framework for the diagnosis of breast cancer using histopathology images

引用

signal image AND VIDEO processing 2023年第8期17卷 4561-4570页

作者： Duzyel, Okan Catal, Mehmet Sergen Kayan, Ceyhun Efe Sevinc, Arda Gumus, Abdurrahman Izmir Inst Technol Dept Elect & Elect Engn TR-35430 Izmir Turkiye

Breast cancer is a major global health concern, and early and accurate diagnosis is crucial for effective treatment. Recent advancements in computer-assisted prediction models have facilitated diagnosis and prognosis using high-resolution histopathology images, which provide detailed information on cancerous tissue. However, these high-resolution images often require resizing, leading to potential data loss. In this study, we demonstrate the effect of a learnable adaptive resizer for breast cancer classification using the BreakHis dataset. Our approach incorporates the adaptive resizer with various convolutional neural network models, including VGG16, VGG19, MobileNetV2, InceptionResnetV2, DenseNet121, DenseNet201, and EfficientNetB0. Despite producing visually less appealing images, the learnable resizer effectively improves classification performance. DenseNet201, when jointly trained with the adaptive resizer, achieves the highest accuracy of 98.96% for input images of 448x448 resolution. Our experimental results demonstrate that the adaptive resizer performs better at a magnification factor of 40x compared to higher magnifications. While its effectiveness becomes less pronounced as image resolution increases to 100x, 200x, and 400x, the adaptive resizer still outperforms bilinear interpolation. In conclusion, this study highlights the potential of adaptive resizers in enhancing performance for medical image classification. By outperforming traditional image resizing methods, our work contributes to the advancement of deep neural networks in the field of breast cancer diagnostics.

关键词： Breast cancer Histopathology images Computer-assisted prediction Deep neural networks Adaptive resizer

来源：评论

学校读者我要写书评

暂无评论

Spatial Adaptive Filter Network With Scale-Sharing Convolution for image Demoireing

引用

IEEE signal processing LETTERS 2024年 31卷 2495-2499页

作者： Xu, Yong Wei, Zhiyu Xu, Ruotao Zhou, Zihan Yu, Zhuliang South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China Guangdong Prov Key Lab Multimodal Big Data Intelli Guangzhou 519085 Peoples R China PaZhou Lab Guangzhou Guangzhou 510335 Peoples R China Inst Super Robot Huangpu Guangzhou 510555 Peoples R China South China Agr Univ Sch Coll Math & Informat Guangzhou Peoples R China South China Univ Technol Shien Ming Wu Sch Intelligent Engn Guangzhou 511442 Peoples R China South China Univ Technol Sch Automat Sci & Engn Guangzhou 510641 Peoples R China

Removing moire patterns is a challenging task as it is a spatially varying degradation that varies in shape, color and scale. Existing image restoration models often rely on static convolutional neural networks (CNNs)-based architectures, and hence potentially suboptimal for addressing the diverse manifestations of moire patterns across different images and spatial positions. To this end, we propose a spatially adaptive neural network for image demoireing. This network introduces a dual-branch filter prediction module engineered to predict pixel-wise adaptive filters that can process moire patterns of varying orientations and color-shift issues. To further tackle the challenge presented by scale variability, a scale-sharing convolution module is proposed, utilizing pixel-wise adaptive filters with multiple dilations to handle moire patterns of different sizes but similar shapes effectively. Upon extensive evaluations of three benchmark datasets, our model consistently outperforms existing methods, yielding a PSNR improvement of over 0.37dB across all evaluated datasets and providing additional benefits in terms of model size.

关键词： Convolution Feature extraction Adaptive filters image color analysis Shape Tensors Semantics Moir & eacute pattern removal dynamic filtering spatial-varying processing

来源：评论

学校读者我要写书评

暂无评论

Advocating Pixel-Level Authentication of Camera-Captured images

引用

IEEE ACCESS 2024年 12卷 45839-45846页

作者： Punnappurath, Abhijith Zhao, Luxi Abdelhamed, Abdelrahman Brown, Michael S. Samsung AI Ctr Toronto Toronto ON M5G 1L7 Canada Google Res Toronto ON M5H 2G4 Canada

The authenticity of digital images posted online and shared on social media is often questioned due to the ability of photo-editing software to alter image content and generative AI methods that can produce visually compelling deepfakes. Only images directly produced by cameras are deemed unaltered and beyond suspicion, as they have not undergone any modifications. However, there is a recent trend among camera manufacturers to integrate AI-based modules into the dedicated onboard hardware, specifically the image signal processor (ISP), responsible for processing the captured sensor image into the final saved image for users. Many of these AI modules utilize perceptual or generative losses during training, which can "hallucinate" image content. While this hallucinated content often manifests as small details and textures, there are instances where these regions unintentionally impact the interpretation of the entire image. This paper aims to bring attention to this issue and advocate for in-camera strategies to validate the authenticity of camera-captured images at a pixel level. We propose the creation of an "authenticity" mask that could be stored as additional metadata with each image. This information can be extracted and overlaid on the image to easily identify the hallucinated regions. Considering the widespread implications of image authenticity (e.g., in courtroom evidence, news broadcasts, and other media forms), we anticipate that authentication metadata will become a standard practice for any ISP utilizing AI.

关键词： Authenticity AI camera metadata neural ISP generative models adversarial loss perceptual loss digital forensics

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：