检索结果-内蒙古大学图书馆

2024 international symposium on Integrated Circuit Design and Integrated Systems, ICDIS 2024

作者： Wan, Tengfei Liu, Huiyi Geng, Lijie College of Computer and Information Hohai University Nanjing China

ISBN: (纸本)9798400718229

This study proposes an innovative approach using multimodal large models for intelligent annotation and automatic caption generation of electronic circuit images. By integrating the Contrastive Language-image Pre-training (CLIP) model with large language models, the proposed method effectively recognizes and annotates key components and signal pathways in circuit diagrams. Through a deep fusion of visual and textual features, this approach not only accurately extracts detailed information from circuit images but also generates comprehensive technical descriptions, significantly enhancing automation in circuit design, debugging, and fault diagnosis. The model was evaluated on the FICS PCB-A dataset, where it achieved an annotation accuracy of 83.3% and a BLEU-4 score of 0.091, outperforming existing models. This research highlights the effectiveness of multimodal information fusion in processing electronic circuit images, providing robust support for automating circuit design and maintenance. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Integrated circuit design

来源：评论

学校读者我要写书评

暂无评论

Advancing SAR image Robustness: Integrating Diffusion Models for Adversarial Purification and Speckle Noise Suppression

Advancing SAR Image Robustness: Integrating Diffusion Models...

引用

2025 IEEE international Conference on Acoustics, Speech, and signal processing, ICASSP 2025

作者： Xie, Ganglin Lu, Haobo Zhang, Xinze He, Kun School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9798350368741

Synthetic Aperture Radar (SAR) is a powerful tool for ground target detection, but SAR images often suffer from coherent speckle noise, which complicates automatic target recognition (ATR). Recent advances in deep learning have shown promise in SAR-ATR, yet there is a notable gap in addressing adversarial attacks and despeckling challenges. This study presents SAR Adversarial Diffusion Purification (SAR-ADP), a novel model designed to enhance SAR image analysis by tackling both adversarial attacks and coherent speckle noise. Our main contributions include pioneering methods for SAR image adversarial defense, simultaneous adversarial noise removal and despeckling, guidance mechanisms within the diffusion model, and a streamlined architecture for SAR image purification. SAR-ADP integrates adversarial defense strategies, diffusion modeling, and despeckling techniques, resulting in significant improvements in image recognition accuracy and post-processing. Empirical results validate its effectiveness, achieving a robustness accuracy of approximately 97% and a standard accuracy of around 99%. © 2025 IEEE.

关键词： adversarial defense despeckling Diffusion model Synthetic Aperture Radar image target recognition

来源：评论

学校读者我要写书评

暂无评论

ICBSP 2024 - Proceedings of the 2024 9th international Conference on Biomedical Imaging, signal processing

ICBSP 2024 - Proceedings of the 2024 9th International Confe...

引用

2024 9th international Conference on Biomedical Imaging, signal processing, ICBSP 2024

ISBN: (纸本)9798400717499

The proceedings contain 24 papers. The topics discussed include: an integrated deep learning model to analyze CT scans for minimally invasive accurate classification of T1a small renal masses;tissue type classification for whole slide histological images with graph convolutional neural network;classifying skin diseases using convolutional neural networks;comparative analysis of YOLO architectures for automated detection of liver disease in histopathological images;bibliometric analysis and research trends in artificial intelligence for medical imaging in Alzheimer’s disease;personalized federated learning using client clustering for medical image classification;data augmentation of domain learning in optic nerve combined cup-disc segmentation with a few labeled data;and a coronary plaque detection and identification method based on medical prior knowledge and hybrid attention unit.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PAFedMIS: Personalized Asynchronous Federated Learning for Medical image Segmentation

PAFedMIS: Personalized Asynchronous Federated Learning for M...

引用

2025 IEEE international Conference on Acoustics, Speech, and signal processing, ICASSP 2025

作者： Li, Yi Hua, Yue Zheng, Xin Guo, Yanqing Wang, Bo Dalian University of Technology China Griffith University Australia

ISBN: (纸本)9798350368741

As privacy protection gains momentum, federated learning has emerged as a cutting-edge approach in medical image analysis. However, the intricacies of medical image segmentation task have led to a dearth of research in this domain, with existing studies falling short in tackling two pivotal challenges: The traditional model with the uniform global model under-performs for certain clients due to the heterogeneity and nonIndependent Identically Distributed(non-IID) data across medical institutions. And the communication between the server and clients often incurs significant time costs. This paper introduces a novel Personalized Asynchronous Federated learning for Medical image Segmentation model, dubbed PAFedMIS, to mitigate the negative impact of the heterogeneous data and fully utilized the waiting time, in medical image segmentation. Comprehensive experiments on ISIC2018 demonstrate the enhanced model accuracy and training efficiency of PAFedMIS. © 2025 IEEE.

关键词： asynchronous federated learning image segmentation medical image personalized federated learning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Driven analysis of Tumor Regions in Magnetic Resonance Imaging 24

Deep Learning-Driven Analysis of Tumor Regions in Magnetic R...

引用

5th international symposium on Artificial Intelligence for Medicine Science, ISAIMS 2024

作者： Cao, Bingyu School of Computer Science Nanjing University of Information Science and Technology Jiangsu Nanjing China

ISBN: (纸本)9798400717826

The early diagnosis of tumor is very important to improve the survival rate of patients. Magnetic resonance imaging (MRI), as a non-invasive advanced imaging technology, has a wide range of potential applications in tumor detection and evaluation. However, MRI images are often complex and susceptible to noise, and traditional manual analysis methods have limitations in accuracy and efficiency. In recent years, deep learning technology has made a breakthrough in the field of image recognition, which brings a new method and perspective for the intelligent analysis of MRI images. This paper studied the application of several classical deep learning models in MRI image tumor region recognition, including Convolutional Neural Network (CNN) models such as VGG, GoogLeNet, ResNet, EfficientNet and Vision Transformer model, combined with data processing technology and parameter adjustment technology. The performance of these models on data sets is discussed, and a high classification accuracy is finally achieved. This study not only provides an effective solution for automated tumor region detection, but also provides important technical support and reference for clinical tumor diagnosis, and further promotes the intelligent development of imaging diagnosis. © 2024 Copyright held by the owner/author(s).

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on the Method for Testing the signal-to-Noise Ratio at the Output of Microchannel Plate Based on Point source of Electron 10

Research on the Method for Testing the Signal-to-Noise Ratio...

引用

10th symposium on Novel Optoelectronic Detection Technology and Applications

作者： Chen, KeJin Qian, Yunsheng Liu, Jian Zeng, JinNeng School of Electronic and Optical Engineering Nanjing University of Science and Technology Nanjing210094 China North Night Vision Technology Co. Ltd. Kunming650217 China

ISBN: (数字)9781510688155

ISBN: (纸本)9781510688148

signal-to-noise ratio(SNR) at the output is one of the key parameters for evaluating the noise characteristics of the image intensifier. The noise from the microchannel plate (MCP) is a critical factor affecting the noise characteristics of image intensifier. To obtain an objective result of the SNR at the output of MCP under the actual operating conditions of the intensifier, This paper proposes a testing method with a point MCP as the core of a point source of electron for measuring the SNR at the output of MCP. Based on this, the testing system was improved. A point source of electron, composed of a deuterium lamp, adjustable aperture, optical filter, integrating sphere, and point MCP, is used as the input source. It emits a 0.2mm electron beam to simulate the noise test conditions of the image intensifier. An optical transmission component, primarily consisting of quartz optical fibers, is placed near the phosphor screen surface to minimize loss when transmitting light intensity. The photoelectric current is then collected and processed using a photomultiplier tube and signal processing unit. Finally, the SNR at the output of MCP is tested under different input electron beam densities and MCP voltage conditions. Experimental results show that this testing method is relatively objective and accurate, with a repeatability within 5%. The SNR values are consistent with those of the super second generation image intensifier under corresponding conditions, providing an objective evaluation of the MCP noise characteristics. © 2025 SPIE.

关键词： Photomultipliers

来源：评论

学校读者我要写书评

暂无评论

Plaintext-Free Deep Learning for Privacy-Preserving Medical image analysis through Frequency Information Embedding

Plaintext-Free Deep Learning for Privacy-Preserving Medical ...

引用

2025 IEEE international Conference on Acoustics, Speech, and signal processing, ICASSP 2025

作者： Sun, Mengyu Yang, Ziyuan Ran, Maosong Wang, Zhiwen Yu, Hui Zhang, Yi School of Cyber Science and Engineering Sichuan University Chengdu China School of Computer Science Sichuan University Chengdu China

ISBN: (纸本)9798350368741

In the fast-evolving field of medical image analysis, deep Learning (DL)-based methods have achieved tremendous success. However, these methods require plaintext data for training and inference stages, raising privacy concerns, especially in the sensitive area of medical data. To tackle these concerns, this paper proposes a novel framework that uses surrogate images for analysis, eliminating the need for plaintext images. This approach is called Frequency-domain Exchange Style Fusion (FESF). The framework includes two main components: image Hidden Module (IHM) and image Quality Enhancement Module (IQEM). The IHM performs in the frequency domain, blending the features of plaintext medical images into host medical images, and then combines this with IQEM to improve and create surrogate images effectively. During the diagnostic model training process, only surrogate images are used, enabling anonymous analysis without any plaintext data during both training and inference stages. Extensive experiments demonstrate that our framework effectively preserves the privacy of medical images and maintains the diagnostic accuracy of DL models at a relatively high level, proving its effectiveness across various datasets and DL-based models. © 2025 IEEE.

关键词： Deep Learning Frequency-Domain Exchange image Concealment Medical image analysis Privacy Preserving

来源：评论

学校读者我要写书评

暂无评论

Enhancement of Coastline Video Monitoring System Using Structuring Element Morphological Operations

引用

international JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS 2025年第1期16卷 603-611页

作者： Pawana, I. Gusti Ngurah Agung Widyantara, I. Made Oka Sudarma, Made Wiharta, Dewa Made Jayantari, Made Widya Udayana Univ Elect Engn Dept Badung Indonesia Udayana Univ Civil Engn Dept Badung Indonesia

Coastal monitoring is vital in environmental management, disaster mitigation, and addressing climate change impacts. Traditional methods are time-consuming and error- prone, prompting the need for innovative systems. This study introduces the Coastal Video Monitoring System (CoViMos), a novel framework for real-time shoreline detection in tropical regions, specifically at Kedonganan Beach, Bali. The CoViMos framework utilizes advanced video monitoring and optimized morphological operations to address challenges such as environmental noise and dynamic shoreline behavior. Key innovations include Kapur's entropy thresholding enhanced with the Grasshopper Optimization Algorithm (GOA) and structuring elements tailored to the beach's unique features. Sensitivity analysis reveals that a structuring element size of five pixels offers optimal performance, balancing efficiency, and image fidelity. This configuration achieves peak values in quality metrics such as the Peak signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Complex Wavelet SSIM (CWSSIM), and Feature Similarity Index (FSIM) while minimizing Mean Squared Error (MSE) and reducing processing time. The results demonstrate significant improvements in shoreline detection accuracy, with PSNR increasing by 9.3%, SSIM by 1.4%, CWSSIM by 1.7%, and FSIM by 1.6%. processing time decreased by 1.3%, emphasizing the system's computational efficiency. These enhancements ensure more precise shoreline mapping, even in noisy and dynamic environments.

关键词： Coastline detection image processing Video Monitoring System (CoViMoS) morphological operations

来源：评论

学校读者我要写书评

暂无评论

Local Feature Alignment Prompt-Tuning for Few-shot Multimodal Aspect Sentiment analysis

Local Feature Alignment Prompt-Tuning for Few-shot Multimoda...

引用

2025 IEEE international Conference on Acoustics, Speech, and signal processing, ICASSP 2025

作者： Ding, Meirong Zou, Chuang Lin, Hongyi School of Artificial Intelligence South China Normal University Guangzhou China

ISBN: (纸本)9798350368741

Multi-modal aspect-oriented sentiment classification (MASC) is a fine-grain task, which aims to detect the sentiment polarity of specific aspect. However, conventional studies suffer from two issues. It is difficult to collect the annotated multi-modal data in fine-grained domains. Meanwhile, the local information corresponding to aspect words in the image has not been mined, and redundant information can affect the accuracy of fine-grained sentiment analysis. To alleviate the above two issues, we propose a Prompt-tuning method based on Alignment between Aspect and Local images (PAALI). Our approach introduces a novel multi-modal prompt template to bridge the gap between text and visual data modalities. Furthermore, we employ a strategy of randomly masking image patches to align them with aspect word features, capturing deeper semantic information. Extensive experiments on multiple benchmark datasets in few-shot settings consistently demonstrate the superiority and robustness of our PAALI method over state-of-the-art competitors. © 2025 IEEE.

关键词： Few-shot learning Multi-modal aspect-based sentiment analysis Prompt tuning

来源：评论

学校读者我要写书评

暂无评论

A new quality assessment method for airborne SAR images using point, edge and polygon targets

引用

international JOURNAL OF REMOTE SENSING 2025年第4期46卷 1840-1863页

作者： Wang, Jiafeng Feng, Yongjiu Lei, Zhenkun Xi, Mengrong Wang, Chao Xie, Huan Tong, Xiaohua Zhou, Yi Cao, Yuze Tang, Panli Tang, Xiaoyan Tongji Univ Coll Surveying & Geoinformat Shanghai 200092 Peoples R China Tongji Univ Shanghai Key Lab Planetary Mapping & Remote Sensin Shanghai Peoples R China

Accurate assessment and analysis of Synthetic Aperture Radar (SAR) image quality is essential for improving image processing and application potential. To address the limitations of traditional assessment methods that rely solely on point target response energy and consider only single-scale polygon targets, we propose an integrated method for SAR image quality assessment that combines point target energy and morphology, multi-scale edge features and integration of land use categories with multi-scale polygon targets. The assessment results indicate that airborne SAR images exhibit high spatial resolution, better than 1.5 m, with a peak signal-to-noise ratio better than -15 dB. Morphological assessments show that SAR images have morphological consistency above 0.75 and angular consistency above 0.9. With increasing wavelength, the percentage of correctly detected edge features rises from 2.55 in the Ka-band to 3.16 in the S-band. The multi-scale polygon target combined with land use categories indicates that for the same land use category, the difference is less than 0.3 dB, further demonstrating the stability and reliability of the SAR imaging system. Land use categories in cross-polarized SAR images are less affected by noise compared to co-polarized modes. The proposed method exploits the geometric compositional structures within SAR images, improving the accuracy of quality assessment..

关键词： SAR image quality assessment compositional structures multiscale targets land use category

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：