检索结果-内蒙古大学图书馆

21st IEEE International Symposium on Biomedical Imaging (ISBI)

作者： Zhao, Yang Lu, Ke Wang, Shuhua Lu, Jian Jian, Xue Univ Chinese Acad Sci Sch Engn Sci Beijing Peoples R China Peng Cheng Lab Shenzhen Peoples R China Peking Univ Third Hosp Dept Urol Beijing Peoples R China

ISBN: (纸本)9798350313345;9798350313338

Unsupervised Domain Adaptation (UDA) has received widespread attention in medical image processing, as it can efficiently handle cross-domain data while diminishing the algorithm's reliance on data annotations. The existing UDA approaches based on image alignment have widely used adversarial learning to minimize the appearance differences of cross-domain data, but are prone to corrupting the original structural semantics. conditional diffusion model (CDM) possesses the ability to generate images that adhere to specific distributions, showcasing its tremendous potential in various research areas. In this paper, we propose a UDA method with CDM for magnetic resonance (MR) images that generates prediction masks for target domain data. To enhance the generalization ability of the model to target domain images, a Cross-domain Frequency-mixed Consistency (CFC) method is proposed, which fuses portions of the spectrum from both source and target domain images. Additionally, we propose an Intra-domain Structural Consistency (ISC) method, which utilizes the CutMix strategy to introduce structure-level perturbations to the frequency-mixed image, thereby improving structural consistency in predicting target domain data. Extensive experiments on two MRI datasets demonstrate that our method outperforms other state-of-the-art UDA methods, and also showcased the promising potential of the CDM method for handling medical images for the first time.

关键词： Unsupervised domain adaptation conditional diffusion model Consistency constraint

来源：评论

学校读者我要写书评

暂无评论

SPECKLE REDUCTION IN DUAL-POLARIMETRIC SAR IMAGES BASED ON conditional diffusion model

SPECKLE REDUCTION IN DUAL-POLARIMETRIC SAR IMAGES BASED ON C...

引用

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

作者： Ma, Yaobin Aghababaei, Hossein Chang, Ling PengKe Wei, Jingbo Nanchang Univ Sch Resources & Environm Nanchang Jiangxi Peoples R China Univ Twente Fac Geoinformat Sci & Earth Observat Enschede Netherlands Nanchang Univ Sch Informat Engn Nanchang Jiangxi Peoples R China Nanchang Univ Inst Space Sci & Technol Nanchang Jiangxi Peoples R China

ISBN: (纸本)9798350360332;9798350360325

Reducing speckle while preserving complex structures in images has always been a significant challenge in processing of Synthetic Aperture Radar (SAR) images. This paper proposes a new despeckling method for dual-polarimetric SAR images based on the conditional diffusion model. By explicitly learning specific distributions from the training data, this method better restores the image structures. To support this research, a VV-VH dual-polarimetric dataset is constructed using multitemporal fusion techniques with data obtained from the Sentinel-1 satellite. The proposed method is compared with five other SAR despeckling methods. The results show that this method performs better in preserving image details and effectively removing speckle. Furthermore, this paper introduces a new sampling method for SAR despeckling. Compared to the two existing methods, it achieves better despeckling results and superior structural preservation.

关键词： SAR Despeckling Dual-Polarimetric conditional diffusion model Sentinel-1

来源：评论

学校读者我要写书评

暂无评论

Masked conditional diffusion model for Enhancing Deepfake Detection

Masked Conditional Diffusion Model for Enhancing Deepfake De...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Chen, Tiewen Yang, Shanmin Hu, Shu Fang, Zhenghan Fu, Ying Wu, Xi Wang, Xin Chengdu Univ Informat Technol Chengdu Sichuan Peoples R China Purdue Univ Indianapolis Indianapolis IN USA Johns Hopkins Univ Baltimore MD USA SUNY Albany Albany NY 12222 USA

ISBN: (纸本)9798350359329;9798350359312

Recent studies on deepfake detection have achieved promising results when training and testing faces are from the same dataset. However, their results severely degrade when confronted with forged samples that the model has not yet seen during training. In this paper, deepfake data to help detect deepfakes. this paper present we put a new insight into diffusion model-based data augmentation, and propose a Masked conditional diffusion model (MCDM) for enhancing deepfake detection. It generates a variety of forged faces from a masked pristine one, encouraging the deepfake detection model to learn generic and robust representations without overfitting to special artifacts. Extensive experiments demonstrate that forgery images generated with our method are of high quality and helpful to improve the performance of deepfake detection models.

关键词： conditional diffusion model Deepfake Detection Data Augmentation

来源：评论

学校读者我要写书评

暂无评论

DisC-Diff: Disentangled conditional diffusion model for Multi-contrast MRI Super-Resolution 26th

DisC-Diff: Disentangled Conditional Diffusion Model for Mult...

引用

26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)

作者： Mao, Ye Jiang, Lan Chen, Xi Li, Chao Univ Cambridge Dept Clin Neurosci Cambridge England Univ Dundee Sch Sci & Engn Dundee Scotland Univ Bath Dept Comp Sci Bath Avon England Univ Dundee Sch Med Dundee Scotland

ISBN: (纸本)9783031439988;9783031439995

Multi-contrast magnetic resonance imaging (MRI) is the most common management tool used to characterize neurological disorders based on brain tissue contrasts. However, acquiring high-resolution MRI scans is time-consuming and infeasible under specific conditions. Hence, multi-contrast super-resolution methods have been developed to improve the quality of low-resolution contrasts by leveraging complementary information from multi-contrast MRI. Current deep learning-based super-resolution methods have limitations in estimating restoration uncertainty and avoiding mode collapse. Although the diffusion model has emerged as a promising approach for image enhancement, capturing complex interactions between multiple conditions introduced by multi-contrast MRI super-resolution remains a challenge for clinical applications. In this paper, we propose a disentangled conditional diffusion model, DisC-Diff, for multi-contrast brain MRI super-resolution. It utilizes the sampling-based generation and simple objective function of diffusion models to estimate uncertainty in restorations effectively and ensure a stable optimization process. Moreover, DisC-Diff leverages a disentangled multi-stream network to fully exploit complementary information from multi-contrast MRI, improving model interpretation under multiple conditions of multi-contrast inputs. We validated the effectiveness of DisC-Diff on two datasets: the IXI dataset, which contains 578 normal brains, and a clinical dataset with 316 pathological brains. Our experimental results demonstrate that DisC-Diff outperforms other state-of-the-art methods both quantitatively and visually.

关键词： Magnetic resonance imaging Multi-contrast super-resolution conditional diffusion model

来源：评论

学校读者我要写书评

暂无评论

A Short-Term Wind Power Scenario Generation Method Based on conditional diffusion model

A Short-Term Wind Power Scenario Generation Method Based on ...

引用

2023 IEEE Sustainable Power and Energy Conference, iSPEC 2023

作者： Yan, Jinghao Li, Pai Huang, Yuehui Beijing100192 China

ISBN: (纸本)9798350322699

With the increasing integration of wind power into power systems, wind scenarios can pave the way for improved decision-making in the power system day-ahead scheduling to help manage the inherent uncertainty and variability of wind power. In this paper, a novel deep learning method that uses conditional diffusion model to address the challenges of wind scenario generation. The incorporation of the conditional term by considering the forecast errors distribution and the residual block with U-net structure was employed to handle complex non-linear relationships in temporal dependencies of short-term wind variations, enabling high-quality scenario generated outputs. A case study was conducted where four metrics was utilized to verify the effectiveness of proposed method, and a comparison with CGAN method was laid to demonstrate superiority of model performance. © 2023 IEEE.

关键词： conditional diffusion model Forecast errors distribution Scenario generation Time series evaluation

来源：评论

学校读者我要写书评

暂无评论

Image rain removal algorithm based on conditional diffusion model 2

Image rain removal algorithm based on conditional diffusion ...

引用

2nd International Conference on Image Processing, Computer Vision and Machine Learning, ICICML 2023

作者： Li, Yuxin Liu, Lijun Ma, Baoying Dalian Minzu University School of Science Dalian China

ISBN: (纸本)9798350331417

Rainfall image restoration is one of the research hotspots in the field of computer vision. For raindrops attached to the camera lens or glass, it will hinder the visualization of the complete image and also lead to image degradation. Most of the existing methods for converting raindrop-degraded images into clean images use generative adversarial neural networks (GANs) to solve this problem. At present, the denoising probability diffusion model has been widely used in various fields. Therefore, this paper proposes a raindrop image restoration algorithm based on conditional diffusion model, which improves the U-Net network used for noise prediction in diffusion model. The model is verified and evaluated on the raindrop removal dataset, and has strong generalization ability for raindrop image restoration in the real world. © 2023 IEEE.

关键词： component conditional diffusion model Raindrop Removal U-Net network

来源：评论

学校读者我要写书评

暂无评论

PV-Diff: Probability Prediction of Short-term Photovoltaic Power Interval Based on conditional diffusion model

PV-Diff: Probability Prediction of Short-term Photovoltaic P...

引用

第43届中国控制会议

作者： Liu Jinxiang Liu Meiqin Dong Shanling Zhang Senlin College of Electrical Engineering Zhejiang University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Xi 'an Jiaotong University Jinhua Research Institute of Zhejiang University

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

In order to reduce the impact of photovoltaic power generation uncertainty on power system, a short-term photovoltaic power interval probability prediction method based on conditional diffusion model is proposed. Firstly, the network architecture of diffusion model based on Time U-Net is constructed, which can deeply learn the ability of noise analysis and conditional guidance. Secondly, the reverse diffusion process guided by conditional enhancement is constructed to make the data generated by the model have higher accuracy. Finally, on the basis of generating multiple sets of data, the probability prediction results of photovoltaic power interval are obtained by using kernel density estimation method. Experiments are carried out on the actual data set of Australian Desert Knowledge Solar Center, and the results show that the proposed method has higher probability prediction performance in sunny, cloudy and cloudy weather conditions.

关键词： probability prediction photovoltaic power forecasting conditional diffusion model generation model attention mechanism

来源：评论

学校读者我要写书评

暂无评论

FastSTI: A Fast conditional Pseudo Numerical diffusion model for Spatio-Temporal Traffic Data Imputation

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2024年第12期25卷 20547-20560页

作者： Cheng, Shaokang Osman, Nada Qu, Shiru Ballan, Lamberto Northwestern Polytech Univ Sch Automat Xian 710072 Peoples R China Univ Padua Dept Math Tullio Levi Civita I-35131 Padua Italy

High-quality spatiotemporal traffic data is crucial for intelligent transportation systems (ITS) and their data-driven applications. Inevitably, the issue of missing data caused by various disturbances threatens the reliability of data acquisition. Recent studies of diffusion probability models have demonstrated the superiority of deep generative models in imputation tasks by precisely capturing the spatio-temporal correlation of traffic data. One drawback of diffusion models is their slow sampling/denoising process. In this work, we aim to accelerate the imputation process while retaining the performance. We propose a fast conditional diffusion model for spatiotemporal traffic data imputation (FastSTI). To speed up the process yet, obtain better performance, we propose the application of a high-order pseudo-numerical solver. Our method further revs the imputation by introducing a predefined alignment strategy of variance schedule during the sampling process. Evaluating FastSTI on two types of real-world traffic datasets (traffic speed and flow) with different missing data scenarios proves its ability to impute higher-quality samples in only six sampling steps, especially under high missing rates (60% similar to 90%). The experimental results illustrate a speed-up of 8.3x faster than the current state-of-the-art model while achieving better performance.

关键词： Traffic data imputation conditional diffusion model pseudo numerical methods fast sampling

来源：评论

学校读者我要写书评

暂无评论

Boosting Black-Box Attack to Deep Neural Networks With conditional diffusion models

引用

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 2024年 19卷 5207-5219页

作者： Liu, Renyang Zhou, Wei Zhang, Tianwei Chen, Kangjie Zhao, Jun Lam, Kwok-Yan Nanyang Technol Univ Sch Comp Sci & Engn Singapore 639798 Singapore Yunnan Univ Informat Sci & Engn Sch Kunming 650500 Peoples R China Yunnan Univ Sch Software Kunming 650500 Peoples R China Yunnan Univ Engn Res Ctr Cyberspace Kunming 650500 Peoples R China Nanyang Technol Univ Sch Comp Sci & Engn Singapore 639798 Singapore

Existing black-box attacks have demonstrated promising potential in creating adversarial examples (AE) to deceive deep learning models. Most of these attacks need to handle a vast optimization space and require a large number of queries, hence exhibiting limited practical impacts in real-world scenarios. In this paper, we propose a novel black-box attack strategy, conditional diffusion model Attack (CDMA), to improve the query efficiency of generating AEs under query-limited situations. The key insight of CDMA is to formulate the task of AE synthesis as a distribution transformation problem, i.e., benign examples and their corresponding AEs can be regarded as coming from two distinctive distributions and can transform from each other with a particular converter. Unlike the conventional query-and-optimization approach, we generate eligible AEs with direct conditional transform using the aforementioned data converter, which can significantly reduce the number of queries needed. CDMA adopts the conditional Denoising diffusion Probabilistic model as the converter, which can learn the transformation from clean samples to AEs, and ensure the smooth development of perturbed noise resistant to various defense strategies. We demonstrate the effectiveness and efficiency of CDMA by comparing it with nine state-of-the-art black-box attacks across three benchmark datasets. On average, CDMA can reduce the query count to a handful of times;in most cases, the query count is only ONE. We also show that CDMA can obtain >99% attack success rate for untargeted attacks over all datasets and targeted attack over CIFAR-10 with the noise budget of is an element of =16.

关键词： Multiaccess communication Closed box Perturbation methods Optimization Estimation Task analysis Robustness Adversarial example adversarial attack black-box attack generative-based attack conditional diffusion model

来源：评论

学校读者我要写书评

暂无评论

BS-DIFF: EFFECTIVE BONE SUPPRESSION USING conditional diffusion modelS FROM CHEST X-RAY IMAGES 21

BS-DIFF: EFFECTIVE BONE SUPPRESSION USING CONDITIONAL DIFFUS...

引用

21st IEEE International Symposium on Biomedical Imaging (ISBI)

作者： Chen, Zhanghao Sun, Yifei Ge, Ruiquan Qin, Wenjian Pan, Cheng Deng, Wenming Liu, Zhou Min, Wenwen Elazab, Ahmed Wan, Xiang Wang, Changmiao Hangzhou Dianzi Univ Hangzhou Peoples R China Chinese Acad Sci Shenzhen Inst Adv Technol Shenzhen Peoples R China Sanda Univ Shanghai Peoples R China Chinese Acad Med Sci & Peking Union Med Coll Natl Canc Ctr Natl Clin Res Ctr Canc Canc Hosp Shenzhen Peoples R China Shenzhen Hosp Shenzhen Peoples R China Yunnan Univ Kunming Yunnan Peoples R China Shenzhen Univ Shenzhen Peoples R China Shenzhen Res Inst Big Data Shenzhen Peoples R China Hangzhou Inst Adv Technol Hangzhou Peoples R China

ISBN: (纸本)9798350313345;9798350313338

Chest X-rays (CXRs) are commonly utilized as a low-dose modality for lung screening. Nonetheless, the efficacy of CXRs is somewhat impeded, given that approximately 75% of the lung area overlaps with bone, which in turn hampers the detection and diagnosis of diseases. As a remedial measure, bone suppression techniques have been introduced. The current dual-energy subtraction imaging technique in the clinic requires costly equipment and subjects being exposed to high radiation. To circumvent these issues, deep learning-based image generation algorithms have been proposed. However, existing methods fall short in terms of producing high-quality images and capturing texture details, particularly with pulmonary vessels. To address these issues, this paper proposes a new bone suppression framework, termed BS-Diff, that comprises a conditional diffusion model equipped with a U-Net architecture and a simple enhancement module to incorporate an autoencoder. Our proposed network cannot only generate soft tissue images with a high bone suppression rate but also possesses the capability to capture fine image details. Additionally, we compiled the largest dataset since 2010, including data from 120 patients with high-definition, high-resolution paired CXRs and soft tissue images collected by our affiliated hospital. Extensive experiments, comparative analyses, ablation studies, and clinical evaluations indicate that the proposed BS-Diff outperforms several bone-suppression models across multiple metrics. Our code can be accessed at https://***/Benny0323/BS-Diff.

关键词： Bone suppression CXRs conditional diffusion model Clinical evaluation

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：