检索结果-内蒙古大学图书馆

25th International Conference on Pattern Recognition (ICPR)

作者： Soh, Jae Woong Cho, Nam Ik Seoul Natl Univ INMC Dept ECE Seoul South Korea

ISBN: (纸本)9781728188089

image denoising is an essential part of many image processing and computer vision tasks due to inevitable noise corruption during image acquisition. Traditionally, many researchers have investigated image priors for the denoising, within the Bayesian perspective based on image properties and statistics. Recently, deep convolutional neural networks (CNNs) have shown great success in image denoising by incorporating large-scale synthetic datasets. However, they both have pros and cons. While the deep CNNs are powerful for removing the noise with known statistics, they tend to lack flexibility and practicality for the blind and real-world noise. Moreover, they cannot easily employ explicit priors. On the other hand, traditional nonlearning methods can involve explicit image priors, but they require considerable computation time and cannot exploit large-scale external datasets. In this paper, we present a CNN-based method that leverages the advantages of both methods based on the Bayesian perspective. Concretely, we divide the blind i mage denoising problem into sub-problems and conquer each inference problem separately. As the CNN is a powerful tool for inference, our method is rooted in CNNs and propose a novel design of network for efficient inference. With our proposed method, we can successfully remove blind and real-world noise, with a moderate number of parameters of universal CNN.

关键词： image denoising

来源：评论

学校读者我要写书评

暂无评论

Employing Two-Dimensional Word Embedding for Difficult Tabular Data Stream Classification

arXiv

引用

arXiv 2024年

作者： Zyblewski, Pawel Department of Systems and Computer Networks Faculty of Information and Communication Technology Wroclaw University of Science and Technology Wybrzeże Wyspiańskiego 27 Wroclaw50-370 Poland

Rapid technological advances are inherently linked to the increased amount of data, a substantial portion of which can be interpreted as data stream, capable of exhibiting the phenomenon of concept drift and having a high imbalance ratio. Consequently, developing new approaches to classifying difficult data streams is a rapidly growing research area. At the same time, the proliferation of deep learning and transfer learning, as well as the success of convolutional neural networks in computer vision tasks, have contributed to the emergence of a new research trend, namely Multi-Dimensional Encoding (MDE), focusing on transforming tabular data into a homogeneous form of a discrete digital signal. This paper proposes Streaming Super Tabular Machine learning (SSTML), thereby exploring for the first time the potential of MDE in the difficult data stream classification task. SSTML encodes consecutive data chunks into an image representation using the STML algorithm and then performs a single ResNet-18 training epoch. Experiments conducted on synthetic and real data streams have demonstrated the ability of SSTML to achieve classification quality statistically significantly superior to state-of-the-art algorithms while maintaining comparable processing time. © 2024, CC BY-NC-SA.

关键词： deep learning

来源：评论

学校读者我要写书评

暂无评论

real- time Compressed Domain Face Recognition Using deep learning

Real- Time Compressed Domain Face Recognition Using Deep Lea...

引用

Recent Trends in Microelectronics, Automation, Computing and Communications Systems (ICMACC), International Conference on

作者： Ranjan K Senapati Indhu Gondra Pragna Panyala PMK Prasad Dept of ECE VNRVJIET Bachupally Nizampet Hyderabad India Dept of ECE GVPCE for Women Madhuravada Vishakapatnam Andra Pradesh India

ISBN: (数字)9798350366570

ISBN: (纸本)9798350366587

The growing demand for facial recognition systems necessitates efficient processing of images, particularly in resource-constrained environments. Compressed images offer a significant advantage in terms of storage and transmission bandwidth, but traditional face recognition algorithms struggle with the information loss inherent in compression techniques like JPEG. This paper explores the feasibility of using a VGG-16 based approach for real-time face recognition directly in the compressed domain. We leverage transfer learning by fine-tuning a pretrained VGG-16 model on a dataset of compressed facial images. This approach capitalizes on the powerful feature extraction capabilities of VGG-16 while adapting it to the challenges of compressed images, such as artifacts and information loss. Our evaluation demonstrates that the trained model achieves accurate face recognition using compressed images, paving the way for real-world applications with bandwidth and storage limitations. Future work could involve exploring the impact of different compression techniques on recognition accuracy and potentially improving robustness against compression artifacts. This work acknowledges the inherent tradeoff between the compression ratio of images and the achievable recognition accuracy. In this thesis, we are investigating this relationship to identify the optimal balance between storage efficiency and recognition performance for real-world deployments.

关键词： Adaptation models image coding image recognition Accuracy Face recognition Transfer learning Transform coding Bandwidth Feature extraction Robustness

来源：评论

学校读者我要写书评

暂无评论

deep learning for fast denoising filtering in ultrasound localization microscopy

引用

PHYSICS IN MEDICINE AND BIOLOGY 2023年第20期68卷 205002-205002页

作者： Yu, Xiangyang Luan, Shunyao Lei, Shuang Huang, Jing Liu, Zeqing Xue, Xudong Ma, Teng Ding, Yi Zhu, Benpeng Huazhong Univ Sci & Technol Wuhan Natl Lab Optoelect Sch Integrated Circuit Wuhan Peoples R China Huazhong Univ Sci & Technol Hubei Canc Hosp TongJi Med Coll Dept Radiat Oncol Wuhan Hubei Peoples R China Chinese Acad Sci Inst Biomed & Hlth Engn Shenzhen Inst Adv Technol Beijing Peoples R China

Objective. Addition of a denoising filter step in ultrasound localization microscopy (ULM) has been shown to effectively reduce the error localizations of microbubbles (MBs) and achieve resolution improvement for super-resolution ultrasound (SR-US) imaging. However, previous image-denoising methods (e.g. block-matching 3D, BM3D) requires long data processing times, making ULM only able to be processed offline. This work introduces a new way to reduce data processing time through deep learning. Approach. In this study, we propose deep learning (DL) denoising based on contrastive semi-supervised network (CS-Net). The neural network is mainly trained with simulated MBs data to extract MB signals from noise. And the performances of CS-Net denoising are evaluated in both in vitro flow phantom experiment and in vivo experiment of New Zealand rabbit tumor. Main results. For in vitro flow phantom experiment, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of single microbubble image are 26.91 dB and 4.01 dB, repectively. For in vivo animal experiment , the SNR and CNR were 12.29 dB and 6.06 dB. In addition, single microvessel of 24 mu m and two microvessels separated by 46 mu m could be clearly displayed. Most importantly,, the CS-Net denoising speeds for in vitro and in vivo experiments were 0.041 s frame-1 and 0.062 s frame-1, respectively. Significance. DL denoising based on CS-Net can improve the resolution of SR-US as well as reducing denoising time, thereby making further contributions to the clinical real-time imaging of ULM.

关键词： deep learning (DL) denoising contrastive semi-supervised network (CS-Net) microvessels super-resolution ultrasound (SR-US)

来源：评论

学校读者我要写书评

暂无评论

real-time, deep Synthetic Aperture Sonar (SAS) Autofocus

Real-Time, Deep Synthetic Aperture Sonar (SAS) Autofocus

引用

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

作者： Gerg, Isaac D. Monga, Vishal Penn State Univ Appl Res Lab University Pk PA 16802 USA Penn State Univ Sch EECS University Pk PA 16802 USA

ISBN: (纸本)9781665403696

Synthetic aperture sonar (SAS) requires precise time-of-flight measurements of the transmitted/received waveform to produce well-focused imagery. It is not uncommon for errors in these measurements to be present resulting in image defocusing. To overcome this, an autofocus algorithm is employed as a post-processing step after image reconstruction to improve image focus. A particular class of these algorithms can be framed as a sharpness/contrast metric-based optimization. To improve convergence, a hand-crafted weighting function to remove "bad" areas of the image is sometimes applied to the image-undertest before the optimization procedure. Additionally, dozens of iterations are necessary for convergence which is a large compute burden for low size, weight, and power (SWaP) systems. We propose a deep learning technique to overcome these limitations and implicitly learn the weighting function in a data-driven manner. Our proposed method, which we call deep Autofocus, uses features from the single-look-complex (SLC) to estimate the phase correction which is applied in k-space. Furthermore, we train our algorithm on batches of training imagery so that during deployment, only a single iteration of our method is sufficient to autofocus. We show results demonstrating the robustness of our technique by comparing our results to four commonly used image sharpness metrics. Our results demonstrate deep Autofocus can produce imagery perceptually better than common iterative techniques but at a lower computational cost. We conclude that deep Autofocus can provide a more favorable cost-quality tradeoff than alternatives with significant potential of future research.

关键词： Training deep learning Measurement uncertainty Geoscience and remote sensing Robustness real-time systems Synthetic aperture sonar

来源：评论

学校读者我要写书评

暂无评论

Redefining Cystoscopy With AI: Bladder Cancer Diagnosis Using an Efficient Hybrid CNN-Transformer Model

Redefining Cystoscopy With AI: Bladder Cancer Diagnosis Usin...

引用

IEEE International Conference on image processing

作者： Meryem Amaouche Ouassim Karrakchou Mounir Ghogho Anouar El Ghazzaly Mohamed Alami Ahmed Ameur International University of Rabat TICLab Morocco Mohammed V Military Training Hospital Rabat Morocco Faculty of Engineering University of Leeds UK

ISBN: (数字)9798350349399

ISBN: (纸本)9798350349405

Bladder cancer ranks within the top 10 most diagnosed cancers worldwide and is among the most expensive cancers to treat due to the high recurrence rates which require lifetime follow-ups. The primary tool for diagnosis is cystoscopy, which heavily relies on doctors’ expertise and interpretation. Therefore, annually, numerous cases are either undiagnosed or misdiagnosed and treated as urinary infections. To address this, we suggest a deep learning approach for bladder cancer detection and segmentation which combines CNNs with a lightweight positional-encoding-free transformer and dual attention gates that fuse self and spatial attention for feature enhancement. The architecture suggested in this paper is efficient making it suitable for medical scenarios that require real time inference. Experiments have proven that this model addresses the critical need for a balance between computational efficiency and diagnostic accuracy in cystoscopic imaging as despite its small size it rivals large models in performance.

关键词： Training image segmentation Accuracy Computational modeling Bladder Transformers real-time systems

来源：评论

学校读者我要写书评

暂无评论

Novel Hybrid deep learning Model for Enhancing Skin Cancer Detection

Novel Hybrid Deep Learning Model for Enhancing Skin Cancer D...

引用

International image processing, Applications and Systems Conference (IPAS)

作者： Jamel Smida Mohamed Karim Azizi Anandh Sam Chandra Bose Amor Smida College of Applied Science AlMaarefa University Riyadh Kingdom of Saudi Arabia Microwave Electronics Research Laboratory Faculty of Sciences of Tunis University Tunis El Manar Tunis Tunisia Department of Medical Equipment Technology College of Applied Medical Sciences Al-Majmaah Kingdom of Saudi Arabia

ISBN: (数字)9798331506520

ISBN: (纸本)9798331506537

Skin Cancer is the most serious form of cancer, often spreading to other parts of the body over time. Early and accurate identification of skin lesions can improve the treatment options and increase the probability of curing the disease before it progresses. However, automatic detection of skin cancer faces difficulties because of the imbalanced datasets, affected areas covered by hair or moles, and other visual obstructions. In recent years, deep learning (DL) techniques have been widely used to address these challenges. Despite advancements, achieving consistently high accuracy remains difficult. To tackle these challenges, the research proposes a novel hybrid DL model for efficient skin cancer detection from dermoscopic images. The proposed network consists of three key modules: feature extraction using the Xception network, feature enhancement through the Squeeze and Excitation network (SENet), and final classification using a transformer module. This combination of DL models effectively identifies SC. The proposed network is evaluated using benign and malignant dermoscopic images collected from the Kaggle website. Necessary preprocessing steps, such as resizing, grayscale conversion, hair removal, and data balancing, are applied. The proposed network is compared with popular DL networks such as MobileNet, Xception, and VGG-16. The experimental outcome demonstrates that the proposed network outperforms these models, achieving a higher accuracy, recall, precision, and F1 score of 96.83%, 97%, 96.67%, and 96.83%, respectively. The metrics value shows that the proposed network is promising and can be helpful for dermatologists in identifying skin cancer with minimal effort.

关键词： deep learning Hair Visualization Accuracy Feature extraction Transformers Skin real-time systems Mobile applications Skin cancer

来源：评论

学校读者我要写书评

暂无评论

real-time, wide-field and high-quality single snapshot imaging of optical properties with profile correction using deep learning

引用

BIOMEDICAL OPTICS EXPRESS 2020年第10期11卷 5701-5716页

作者： Aguenounon, Enagnon Smith, Jason T. Al-Taher, Mahdi Diana, Michele Intes, Xavier Gioux, Sylvain Univ Strasbourg ICube Lab 300 Blvd Sebastien Brant F-67412 Illkirch Graffenstaden France Rensselaer Polytech Inst Dept Biomed Engn Troy NY 12180 USA IHU Strasbourg Inst Image Guided Surg Strasbourg France Maastricht Univ Med Ctr Maastricht Netherlands IRCAD Res Inst Digest Canc Strasbourg France

The development of real-time, wide-field and quantitative diffuse optical imaging methods to visualize functional and structural biomarkers of living tissues is a pressing need for numerous clinical applications including image-guided surgery. In this context, Spatial Frequency Domain Imaging (SFDI) is an attractive method allowing for the fast estimation of optical properties using the Single Snapshot of Optical Properties (SSOP) approach. Herein, we present a novel implementation of SSOP based on a combination of deep learning network at the filtering stage and Graphics processing Units (GPU) capable of simultaneous high visual quality image reconstruction, surface profile correction and accurate optical property (OP) extraction in real-time across large fields of view. In the most optimal implementation, the presented methodology demonstrates megapixel profile-corrected OP imaging with results comparable to that of profile-corrected SFDI, with a processing time of 18 ms and errors relative to SFDI method less than 10% in both profilometry and profile-corrected OPs. This novel processing framework lays the foundation for real-time multispectral quantitative diffuse optical imaging for surgical guidance and healthcare applications. All code and data used for this work is publicly available at *** under the resources tab. (C) 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

关键词： Diffuse optical imaging image reconstruction Imaging techniques Optical properties Small animal imaging Spatial frequency

来源：评论

学校读者我要写书评

暂无评论

Automatic Recognition of Concealed Fish Bones under Laryngoscopy: A Practical AI Model Based on YOLO-V5

引用

LARYNGOSCOPE 2024年第5期134卷 2162-2169页

作者： Tao, Xiaoyao Zhao, Xu Liu, Hairui Wang, Jinqiao Tian, Chunhui Liu, Longsheng Ding, Yujie Chen, Xue Liu, Yehai Anhui Med Univ Affiliated Hosp 1 Otorhinolaryngol Head & Neck Surg Dept Hefei Peoples R China Chinese Acad Sci Inst Automat Beijing Peoples R China Chinese Acad Sci Beijing Peoples R China China Univ Geosci Sch Informat Engn Beijing Peoples R China China Univ Geosci Beijing Peoples R China Anhui Med Univ Suzhou Hosp Otolaryngol Head & Neck Surg Dept Suzhou Peoples R China Anhui Med Univ Suzhou Hosp Suzhou Peoples R China Anhui Med Univ Chaohu Hosp Otolaryngol Head & Neck Surg Dept Hefei Peoples R China Feixi Cty Peoples Hosp Otolaryngol Head & Neck Surg Dept Hefei Peoples R China Feidong Cty Peoples Hosp Otolaryngol Head & Neck Surg Dept Hefei Peoples R China Anhui Med Univ Affiliated Hosp 1 Otorhinolaryngol Head & Neck Surg Dept 218 Jixi Rd Hefei 230031 Anhui Peoples R China

Background: Fish bone impaction is one of the most common problems encountered in otolaryngology emergencies. Due to their small and transparent nature, as well as the complexity of pharyngeal anatomy, identifying fish bones efficiently under laryngoscopy requires substantial clinical experience. This study aims to create an AI model to assist clinicians in detecting pharyngeal fish bones more efficiently under ***: Totally 3133 laryngoscopic images related to fish bones were collected for model training and validation. The images in the training dataset were trained using the YOLO-V5 algorithm model. After training, the model was validated and its performance was evaluated using a test dataset. The model's predictions were compared to those of human experts. Seven laryngoscopic videos related to fish bone were used to validate real-time target detection by the ***: The model trained in YOLO-V5 demonstrated good generalization and performance, with an average precision of 0.857 when the intersection over union (IOU) threshold was set to 0.5. The precision, recall rate, and F1 scores of the model are 0.909, 0.818, and 0.87, respectively. The overall accuracy of the model in the validation set was 0.821, comparable to that of ENT specialists. The model processed each image in 0.012 s, significantly faster than human processing (p < 0.001). Furthermore, the model exhibited outstanding performance in video ***: Our AI model based on YOLO-V5 effectively identifies and localizes fish bone foreign bodies in static laryngoscopic images and dynamic videos. It shows great potential for clinical application.

关键词： deep learning YOLO fish bones laryngoscopy computer vision

来源：评论

学校读者我要写书评

暂无评论

GastroNet FusionAI: Artificial Intelligence Diagnosis of Clinical Gastric Cancer Tumors

GastroNet FusionAI: Artificial Intelligence Diagnosis of Cli...

引用

IEEE International Conference on Information Communication and Signal processing (ICICSP)

作者： Jialun Xu Xiayang Wu Zhenyuan Xu School of Statistics Nanjing Audit University Nanjing China School of Statistics Nanjing University of Posts and Telecommunications Nanjing China

ISBN: (数字)9798350355895

ISBN: (纸本)9798350355901

The significant advancements in natural image processing have sparked widespread interest in applying similar deep learning techniques to medical imaging. Unlike natural images, detecting details in medical images requires capturing finer details, necessitating the use of specially designed network architectures rather than standard ones developed for large-scale, downscaled natural images. Additionally, medical imaging often requires the fusion of images from multiple perspectives to form an accurate diagnosis. Therefore, we propose a novel AI algorithm called GastroNet FusionAI, an advanced algorithm specifically designed for gastric cancer diagnosis. It employs an innovative image fusion strategy that effectively integrates images from four different imaging perspectives in gastric cancer detection. This approach not only provides a more comprehensive view but also significantly enhances the accuracy of disease detection by enriching the image content. The development of GastroNet FusionAI aims to improve the early detection rate of gastric cancer, which is crucial for increasing patient survival rates. Gastric cancer is one of the most common types of cancer globally and has a high incidence in certain regions. Despite the enormous potential of AI technology in gastric cancer detection, research on its effectiveness and reliability remains relatively scarce. Existing studies are often limited by small sample sizes, homogeneous datasets, or a lack of rare gastric cancer subtypes. To overcome these challenges, we employed real-time Test-time Augmentation (R-TTA) to expand the original dataset (consisting of 54,706 images) to 273,530 images. This method not only improved the model's learning efficiency but also effectively reduced the risk of overfitting. Our model was trained on a preprocessed dataset from three different hospitals and four different imaging perspectives, and its performance was evaluated using cross-validation to ensure the broad applicability of

关键词： Accuracy Signal processing algorithms Cancer detection real-time systems Artificial intelligence Medical diagnostic imaging image fusion Cancer Overfitting Tumors

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：