检索结果-内蒙古大学图书馆

From image classification to segmentation: a comprehensive empirical review

Iran Journal of computer science 2025年 1-31页

作者： Ruksana Jim, Jamin Rahman Hossain, Md Jonair Das, Aritra Kabir, Md Mohsin Mridha, M.F. Bangladesh University of Business and Technology Dhaka Bangladesh Department of Computer Science and Engineering American International University-Bangladesh Dhaka 1229 Bangladesh Advanced Machine Intelligence Research Lab Dhaka 1207 Bangladesh Department of Computer Science and Engineering Bangladesh Open University Gazipur 1700 Bangladesh

With the development of the machine learning age, deep learning techniques for computer vision tasks use real-world data to analyze problems across various fields. In particular, it has achieved significant success in image classification, segmentation, and detection on different datasets. As the area of deep learning expands, annotated datasets can be challenging owing to various factors such as the labeling process or difficulty in correctly classifying data, even for experts. Therefore, an extensive survey of machine learning techniques is introduced. Through empirical analysis, several deep learning architectures were applied to benchmark datasets, such as ImageNet, CIFAR-10, COCO, and MNIST. This study includes Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and their variants in these datasets. Additionally, this work reviews the experimental results from modern studies and explains the methodological nuances and comparative performance. In addition to examining the relevance of model performance to preprocessing techniques, various evaluation metrics and promising challenges with future directions for new frontiers are also discussed. Subsequently, this research investigation addresses challenges such as dataset characteristics, model complexity, and computational efficiency and provides insights into the practical implementation of deep learning solutions for real-world applications. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： CIFAR-10 Classification COCO Convolutional neural networks Deep learning Detection ImageNET MNIST Segmentation

来源：评论

学校读者我要写书评

暂无评论

Self-supervised classification of surfaces using reflectance transformation imaging 12

Self-supervised classification of surfaces using reflectance...

引用

12th Colour and Visual Computing Symposium, CVCS 2024

作者： Khawaja, Muhammad Arsalan George, Sony Marzani, Franck Hardeberg, Jon Yngve Mansouri, Alamin ImViA Lab Université de Bourgogne Dijon21000 France Colourlab Department of Computer Science Norwegian University of Science and Technology Gjøvik2815 Norway

Reflectance Transformation Imaging (RTI) is an imaging technique used to analyze objects or surfaces by capturing their appearance under varying illumination directions. This paper proposes two self-supervised learning algorithms to classify surfaces according to their reflectance profiles. The classification problem is addressed using K-means and Self Organizing Map (SOM) neural networks. The proposed methodology is evaluated using both real and synthetic datasets. The primary motivation for our approach is to exploit illumination variation data to enhance surface understanding and detect anomalies. Given the exploratory nature of this task and the lack of ground truth for comparison, a self-supervised method was deemed most suitable. The classification of surfaces using reflectance information has immense applications in fields such as Cultural Heritage (CH) preservation, digitization, and industrial quality control. © 2024 Copyright for this paper by its authors.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Variance-Guided Structured Sparsity in Deep Neural Networks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第6期4卷 1714-1723页

作者： Pandit, Mohammad Khalid Banday, Mahroosh Indian Institute of Technology Delhi Department of Computer Science and Engineering New Delhi110016 India Indian Institute of Technology Delhi Samsung Innovation Lab Bharti School of Telecommunication Technology and Management New Delhi110016 India

The success of deep neural networks, especially convolutional neural networks in various applications, has greatly been possible by the presence of an enormous number of learnable parameters. These parameters increase the learning capacity of the model, but at the same time, it also significantly increases the computational and memory costs. This severely hinders the scalability of these models to limited resource environments, such as IoT devices. The majority of the network weights are known to be redundant and can be removed from the network. This article introduces a regularization scheme, which is the combination of structured sparsity regularization and variance regularization. It simultaneously helps to obtain computationally sparse models by making the majority of parameter groups zero and increasing the variance of nonzero groups to compensate for the accuracy drop. We use sparse group lasso, group sparsity variant of 1 (lasso) regularization, which removes redundant connections and unnecessary neurons from the network. For variance regularization, the KL divergence of the current parameter distribution and the target distribution is minimized, which aims to have the concentration of weights toward zero and a high variance of nonzero weights (skewed distribution). To check the effectiveness of the proposed regularizer, the experiments are performed on various benchmark datasets and it is noticed that variance regularization helps to reduce the accuracy drop caused by sparsity regularization. On MNIST, the trainable parameters are reduced from 331 984 (baseline model) to 57 327 and managed to obtain better accuracy than the baseline (99.6%). Also, on Fashion-MNIST, Cifar-10, and ImageNet the proposed scheme achieved state-of-the-art sparsity with almost no drop in accuracy. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Hardware Security in Evolving FinTech Landscape 5th

Hardware Security in Evolving FinTech Landscape

引用

5th International Conference on Computing and Network Communications, CoCoNet 2023

作者： Bhowmik, Biswajit Dongala, Joshitha Reddy Sudhama, Kruthika K. Antony, Reshma Tresa Girish, K.K. Maharshi Sushrut CAS Lab BRICS Laboratory Department of Computer Science and Engineering National Institute of Technology Karnataka Mangalore Surathkal575025 India

ISBN: (纸本)9789819745395

The assimilation of technology into the financial sector, often referred to as FinTech, has brought about a significant transformation. This shift has not only widened the scope of financial inclusivity but has also fundamentally reshaped the contours of financial solutions delivered. As FinTech solutions continue to empower individuals with greater control over their finances through mobile banking, digital wallets, and advanced data analytics, the security of these innovations becomes paramount. While software security has traditionally received more attention, this paper underscores the significance of hardware security, which serves as the foundational infrastructure for software security measures. It delves into the factors used to evaluate hardware security and outlines various categories of hardware attacks. A case study, focusing on point-of-sale (PoS) systems, exemplifies the importance of hardware security in FinTech. Ultimately, this research contributes to a comprehensive understanding of the evolving FinTech landscape and its implications for both financial inclusion and cybersecurity. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

LLaVA-SG: Leveraging Scene Graphs as Visual Semantic Expression in Vision-Language Models

LLaVA-SG: Leveraging Scene Graphs as Visual Semantic Express...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Wang, Jingyi Ju, Jianzhong Luan, Jian Deng, Zhidong Department of Computer Science and Technology BNRist THUAI Tsinghua University Beijing China Xiaomi AI Lab Beijing China

ISBN: (纸本)9798350368741

Recent advances in large vision-language models (LVLMs) typically employ vision encoders based on the Vision Transformer (ViT) architecture. The division of the images into patches by ViT results in a fragmented perception, thereby hindering the visual understanding capabilities of LVLMs. In this paper, we propose an innovative enhancement to address this limitation by introducing a Scene Graph Expression (SGE) module in LVLMs. This module extracts and structurally expresses the complex semantic information within images, thereby improving the foundational perception and understanding abilities of LVLMs. Extensive experiments demonstrate that integrating our SGE module significantly enhances the LVLM's performance in vision-language tasks, indicating its effectiveness in preserving intricate semantic details and facilitating better visual understanding. © 2025 IEEE.

关键词： Large Language Model Scene Graph Vision-Language Model

来源：评论

学校读者我要写书评

暂无评论

Recognizing Breast Cancer Using Edge-Weighted Texture Features of Histopathology Images

引用

computers, Materials & Continua 2023年第10期77卷 1081-1101页

作者： Arslan Akram Javed Rashid Fahima Hajjej Sobia Yaqoob Muhammad Hamid Asma Arshad Nadeem Sarwar Department of Computer Science and Information Technology Superior UniversityLahore54000Pakistan MLC Lab Maharban HouseHouse#209Zafar ColonyOkara56300Pakistan Information Technology Services University of OkaraOkara56300Pakistan Department of CS&SE International Islamic UniversityIslamabad44000Pakistan Department of Information Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia Department of Computer Science University of OkaraOkara56300Pakistan Department of Statistics and Computer Science University of Veterinary and Animal SciencesLahorePunjab54000Pakistan School of Biochemistry and Biotechnology University of the PunjabLahore54000Pakistan Department of Computer Science Bahria UniversityLahore CampusLahore54600Pakistan

Around one in eight women will be diagnosed with breast cancer at some *** patient outcomes necessitate both early detection and an accurate *** images are routinely utilized in the process of diagnosing breast *** proposed in recent research only focus on classifying breast cancer on specific magnification *** study has focused on using a combined dataset with multiple magnification levels to classify breast cancer.A strategy for detecting breast cancer is provided in the context of this *** image texture data is used with the wavelet transform in this *** proposed method comprises converting histopathological images from Red Green Blue(RGB)to Chrominance of Blue and Chrominance of Red(YCBCR),utilizing a wavelet transform to extract texture information,and classifying the images with Extreme Gradient Boosting(XGBOOST).Furthermore,SMOTE has been used for resampling as the dataset has imbalanced *** suggested method is evaluated using 10-fold cross-validation and achieves an accuracy of 99.27%on the BreakHis 1.040X dataset,98.95%on the BreakHis 1.0100X dataset,98.92%on the BreakHis 1.0200X dataset,98.78%on the BreakHis 1.0400X dataset,and 98.80%on the combined *** findings of this study imply that improved breast cancer detection rates and patient outcomes can be achieved by combining wavelet transformation with textural signals to detect breast cancer in histopathology images.

关键词： Benign and malignant color conversion wavelet domain texture features xgboost

来源：评论

学校读者我要写书评

暂无评论

NeuroBooster: A Robust Classifier for the Discovery of Neuropeptide Sequences based on Meta-learning Approach

NeuroBooster: A Robust Classifier for the Discovery of Neuro...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Hossain Shaon, Md. Shazzad Fahim Sultan, Md. Karim, Tasmin Hossain Alshan, Md. Shoaib Cuzzocrea, Alfredo Akter, Mst Shapna Oakland University Department of Computer Science and Engineering RochesterMI48309 United States Shanto-Mariam University of Creative Technology Department of Computer Science and Engineering Uttara Dhaka1230 Bangladesh University of Calabria IDEA Lab Rende Italy University of Paris City Department of Computer Science Paris France

ISBN: (纸本)9798350362480

Neuropeptides (NPs) are fragile proteins that serve as essential signaling molecules in the neurological system, playing a key role in modulating various physiological processes. Identifying particular neuropeptide sequences relevant to specific disorders would be beneficial for accelerating the development of diagnostic tools. The study proposed another approach to detecting NPs with multi-layer perception (MLP) and a bagging classifier-based meta-learning method called NeuroBooster. This investigation initially focused on five feature extractions based on composition, such as AAC, PAAC, physicochemical properties, QSO, and transfer-learning, such as Bert, and F2V strategies. Subsequently, we used the XGB feature selection method in the Bert and F2V methods to obtain the most 100D crucial features. The predicted probabilistic outcomes of NPs from the 8 preliminary models merged and derived a two-stage dataset with 40 dimensions of features and transmitted them into three classic models and two meta-models, through rigorous criteria for evaluation. Compared with the existing predictor, our proposed model NeuroBooster achieved a higher accuracy of 91.91% in the independent test method. Consequently, we discovered important features in these five models underscoring that physicochemical properties are potential targets for identification, thereby revealing new avenues for therapies. © 2024 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Supervised Relation Extraction Based on labels Contrastive Learning 16

Supervised Relation Extraction Based on Labels Contrastive L...

引用

16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2023

作者： Wang, Yuanru Zhao, Yanhui Cui, Rong-Yi Jin, Guo-Zhe Yanbian University Intelligent Information Processing Lab. Department of Computer Science & Technology Yanji China

ISBN: (纸本)9798350330755

Relation extraction is an important task in natural language processing. Existing relation extraction tasks usually use data augmentation to construct positive and negative samples for contrastive learning training. And previous works lack the accurate selection of positive and negative samples and the full use of relation labels. To overcome these defects, for supervised relational extraction, we propose a labels Contrastive Learning framework (labelsCL) that considers both global and local perspectives, which makes full use of labels for design. Specifically, from a global perspective, we use label and cosine similarity to select the most easily misclassified samples as positive and negative examples for training. From a local perspective, positive and negative samples are selected according to batch and label, and this way of sample selection increases the randomness of sample selection. The contrastive learning framework labelsCL proposed by us makes full use of labels and synthesizes two perspectives to construct contrastive learning labelsCL, which increases the tolerance of the framework. Finally, the experimental results demonstrate the effectiveness of our proposed framework, which is significantly improved on the Semeval-2010 Task 8 dataset compared with the baselines. © 2023 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

On The Design and Training Strategies for Rnn-Based Online Neural Speech Separation Systems 48

On The Design and Training Strategies for Rnn-Based Online N...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Li, Kai Luo, Yi Tsinghua University BNRist Department of Computer Science and Technology China Tencent Ai Lab Shenzhen China

ISBN: (纸本)9781728163277

While the performance of offline neural speech separation systems has been greatly advanced by the recent development of novel neural network architectures, there is typically an inevitable performance gap between the systems and their online variants. In this paper, we investigate how RNN-based offline neural speech separation systems can be changed into their online counterparts while mitigating the performance degradation. We decompose or reorganize the forward and backward RNN layers in a bidirectional RNN layer to form an online path and an offline path, which enables the model to perform both online and offline processing with a same set of model parameters. We further introduce two training strategies for improving the online model via either a pretrained offline model or a multitask training objective. Experiment results show that compared to the online models that are trained from scratch, the proposed layer de-composition and reorganization schemes and training strategies can effectively mitigate the performance gap between two RNN-based offline separation models and their online variants. © 2023 IEEE.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

PVT v2:Improved baselines with Pyramid Vision Transformer

引用

Computational Visual Media 2022年第3期8卷 415-424页

作者： Wenhai Wang Enze Xie Xiang Li Deng-Ping Fan Kaitao Song Ding Liang Tong Lu Ping Luo Ling Shao Shanghai AI Laboratory Shanghai 200232China Department of Computer Science and Technology NanjingUniversityNanjing 210023China Department of Computer Science the University ofHong KongHong Kong 999077China School of Computer Science and Engineering Nanjing University of Science and TechnologyNanjing 210014China Computer Vision Lab ETH ZurichZurich 8092Switzerland SenseTime Beijing 100080China Inception Institute of Artificial Intelligence Abu DhabiUnited Arab Emirates

Transformers have recently lead to encouraging progress in computer *** this work,we present new baselines by improving the original Pyramid Vision Transformer(PVT v1)by adding three designs:(i)a linear complexity attention layer,(ii)an overlapping patch embedding,and(iii)a convolutional feed-forward *** these modifications,PVT v2 reduces the computational complexity of PVT v1 to linearity and provides significant improvements on fundamental vision tasks such as classification,detection,and *** particular,PVT v2 achieves comparable or better performance than recent work such as the Swin *** hope this work will facilitate state-ofthe-art transformer research in computer *** is available at https://***/whai362/PVT.

关键词： transformers dense prediction image classification object detection semantic segmentation

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

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

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：