检索结果-内蒙古大学图书馆

A machine learning approach for skin lesion classification on iOS: implementing and optimizing a convolutional transfer learning model with Create ML

引用

International Journal of computers and Applications 2024年第8期46卷 666-685页

作者： Benedek Szabo, Aron Katona, Jozsef Department of Software Development and Application Institute of Computer Engineering University of Dunaujvaros Dunaujvaros Hungary Department of Instrumentation and Automation Institute of Electronics and Communication Systems Kandó Kálmán Faculty of Electrical Engineering Obuda University Budapest Hungary Department of Informatics GAMF Faculty of Engineering and Computer Science Kecskemét Hungary

The integration of machine learning (ML) into mobile applications presents unique challenges, particularly in resource-constrained environments such as iOS devices. Skin lesion classification is a critical task in dermatology, where accurate and efficient diagnostic tools can significantly aid in early detection of malignant lesions. This study aims to implement a machine learning-based iOS mobile application and develop a binary classification model for skin lesion images to determine whether a lesion is malignant. The research utilized Create ML to develop a convolutional neural network optimized for iOS, employing a transfer learning approach. A logistic regression model was cascaded with the convolutional neural network N to enhance classification accuracy. The model's performance was assessed through various validation metrics, ensuring its robustness and efficiency within the constraints of mobile hardware. A curated dataset from the International Skin Imaging Collaboration archive was used for training and testing. The model achieved an accuracy of 92%, a precision of 90%, a recall of 93%, and an F1-score of 91% in classifying skin lesions. These metrics validate the model's efficacy in identifying malignant lesions. Data curation involved collecting, labeling, and preparing a dataset from publicly available sources, ensuring the inclusion of diagnostically relevant features. The final model was integrated into an iOS application using Core ML and Vision frameworks. The developed application demonstrates reliable performance in classifying skin lesions with high accuracy on iOS devices. The inclusion of comprehensive performance metrics justifies the efficacy of the proposed approach. Future work will explore enhancements in model architecture and object detection capabilities to further improve diagnostic precision and application usability. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

Heterogeneous Domain Adaptation via Correlative and Discriminative Feature Learning

引用

IEEE Transactions on Multimedia 2025年 27卷 3447-3461页

作者： Lu, Yuwu Lin, Dewei Shen, Linlin Zhou, Yicong Pan, Jiahui South China Normal University School of Artificial Intelligence Foshan528225 China Shenzhen University College of Computer Science and Software Engineering Shenzhen518055 China Shenzhen University Guangdong Key Laboratory of Intelligent Information Processing Shenzhen518060 China University of Nottingham Ningbo China Department of Computer Science Ningbo315100 China University of Macau Department of Computer and Information Science 999078 China

Heterogeneous domain adaptation seeks to learn an effective classifier or regression model for unlabeled target samples by using the well-labeled source samples but residing in different feature spaces and lying different distributions. Most recent works have concentrated on learning domain-invariant feature representations to minimize the distribution divergence via target pseudo-labels. However, two critical issues need to be further explored: 1) new feature representations should be not only domain-invariant but also category-correlative and discriminative and 2) alleviating the negative transfer caused by the incorrect pseudo-labeling target samples could boost the adaptation performance during the iterative learning process. To address these issues, in this paper, we put forward a novel heterogeneous domain adaptation method to learn category-correlative and discriminative representations, referred to as correlative and discriminative feature learning (CDFL). Specifically, CDFL aims to learn a feature space where class-specific feature correlations between the source and target domains are maximized, the divergences of marginal and conditional distribution between the source and target domains are minimized, and the distances of inter-class distribution are forced to be maximized to ensure the discriminative ability. Meanwhile, a selective pseudo-labeling procedure based on the correlation coefficient and classifier prediction is introduced to boost class-specific feature correlation and discriminative distribution alignment in an iteration way. Extensive experiments certify that CDFL outperforms the State-of-the-Art algorithms on five standard benchmarks. © 1999-2012 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A Multi-Metric Ranking with Label Correlations Approach for Library Migration Recommendations 31

A Multi-Metric Ranking with Label Correlations Approach for ...

引用

31st IEEE International Conference on software Analysis, Evolution and Reengineering, SANER 2024

作者： Zhang, Jiancheng Luo, Qin Wu, Peng School of Computer Science and Software Engineering SouthWest Petroleum University Chengdu China School of Information and Engineering Sichuan Tourism University Chengdu China

ISBN: (纸本)9798350330663

While third-party libraries provide benefit to software systems, they also bring unique challenges. It often happens that developers need to replace some already-used libraries with other functionality-equivalent libraries. However, it is not easy to find a relevant candidate from overwhelming libraries. Despite several approaches have been proposed to mine library migrations from historical data, the study of library recommendation from the perspective of both open-source projects and third-party libraries is lacking. Therefore, conducting such research may assist developer better select suitable third-party libraries. In this paper, we propose a multi-metric ranking with label correlations (MMRLC) algorithm, which can recommend libraries holistically from the both perspectives. Not only does it mine library migrations from existing software data, MMRLC further leverages label correlations of libraries in Maven Central Repository to make recommendations. To demonstrate the usefulness, three popular algorithms were conducted on a benchmark dataset for comparison. The results show that our approach can recommend libraries with precision @ 1 of 0.8454 and recall @20 of 0.9301. Moreover, to demonstrate the generality, we select 366 libraries and resort to TagWiki to generate the related library labels, and the results show that our approach still has comparable performance. © 2024 IEEE.

关键词： Libraries

来源：评论

学校读者我要写书评

暂无评论

Generation of hyperspectral point clouds: Mapping, compression and rendering

引用

computers and Graphics (Pergamon) 2022年 106卷 267-276页

作者： López, Alfonso Jurado, Juan M. Jiménez-Pérez, J. Roberto Feito, Francisco R. Department of Computer Science University of Jaén Spain Department of Software Engineering University of Granada Spain

Hyperspectral data are being increasingly used for the characterization and understanding of real-world scenarios. In this field, UAV-based sensors bring the opportunity to collect multiple samples from different viewpoints. Thus, light-material interactions of real objects may be observed in outdoor scenarios with a significant spatial resolution (5 cm/pixel). Nevertheless, the generation of hyperspectral 3D data still poses challenges in post-processing due to the high geometric deformation of images. Most of the current solutions use both LiDAR (Light Detection and Ranging) and hyperspectral sensors, which are integrated into the same acquisition system. However, these present several limitations due to errors derived from inertial measurements for data fusion and the spatial resolution according to the LiDAR capabilities. In this work, a method is proposed for the generation of hyperspectral point clouds. Input data are formed by push-broom hyperspectral images and 3D point clouds. On the one hand, the point clouds may be obtained by applying a typical photogrammetric workflow or LiDAR technology. Then, hyperspectral images are geometrically corrected and aligned with the RGB orthomosaic. Accordingly, hyperspectral data are ready to be mapped on the 3D point cloud. This procedure is implemented on the GPU by testing which points are visible for each pixel of the hyperspectral imagery. This work also provides a novel solution to generate, compress and render 3D hyperspectral point clouds, enabling the study of geometry and the hyperspectral response of natural and artificial materials in the real world. © 2022 Elsevier Ltd

关键词： Data fusion

来源：评论

学校读者我要写书评

暂无评论

A Trust-based Framework for the Detection and Avoidance of Blackhole Nodes in Opportunistic Networks 9

A Trust-based Framework for the Detection and Avoidance of B...

引用

9th International Conference on Signal Processing and Communication, ICSC 2023

作者： Kundu, Protyasha Saha, Swarnadeep Kundu, Anindita Barai, Smritikona Department of Computer Science and Engineering Heritage Institute of Technology Kolkata India Department of Software Systems School of Computer Science and Engineering Vellore Institute of Technology Vellore India

ISBN: (纸本)9798350383201

An Opportunistic Network (OppNet), as opposed to a ubiquitous centralized network, relies on sporadic and opportunistic encounters between nodes to facilitate communication. The uncertainty about the node's nature presents challenges when nodes with malicious intents are present in the network. One category of malicious nodes includes blackhole nodes, which receive messages but drop them intentionally. In this paper, we introduce a novel trust function designed to detect blackhole nodes within the network. Our function has been widely tested and simulated to validate and increase the efficacy of transmission of messages in the network. We have performed a comparative analysis with various protocols (PROPHET, Epidemic and Spray And Wait) on several parameters such as Delivery Ratio, Overhead Ratio and Number of Messages dropped. The results demonstrate that our trust function is able to correctly detect and avoid blackhole nodes. We observed significant improvement in delivery ratio in Epidemic Routing (47%), Prophet (41%), and Spray And Wait Routing (17%) protocols when adjusting blackhole nodes while keeping the total network nodes constant. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

HSM:Hand Surface Menu Solution for Interaction in VR 10

HSM:Hand Surface Menu Solution for Interaction in VR

引用

10th International Conference on Virtual Reality, ICVR 2024

作者： Zhang, Hailong Li, Lin Xue, Chang Wei, XiaoCai Zhou, Yin Hefei University of Technology School of Computer Science and Information Engineering Hefei China Hefei University of Technology School of The Software Engineering Hefei China

ISBN: (纸本)9798350364231

The menu interaction methods in VR, such as floating menus, are still considered unnatural. A solution is proposed in this paper where menus are tightly attached to the user's palm. Firstly, use UV mapping technology to attach the menu to the palm. Secondly, optimal layout parameters for the hand surface menu were determined through user experiments. Finally, a dataset was built to train the interaction error detection model, aiming to improve interaction success rates under various hand poses in VR. Experimental results demonstrate that compared to other VR menu methods, HSM reduces the average interaction error rate by 30.1% and shortens the average completion time by 12.4% under the optimal hand pose. © 2024 IEEE.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

SRGAN Optimization through Enhanced Network Structure and Multi-Loss Functions 4

SRGAN Optimization through Enhanced Network Structure and Mu...

引用

4th IEEE International Conference on Data science and computer Application, ICDSCA 2024

作者： Chen, Zexin Chen, Xiaoxin College of Computer Science and Software Engineering Shenzhen University Shenzhen China College of Medical Information Engineering Guangdong Pharmaceutical University Guangzhou China

ISBN: (纸本)9798350368239

An enhanced SRGAN image super-resolution algorithm is proposed to address SRGAN's training instability and limited global information capture. This enhancement optimizes both the network structure and loss function. In network structure optimization, group normalization is introduced as an alternative to batch normalization, addressing its dependency on batch size and improving stability during small-batch training. Additionally, Global Context Block (GCBlock) is integrated into the residual block, enhancing the network’s global information capture and detail representation capabilities. In regard to the loss function, this paper introduces L1 loss and MS-SSIM loss as alternatives to the conventional MSE loss. This approach aims to enhance the quality of the generated image in both pixel space and structural information while reducing the smoothing effect. In the experimental section, the ablation experiments evaluate the performance of models with and without GCBlock on five test sets, including Urban100 and Manga109. The results demonstrate that GCBlock improves both detail expressiveness and global information capture, with an average PSNR gain of 0.1388 and SSIM improvement of 0.008. In comparison experiments, the improved algorithm achieved an average PSNR increase of 0.270 and an average SSIM increase of 0.018 on the aforementioned test set, indicating a significant improvement in detail expression and visual quality. Finally, the limitations of the enhanced algorithm are discussed, and potential future research directions are proposed. ©2024 IEEE.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Monte Carlo Simulation-Based Robust Workflow Scheduling for Spot Instances in Cloud Environments

引用

Tsinghua science and Technology 2024年第1期29卷 112-126页

作者： Quanwang Wu Jianzhao Fang Jie Zeng Junhao Wen Fengji Luo College of Computer Science Chongqing UniversityChongqing 400044China National Experimental Teaching Demonstration Center Chongqing UniversityChongqing 400044China College of Big Data and Software Engineering Chongqing UniversityChongqing 400044China School of Civil Engineering The University of SydneySydney 2006Australia

When deploying workflows in cloud environments,the use of Spot Instances(SIs)is intriguing as they are much cheaper than on-demand ***,Sls are volatile and may be revoked at any time,which results in a more challenging scheduling problem involving execution interruption and hence hinders the successful handling of conventional cloud workflow scheduling *** some scheduling methods for Sls have been proposed,most of them are no more applicable to the latest Sls,as they have evolved by eliminating bidding and simplifying the pricing *** study focuses on how to minimize the execution cost with a deadline constraint when deploying a workflow on volatile Sls in cloud *** on Monte Carlo simulation and list scheduling,a stochastic scheduling method called MCLS is devised to optimize a utility function introduced for this *** the Monte Carlo simulation framework,MCLS employs sampled task execution time to build solutions via deadline distribution and list scheduling,and then returns the most robust solution from all the candidates with a specific evaluation mechanism and selection *** results show that the performance of MCLS is more competitive comparedwithtraditionalalgorithms.

关键词： constrained optimization Monte Carlo simulation robustness Spot Instances(Sls) workflow scheduling

来源：评论

学校读者我要写书评

暂无评论

Leveraging Transfer Learning for Spatio-Temporal Human Activity Recognition from Video Sequences

引用

computers, Materials & Continua 2023年第3期74卷 5017-5033页

作者： Umair Muneer Butt Hadiqa Aman Ullah Sukumar Letchmunan Iqra Tariq Fadratul Hafinaz Hassan Tieng Wei Koh School of Computer Sciences Universiti Sains MalaysiaPenang1180Malaysia Department of Computer Science The University of ChenabGujrat50700Pakistan Department of Software Engineering and Information System Universiti Putra MalaysiaSelangor43400Malaysia

Human Activity Recognition(HAR)is an active research area due to its applications in pervasive computing,human-computer interaction,artificial intelligence,health care,and social ***,dynamic environments and anthropometric differences between individuals make it harder to recognize *** study focused on human activity in video sequences acquired with an RGB camera because of its vast range of real-world *** uses two-stream ConvNet to extract spatial and temporal information and proposes a fine-tuned deep neural ***,the transfer learning paradigm is adopted to extract varied and fixed frames while reusing object identification *** state-of-the-art pre-trained models are exploited to find the best model for spatial feature *** temporal sequence,this study uses dense optical flow following the two-stream ConvNet and Bidirectional Long Short TermMemory(BiLSTM)to capture *** state-of-the-art datasets,UCF101 and HMDB51,are used for evaluation *** addition,seven state-of-the-art optimizers are used to fine-tune the proposed network ***,this study utilizes an ensemble mechanism to aggregate spatial-temporal features using a four-stream Convolutional Neural Network(CNN),where two streams use RGB *** contrast,the other uses optical flow ***,the proposed ensemble approach using max hard voting outperforms state-ofthe-art methods with 96.30%and 90.07%accuracies on the UCF101 and HMDB51 datasets.

关键词： Human activity recognition deep learning transfer learning neural network ensemble learning spatio-temporal

来源：评论

学校读者我要写书评

暂无评论

FakeDiffer: Distributional Disparity Learning on Differentiated Reconstruction for Face Forgery Detection 39

FakeDiffer: Distributional Disparity Learning on Differentia...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Wang, Bo Zhang, Zhao Zhao, Suiyi Ye, Xianming Zhang, Haijun Wang, Meng School of Computer Science and Information Engineering Hefei University of Technology Hefei China Yunnan Key Laboratory of Software Engineering Yunnan China Department of Electronic and Computer Engineering University of Pretoria South Africa Shenzhen China

ISBN: (纸本)157735897X

Existing face forgery detection methods achieve promising performance when training and testing forgery data are from identical manipulation types, while they fail to generalize well to unseen samples. In this paper, we experimentally investigate and find that the poor generalization of the methods mainly arises from their overfitting on the known fake patterns. Excessively focused on seen fakes, those detectors fail to effectively learn image-intrinsic information and the distributional disparity between real and fake images. Then, to address this issue, we redefine fake learning as real-fake distributional disparity learning. We propose a novel deepfake detection framework learning distributional disparity based on the differentiated reconstruction on real and fake images for improved generalization. Specifically, distributional disparity learning on differentiated reconstruction of the real and fake images, enforces the model to learn image-invariant intrinsic representations. The reconstruction on real and fake images forces the decoders to learn the distribution of real and fake images, respectively. Moreover, to avoid the influence from the specificalization of the known fake patterns, we further propose the information interaction learning on the encoded intrinsic information and the pixel disparity between the input image and its reconstruction to distinguish face forgeries that are even unknown. Extensive experiments on large-scale benchmark datasets demonstrated the effectiveness of addressing the overfitting issue of the classification network, and verified the superior performance of our method. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep learning

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：