检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Tepeli, Yasin I. de Wolf, Mathijs Gonçalves, Joana P. Pattern Recognition & Bioinformatics Department of Intelligent Systems EEMCS Faculty Delft University of Technology Delft2628 XE Netherlands

Selection bias poses a critical challenge for fairness in machine learning, as models trained on data that is less representative of the population might exhibit undesirable behavior for underrepresented profiles. Semi-supervised learning strategies like self-training can mitigate selection bias by incorporating unlabeled data into model training to gain further insight into the distribution of the population. However, conventional self-training seeks to include high-confidence data samples, which may reinforce existing model bias and compromise effectiveness. We propose Metric-DST, a diversity-guided self-training strategy that leverages metric learning and its implicit embedding space to counter confidence-based bias through the inclusion of more diverse samples. Metric-DST learned more robust models in the presence of selection bias for generated and real-world datasets with induced bias, as well as a molecular biology prediction task with intrinsic bias. The Metric-DST learning strategy offers a flexible and widely applicable solution to mitigate selection bias and enhance fairness of machine learning models. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Improving performance of heart rate time series classification by grouping subjects

arXiv

引用

arXiv 2023年

作者： Beekhuizen, Michael Naseri, Arman Tax, David van der Bilt, Ivo Reinders, Marcel Pattern Recognition and Bioinformatics Delft University of Technology Delft Netherlands Department of Cardiology Haga Teaching Hospital The Hague Netherlands

Unlike the more commonly analyzed ECG or PPG data for activity classification, heart rate time series data is less detailed, often noisier and can contain missing data points. Using the BigIdeasLab_STEP dataset, which includes heart rate time series annotated with specific tasks performed by individuals, we sought to determine if general classification was achievable. Our analyses showed that the accuracy is sensitive to the choice of window/stride size. Moreover, we found variable classification performances between subjects due to differences in the physical structure of their hearts. Various techniques were used to minimize this variability. First of all, normalization proved to be a crucial step and significantly improved the performance. Secondly, grouping subjects and performing classification inside a group helped to improve performance and decrease inter-subject variability. Finally, we show that including handcrafted features as input to a deep learning (DL) network improves the classification performance further. Together, these findings indicate that heart rate time series can be utilized for classification tasks like predicting activity. However, normalization or grouping techniques need to be chosen carefully to minimize the issue of subject variability. © 2023, CC BY.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Deep Convolutional Neural Networks as Generic Feature Extractors

arXiv

引用

arXiv 2017年

作者： Hertel, Lars Barth, Erhardt Käster, Thomas Martinetz, Thomas Institute for Signal Processing University of Luebeck Institute for Neuro-and Bioinformatics University of Luebeck Pattern Recognition Company GmbH Luebeck Germany

Recognizing objects in natural images is an intricate problem involving multiple conflicting objectives. Deep convolutional neural networks, trained on large datasets, achieve convincing results and are currently the state-of-Theart approach for this task. However, the long time needed to train such deep networks is a major drawback. We tackled this problem by reusing a previously trained network. For this purpose, we first trained a deep convolutional network on the ILSVRC-12 dataset. We then maintained the learned convolution kernels and only retrained the classification part on different datasets. Using this approach, we achieved an accuracy of 67.68% on CIFAR-100, compared to the previous state-of-The-Art result of 65.43%. Furthermore, our findings indicate that convolutional networks are able to learn generic feature extractors that can be used for different tasks. Copyright © 2017, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Markov Random Field for wind farm planning

Markov Random Field for wind farm planning

引用

IEEE International Conference on Smart Energy Grid Engineering (SEGE)

作者： Hale Cetinay Taygun Kekeç Fernando A. Kuipers D. M. J. Tax Network Architectures and Services Delft University of Technology Delft The Netherlands Pattern Recognition and Bioinformatics Laboratory Delft University of Technology Delft The Netherlands

Many countries aim to integrate a substantial amount of wind energy in the near future. This requires meticulous planning, which is challenging due to the uncertainty in wind profiles. In this paper, we propose a novel framework to discover and investigate those geographic areas that are well suited for building wind farms. We combine the key indicators of wind farm investment using fuzzy sets, and employ multiple-criteria decision analysis to obtain a coarse wind farm suitability value. We further demonstrate how this suitability value can be refined by a Markov Random Field (MRF) that takes the dependencies between adjacent areas into account. As a proof of concept, we take wind farm planning in Turkey, and demonstrate that our MRF modeling can accurately find promising areas.

关键词： Wind farms Investment Markov random fields Wind power generation Decision analysis Planning

来源：评论

学校读者我要写书评

暂无评论

Illumination correction for image stitching

Illumination correction for image stitching

引用

作者： Klement, Sascha Timm, Fabian Barth, Erhardt Institute for Neuro- and Bioinformatics University of Lübeck Ratzeburger Allee 160 23538 Lübeck Germany Pattern Recognition Company GmbH Innovations Campus Lübeck Maria-Goeppert-Straße 1 23562 Lübeck Germany

ISBN: (纸本)9789898425461

Inhomogeneous illumination occurs in nearly every image acquisition system and can hardly be avoided simply by improving the quality of the hardware and the optics. Therefore, software solutions are needed to correct for inhomogeneities, which are particularly visible when combining single images to larger mosaics, e.g. when wrapping textures onto surfaces. Various methods to remove smoothly varying image gradients are available, but often produce artifacts at the image boundary. We present a novel correction method for compensating these artifacts based on the Gaussian pyramid and an appropriate extrapolation of the image boundary. Our framework provides various extrapolation methods and reduces the illumination correction error significantly. Moreover, the correction is done in real-time for high-resolution images and is part of an application for virtual material design.

关键词： Extrapolation

来源：评论

学校读者我要写书评

暂无评论

A preliminary study on visual estimation of taste appreciation

A preliminary study on visual estimation of taste appreciati...

引用

IEEE International Conference on Multimedia and Expo Workshops (ICMEW)

作者： İdil Esen Zülfikar Hamdi Dibeklioğlu Hazim Kemal Ekenel Faculty of Computer and Informatics Istanbul Technical University Istanbul Turkey Pattern Recognition & Bioinformatics Group Delft University of Technology Delft Netherlands

From past to present, individuals' appreciation of taste has always been wondered. Moreover, there is an increasing research interest in measuring taste appreciation. Most of the previous work in this area are psychological studies that rely on manual coding of facial actions and/or emotional expressions. Consequently, these studies depend on human observations. We propose a preliminary study for an automatic visual analysis system that estimates taste liking of individuals. Our results show that the proposed system performs with 56.6% accuracy to classify appreciation in terms of liking, neutral, and disliking categories. In order to explore this result in detail, classification of liking level pairs such as disliking-vs-liking, neutral-vs-liking, and neutral-vs-disliking are also evaluated. Our system achieved 72.5% accuracy for distinguishing between dislike and liking, however, classification accuracy for dislike and neutral, and for liking and neutral have been found to be lower. Our results suggest that reliable and fast automatic systems can be developed to estimate taste appreciation, yet classes of liking and neutral state are not easily separable as indicated in previous studies.

关键词： Feature extraction Videos Databases Support vector machines Visualization Estimation Liquids

来源：评论

学校读者我要写书评

暂无评论

Accurate eye centre localisation by means of gradients

Accurate eye centre localisation by means of gradients

引用

作者： Timm, Fabian Barth, Erhardt Institute for Neuro- and Bioinformatics University of Lübeck Ratzeburger Allee 160 D-23538 Lübeck Germany Pattern Recognition Company GmbH Innovations Campus Lübeck Maria-Goeppert-Strasse 1 D-23562 Lübeck Germany

ISBN: (纸本)9789898425478

The estimation of the eye centres is used in several computer vision applications such as face recognition or eye tracking. Especially for the latter, systems that are remote and rely on available light have become very popular and several methods for accurate eye centre localisation have been proposed. Nevertheless, these methods often fail to accurately estimate the eye centres in difficult scenarios, e.g. low resolution, low contrast, or occlusions. We therefore propose an approach for accurate and robust eye centre localisation by using image gradients. We derive a simple objective function, which only consists of dot products. The maximum of this function corresponds to the location where most gradient vectors intersect and thus to the eye's centre. Although simple, our method is invariant to changes in scale, pose, contrast and variations in illumination. We extensively evaluate our method on the very challenging BioID database for eye centre and iris localisation. Moreover, we compare our method with a wide range of state of the art methods and demonstrate that our method yields a significant improvement regarding both accuracy and robustness.

关键词： Eye tracking

来源：评论

学校读者我要写书评

暂无评论

What Will Your Future Child Look Like? Modeling and Synthesis of Hereditary patterns of Facial Dynamics

What Will Your Future Child Look Like? Modeling and Synthesi...

引用

International Conference on Automatic Face and Gesture recognition

作者： Itir Önal Ertugrul Hamdi Dibeklioglu Department of Computer Engineering Middle East Technical University Ankara Turkey Pattern Recognition & Bioinformatics Group Delft University of Technology Delft The Netherlands

Analysis of kinship from facial images or videos is an important problem. Prior machine learning and computer vision studies approach kinship analysis as a verification or recognition task. In this paper, first time in the literature, we propose a kinship synthesis framework, which generates smile videos of (probable) children from the smile videos of parents. While the appearance of a child's smile is learned using a convolutional encoder-decoder network, another neural network models the dynamics of the corresponding smile. The smile video of the estimated child is synthesized by the combined use of appearance and dynamics models. In order to validate our results, we perform kinship verification experiments using videos of real parents and estimated children generated by our framework. The results show that generated videos of children achieve higher correct verification rates than those of real children. Our results also indicate that the use of generated videos together with the real ones in the training of kinship verification models, increases the accuracy, suggesting that such videos can be used as a synthetic dataset.

关键词： Videos Face Neural networks Eyebrows Nose Mouth Time series analysis

来源：评论

学校读者我要写书评

暂无评论

Building a Deep Learning Model to Detect Osteoporosis from Dental Panoramic X-Ray Image

引用

Open Journal of Applied Sciences 2024年第12期14卷 3480-3489页

作者： Farah Hassan Brangakgi Yasser Khadra Department of Physiology Faculty of Medicine Damascus University Damascus Syria Bioinformatics Syrian Virtual University Damascus Syria Medical Digital Image Processing and Pattern Recognition University of Lyon Lyon France

The project discusses the development of a deep learning model to detect osteoporosis from dental panoramic X-Ray images. It provides an in-depth understanding of human bone structure, osteoporosis, its symptoms, causes, prevalence, and risk factors. The project also explains bone density measurement using dual-energy X-ray absorptiometry (DEXA) and the application of artificial intelligence (AI) and machine learning (ML) in medical imaging. The study uses panoramic dental X-rays to evaluate AI technology in dental imaging and classification of mandible inferior cortical based on Klemetti and Kolmakow criteria. The model architecture consists of convolutional, pooling, fully connected, ReLU, and Softmax layers. Dropout and early stopping are added to the model. The training process uses the train-test approach with 100 epochs and a batch size of 32, and performance evaluation measures such as accuracy, sensitivity, specificity, and F1-score are used to assess the classifier’s performance. The findings and methodology provide a comprehensive understanding of the application of deep learning in the detection of osteoporosis from dental panoramic X-Ray images, and the study demonstrates a robust approach to implementing AI in medical imaging for osteoporosis detection.

关键词： Deep Learning CNN Convolutional Neural Network Python Osteoporosis Dental Panoramic X-Ray

来源：评论

学校读者我要写书评

暂无评论

Conditional random fields for protein function prediction

Conditional random fields for protein function prediction

引用

8th IAPR International Conference on pattern recognition in bioinformatics, PRIB 2013

作者： Gehrmann, Thies Loog, Marco Reinders, Marcel J.T. De Ridder, Dick Delft Bioinformatics Lab Delft University of Technology Mekelweg 4 2628 CD Delft Netherlands Pattern Recognition Lab Delft University of Technology Mekelweg 4 2628 CD Delft Netherlands Netherlands Bioinformatics Centre P.O. Box 9101 6500 HB Nijmegen Netherlands Kluyver Centre for Genomics of Industrial Fermentation P.O. Box 5057 2600 GA Delft Netherlands

ISBN: (纸本)9783642391583

Markov Random Fields (MRF) have been shown to be good predictors of functional annotation, using protein-protein interaction data. Many other sources of data can also be used in this prediction task, but they are typically not *** this study, we extend a method using MRFs in order to allow the use of additional data. A conditional random field (CRF) model is proposed as an alternative to an MRF model in order to remove the requirement of modeling relationships between the sources of data. We observe that a substantial performance improvement is possible using additional data, such as genetic interaction networks. The improvement gained from each source of evidence is not the same for each protein function, indicating that each source supplies different information. We demonstrate that CRFs can be used to efficiently integrate various sources of data to predict functional annotations. © 2013 Springer-Verlag Berlin Heidelberg.

关键词： Forecasting

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：