检索结果-内蒙古大学图书馆

Deep learning Approach for High Recall Pneumonia Classification with Swin Transformer and L2 Regularization 23

学校读者我要写书评

暂无评论

Deep Learning Approach for High Recall Pneumonia Classificat...

8th International Conference on Sustainable Information Engineering and Technology, SIET 2023

作者： Pinasthika, Krisna Sofyanda, Erika Yussi Ulumiyah, Silfiatul Muflikhah, Lailil Computer Vision Research Group Brawijaya University Indonesia Technology Education Learning Research Group Brawijaya University Indonesia Information System Research Group Brawijaya University Indonesia Department of Informatics Engineering Brawijaya University Indonesia

ISBN: (纸本)9798400708503

Pneumonia is one of the lung infections that can be lethal to people. Pneumonia can be diagnosed by conducting an X-ray scan of the human chest. Even though X-ray scanning is straightforward, the process of identifying pneumonia in X-ray data is still difficult to do even by radiologists. Therefore, a computational algorithm is needed that can be used to analyze X-ray image data. The use of Deep learning (DL) methods for analyzing images has been proven to work well in various cases, especially X-ray data. Based on previous research, the transformer architecture has the advantage of analyzing with a larger receptive area if compared to Convolutional Neural Networks (CNN) which have limited kernel size. In this study, we used the Swin transformer in classifying pneumonia diseases through X-ray images. Through the study conducted, the Swin Transformer Architecture is able to provide a fairly good evaluation metric when compared to the use of convolutional neural networks. The Tiny variant of the Swin Transformer architecture is able to provide evaluation metrics with a precision value of 0.870, recall of 0.995, and F1-Score of 0.928. On the other hand, the use of L2 regularization in stage 3 and stage 4 of the Swin Tiny architecture is able to provide a high recall value of 0.997, but the use of L2 regularization does not provide an increase in precision and F1-Score. In biomedical cases that are only concerned with recall values, the use of L2 regularization on Swin Tiny can provide advantages. © 2023 ACM.

关键词： computer vision

A Semiautomatic Image Processing-Based Method for Binary Segmentation of Lungs in Computed Tomography Images

学校读者我要写书评

暂无评论

SN computer Science 2024年第6期5卷 689页

作者： Ramos, Leo Pineda, Israel School of Mathematical and Computational Sciences Yachay Tech University Deep Learning for Autonomous Driving Robotics and Computer Vision (DeepARC) Research Group Yachay Tech University Kauel Inc. Computer Vision Center Universitat Autònoma de Barcelona College of Sciences and Engineering Universidad San Francisco de Quito

Precise biomedical image segmentation is pivotal in medical diagnosis and treatment. Among various methodologies, image processing-based techniques are useful due to their swift and consistent results, free from the dependency on extensive training data required by deep learning approaches. This study introduces a tri-phase, semi-automated pipeline for lung computed tomography image segmentation, beginning with a preprocessing phase that includes contrast enhancement, noise reduction, and edge detection to improve image quality. This is followed by the application of the region growing method for segmentation and is completed with morphological closing to refine the final output. Comparative tests reveal that our technique surpasses prevalent methods like split and merge, Otsu’s technique, and basic region growing in terms of the dice coefficient (DC) and intersection over union (IoU). Specifically, our approach achieved an average DC of 0.9633, contrasting with 0.4668, 0.005, and 0.4067 recorded by Otsu’s, split and merge, and basic region growing techniques respectively. Similarly, the average IoU for our method was 0.9341, in comparison to 0.3105, 0.0660, and 0.2605 for the aforementioned methods. Additionally, comparative analysis with contemporary segmentation models demonstrates that our approach is competitive, even without relying on large training datasets. These results highlight our method’s efficiency and practicality, distinguishing it as a compelling option in medical scenarios where access to extensive training data is limited. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Image analysis Image processing Image segmentation Medical image Pattern recognition Region growing

Spatial and Colour Opponency in Anatomically Constrained Deep Networks

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Harris, Ethan Mihai, Daniela Hare, Jonathon Vision Learning and Control Group Electronics and Computer Science University of Southampton

Colour vision has long fascinated scientists, who have sought to understand both the physiology of the mechanics of colour vision and the psychophysics of colour perception. We consider representations of colour in anatomically constrained convolutional deep neural networks. Following ideas from neuroscience, we classify cells in early layers into groups relating to their spectral and spatial functionality. We show the emergence of single and double opponent cells in our networks and characterise how the distribution of these cells changes under the constraint of a retinal bottleneck. Our experiments not only open up a new understanding of how deep networks process spatial and colour information, but also provide new tools to help understand the black box of deep learning. The code for all experiments is avaialable at https://***/ecs-vlc/opponency. Copyright © 2019, The Authors. All rights reserved.

关键词： Color vision

Avoiding hashing and encouraging visual semantics in referential emergent language games

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Mihai, Daniela Hare, Jonathon Vision Learning and Control Group Electronics and Computer Science University of Southampton

There has been an increasing interest in the area of emergent communication between agents which learn to play referential signalling games with realistic images. In this work we consider the signalling game setting of Havrylov and Titov and investigate the effect of the feature extractor’s weights and of the task being solved on the visual semantics learned or captured by the models. We impose various augmentation to the input images and additional tasks in the game with the aim to induce visual representations which capture conceptual properties of images. Through our set of experiments, we demonstrate that communication systems which capture visual semantics can be learned in a completely self-supervised manner by playing the right types of game. Copyright © 2019, The Authors. All rights reserved.

关键词： Semantics

Robust Autonomous Vehicle Pursuit Without Expert Steering Labels

学校读者我要写书评

暂无评论

IEEE Robotics and Automation Letters 2023年第10期8卷 6595-6602页

作者： Pan, Jiaxin Zhou, Changyao Gladkova, Mariia Khan, Qadeer Cremers, Daniel Technical University of Munich Computer Vision Group Garching85748 Germany Munich Data Science Institute Garching85748 Germany Munich Center for Machine Learning Munchen80333 Germany University of Oxford OxfordOX1 3AZ United Kingdom

In this work, we present a learning method for both lateral and longitudinal motion control of an ego-vehicle for the task of vehicle pursuit. The car being controlled does not have a pre-defined route, rather it reactively adapts to follow a target vehicle while maintaining a safety distance. To train our model, we do not rely on steering labels recorded from an expert driver, but effectively leverage a classical controller as an offline label generation tool. In addition, we account for the errors in the predicted control values, which can lead to a loss of tracking and catastrophic crashes of the controlled vehicle. To this end, we propose an effective data augmentation approach, which allows to train a network that is capable of handling different views of the target vehicle. During the pursuit, the target vehicle is firstly localized using a Convolutional Neural Network. The network takes a single RGB image along with cars' velocities and estimates target vehicle's pose with respect to the ego-vehicle. This information is then fed to a Multi-Layer Perceptron, which regresses the control commands for the ego-vehicle, namely throttle and steering angle. We extensively validate our approach using the CARLA simulator on a wide range of terrains. Our method demonstrates real-time performance, robustness to different scenarios including unseen trajectories and high route completion. © 2016 IEEE.

关键词： Deep learning

LOH and behold: Web-scale visual search, recommendation and clustering using locally optimized hashing 14

学校读者我要写书评

暂无评论

LOH and behold: Web-scale visual search, recommendation and ...

computer vision - ECCV 2016 Workshops, Proceedings

作者： Kalantidis, Yannis Kennedy, Lyndon Nguyen, Huy Mellina, Clayton Shamma, David A. Computer Vision and Machine Learning Group Flickr Yahoo San Francisco United States Futurewei Technologies Inc Santa Clara United States CWI: Centrum Wiskunde and Informatica Amsterdam Netherlands

ISBN: (纸本)9783319466033

We propose a novel hashing-based matching scheme, called Locally Optimized Hashing (LOH), based on a state-of-the-art quantization algorithm that can be used for efficient, large-scale search, recommendation, clustering, and deduplication. We show that matching with LOH only requires set intersections and summations to compute and so is easily implemented in generic distributed computing systems. We further show application of LOH to: (a) large-scale search tasks where performance is on par with other state-of-the-art hashing approaches (b) large-scale recommendation where queries consisting of thousands of images can be used to generate accurate recommendations from collections of hundreds of millions of images;and (c) efficient clustering with a graph-based algorithm that can be scaled to massive collections in a distributed environment or can be used for deduplication for small collections, like search results, performing better than traditional hashing approaches while only requiring a few milliseconds to run. In this paper we experiment on datasets of up to 100 million images, but in practice our system can scale to larger collections and can be used for other types of data that have a vector representation in a Euclidean space. © Springer International Publishing Switzerland 2016.

关键词： Vector spaces

A biologically inspired visual working memory for deep networks

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Harris, Ethan Niranjan, Mahesan Hare, Jonathon Vision Learning and Control Group Department of Electronics and Computer Science University of Southampton

The ability to look multiple times through a series of pose-adjusted glimpses is fundamental to human vision. This critical faculty allows us to understand highly complex visual scenes. Short term memory plays an integral role in aggregating the information obtained from these glimpses and informing our interpretation of the scene. Computational models have attempted to address glimpsing and visual attention but have failed to incorporate the notion of memory. We introduce a novel, biologically inspired visual working memory architecture that we term the Hebb-Rosenblatt memory. We subsequently introduce a fully differentiable Short Term Attentive Working Memory model (STAWM) which uses transformational attention to learn a memory over each image it sees. The state of our Hebb-Rosenblatt memory is embedded in STAWM as the weights space of a layer. By projecting different queries through this layer we can obtain goal-oriented latent representations for tasks including classification and visual reconstruction. Our model obtains highly competitive classification performance on MNIST and CIFAR-10. As demonstrated through the CelebA dataset, to perform reconstruction the model learns to make a sequence of updates to a canvas which constitute a parts-based representation. Classification with the self supervised representation obtained from MNIST is shown to be in line with the state of the art models (none of which use a visual attention mechanism). Finally, we show that STAWM can be trained under the dual constraints of classification and reconstruction to provide an interpretable visual sketchpad which helps open the ‘black-box’ of deep learning. Copyright © 2019, The Authors. All rights reserved.

关键词： Memory architecture

Human-centric indoor scene synthesis using stochastic grammar

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Qi, Siyuan Zhu, Yixin Huang, Siyuan Jiang, Chenfanfu Zhu, Song-Chun UCLA Center for Vision Cognition Learning and Autonomy UPenn Computer Graphics Group

We present a human-centric method to sample and synthesize 3D room layouts and 2D images thereof, to obtain large-scale 2D/3D image data with perfect per-pixel ground truth. An attributed spatial And-Or graph (S-AOG) is proposed to represent indoor scenes. The S-AOG is a probabilistic grammar model, in which the terminal nodes are object entities. Human contexts as contextual relations are encoded by Markov Random Fields (MRF) on the terminal nodes. We learn the distributions from an indoor scene dataset and sample new layouts using Monte Carlo Markov Chain. Experiments demonstrate that our method can robustly sample a large variety of realistic room layouts based on three criteria: (i) visual realism comparing to a state-of-the-art room arrangement method, (ii) accuracy of the affordance maps with respect to groundtruth, and (ii) the functionality and naturalness of synthesized rooms evaluated by human subjects. Copyright © 2018, The Authors. All rights reserved.

关键词： Stochastic systems

A survey of historical document image datasets

学校读者我要写书评

暂无评论

A survey of historical document image datasets

作者： Nikolaidou, Konstantina Seuret, Mathias Mokayed, Hamam Liwicki, Marcus EISLAB Machine Learning Group Luleå University of Technology Aurorum 1 Norrbotten Luleå97187 Sweden Pattern Recognition Lab Computer Vision Group Friedrich-Alexander-Universität Martensstr. 3 Bavaria Erlangen91058 Germany

This paper presents a systematic literature review of image datasets for document image analysis, focusing on historical documents, such as handwritten manuscripts and early prints. Finding appropriate datasets for historical document analysis is a crucial prerequisite to facilitate research using different machine learning algorithms. However, because of the very large variety of the actual data (e.g., scripts, tasks, dates, support systems, and amount of deterioration), the different formats for data and label representation, and the different evaluation processes and benchmarks, finding appropriate datasets is a difficult task. This work fills this gap, presenting a meta-study on existing datasets. After a systematic selection process (according to PRISMA guidelines), we select 65 studies that are chosen based on different factors, such as the year of publication, number of methods implemented in the article, reliability of the chosen algorithms, dataset size, and journal outlet. We summarize each study by assigning it to one of three pre-defined tasks: document classification, layout structure, or content analysis. We present the statistics, document type, language, tasks, input visual aspects, and ground truth information for every dataset. In addition, we provide the benchmark tasks and results from these papers or recent competitions. We further discuss gaps and challenges in this domain. We advocate for providing conversion tools to common formats (e.g., COCO format for computer vision tasks) and always providing a set of evaluation metrics, instead of just one, to make results comparable across studies. © 2022, The Author(s).

关键词： Deterioration