2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)最新文献_第2页

A New Focus+Context Visualization Technique for Inspecting Black Oil Reservoir Models 黑色油藏模型检测的焦点+上下文可视化新技术

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00013

Luiz Felipe Netto, W. Celes

{"title":"A New Focus+Context Visualization Technique for Inspecting Black Oil Reservoir Models","authors":"Luiz Felipe Netto, W. Celes","doi":"10.1109/sibgrapi54419.2021.00013","DOIUrl":"https://doi.org/10.1109/sibgrapi54419.2021.00013","url":null,"abstract":"In this paper, we propose a new visualization technique to inspect simulated black oil reservoir models. Numerical reservoir simulation is widely used in the oil & gas industry to predict and plan the exploration of petroleum fields. Of particular interest, it is crucial to understand the physical phenomena around injector and producer wells: how is the distribution of pressure, how oil/gas/water saturation varies over time, and others. Traditional visualization techniques only provide local insights, being hard to understand the three-dimension physical behavior. Another challenge is to handle today’s massive models. We propose an efficient and effective focus+context visualization technique employing a cutaway approach, having the wells as objects of interest. We explore cone-shaped and box-shaped view-dependent cutting surfaces. We also allow the user to control the cutting surface aperture freely and to freeze the cut to gain motion parallax depth cues during model exploration. The proposed rendering algorithm runs on GPU, delivering real-time frame rate even for large reservoir models.","PeriodicalId":197423,"journal":{"name":"2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128978916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

New hierarchy-based segmentation layer: towards automatic marker proposal 新的基于层次结构的分割层:面向自动标记方案

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00055

G. B. Fonseca, Romain Negrel, B. Perret, J. Cousty, S. Guimarães

{"title":"New hierarchy-based segmentation layer: towards automatic marker proposal","authors":"G. B. Fonseca, Romain Negrel, B. Perret, J. Cousty, S. Guimarães","doi":"10.1109/sibgrapi54419.2021.00055","DOIUrl":"https://doi.org/10.1109/sibgrapi54419.2021.00055","url":null,"abstract":"Image segmentation is an ill-posed problem by definition, as it is not always possible to automatically select which object appearing in an image is the object of interest. To deal with this issue, prior knowledge in the form of human-given markers can be included in the segmentation pipeline. Even though user interaction can drastically improve segmentation results, it is an expensive resource, and finding ways to reduce human effort on an interactive segmentation loop is of great interest. In this work, we propose a new segmentation layer to be used with deep neural networks, which allows us to create and train in an end-to-end fashion a marker creation network. To train the network, we propose a loss function composed of: a segmentation loss using the proposed differentiable segmentation layer; and a set of regularization functions that enforce the desired characteristics on the produced markers. We showed that by using the proposed layer and loss function, we can train the network to automatically generate markers that recover a good segmentation and have desirable shape characteristics. This behavior is observed on the training dataset, as well as on four unseen datasets.","PeriodicalId":197423,"journal":{"name":"2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122227401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An egg image noise model for digital visual counting processing 一种用于数字视觉计数处理的鸡蛋图像噪声模型

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00047

C. Behaine, J. S. Ide

{"title":"An egg image noise model for digital visual counting processing","authors":"C. Behaine, J. S. Ide","doi":"10.1109/sibgrapi54419.2021.00047","DOIUrl":"https://doi.org/10.1109/sibgrapi54419.2021.00047","url":null,"abstract":"Contactless counting is a suitable technique for the measurement of fragile commodities, acting as a successful tool for industrial production control. Visual counting processing is one of the most common contactless methods for non-invasive measurements. However, the creation of accurate models for processing images in realistic scenarios is still challenging due to the existence of noise in optical sensors. This paper proposes an egg image noise model for digital visual counting processing that incorporates particular aspects of real images in such acquisition systems. The matching function is defined in hue saturation value (HSV) color space, and a classical nearest neighbor cluster classification is utilized for the counting. Validation experiments are executed with low and high diversity test images, and the performance of the proposed model is compared to existing methods. The matching function results suggest that the introduced egg image noise model is able to represent more accurately complex aspects of egg images in an industrial environment. The comparative results show that the proposed model significantly improves digital visual counting, in terms of egg counting errors, and outperforms in 9% the second best method.","PeriodicalId":197423,"journal":{"name":"2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131266600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Musical Hyperlapse: A Multimodal Approach to Accelerate First-Person Videos 音乐超缩:加速第一人称视频的多模式方法

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00033

Diognei de Matos, W. Ramos, Luiz Romanhol, Erickson R. Nascimento

引用次数: 1

One-Class Classifiers for Novelties Detection in Electrical Submersible Pumps 用于电潜泵新特性检测的单类分类器

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/SIBGRAPI54419.2021.00061

Gabriel Soares Baptista, L. H. S. Mello, Thiago Oliveira-Santos, F. M. Varejão, M. Ribeiro, A. Rodrigues

{"title":"One-Class Classifiers for Novelties Detection in Electrical Submersible Pumps","authors":"Gabriel Soares Baptista, L. H. S. Mello, Thiago Oliveira-Santos, F. M. Varejão, M. Ribeiro, A. Rodrigues","doi":"10.1109/SIBGRAPI54419.2021.00061","DOIUrl":"https://doi.org/10.1109/SIBGRAPI54419.2021.00061","url":null,"abstract":"Detecting anomalies and fault novelties is of high interest in the industry due to the scarcity of fault examples to train classification systems. In this article two algorithms for anomaly detection, One-Class SVM and Isolation Forest, are successfully used as effective methods for detecting fault novelties in problems of electrical submersible pumps. Faults in submersible electric pumps generate an enormous cost for companies in the oil and gas sector, since the cost of stopping production to change the equipment is excessive, which makes it necessary to identify problems before implementation. Empirical evaluation shows that both one-class classifiers performed satisfactorily, obtaining macro f-measure values of approximately 0.86. For comparison purposes, a Random Forest trained in a conventional binary classification manner is tested and achieved a macro f-measure of 0.95. Results show that the proposed solutions can have practical applications in the classification of problems in electrical submersible pumps, changing the way the oil and gas industry addresses this difficulty.","PeriodicalId":197423,"journal":{"name":"2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129348832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Gravity Alignment for Single Panorama Depth Inference 单全景深度推断的重力对准

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00015

Matheus A. Bergmann, Paulo G. L. Pinto, T. L. T. D. Silveira, C. Jung

引用次数: 3

Descriptive Image Gradient from Edge-Weighted Image Graph and Random Forests 基于边缘加权图像图和随机森林的描述性图像梯度

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00053

R. Almeida, Zenilton K. G. Patrocínio, A. Araújo, Ewa Kijak, Simon Malinowski, S. Guimarães

{"title":"Descriptive Image Gradient from Edge-Weighted Image Graph and Random Forests","authors":"R. Almeida, Zenilton K. G. Patrocínio, A. Araújo, Ewa Kijak, Simon Malinowski, S. Guimarães","doi":"10.1109/sibgrapi54419.2021.00053","DOIUrl":"https://doi.org/10.1109/sibgrapi54419.2021.00053","url":null,"abstract":"Creating an image gradient is a transformation process that aims to enhance desirable properties of an image, whilst leaving aside noise and non-descriptive characteristics. Many algorithms in image processing rely on a good image gradient to perform properly on tasks such as edge detection and segmentation. In this work, we propose a novel method to create a very descriptive image gradient using edge-weighted graphs as a structured input for the random forest algorithm. On the one side, the spatial connectivity of the image pixels gives us a structured representation of a grid graph, creating a particular transformed space close to the spatial domain of the images, but strengthened with relational aspects. On the other side, random forest is a fast, simple and scalable machine learning method, suited to work with high-dimensional and small samples of data. The local variation representation of the edge-weighted graph, aggregated with the random forest implicit regularization process, serves as a gradient operator delimited by the graph adjacency relation in which noises are mitigated and desirable characteristics reinforced. In this work, we discuss the graph structure, machine learning on graphs and the random forest operating on graphs for image processing. We tested the created gradients on the hierarchical watershed algorithm, a segmentation method that is dependent on the input gradient. The segmentation results obtained from the proposed method demonstrated to be superior compared to other popular gradients methods.","PeriodicalId":197423,"journal":{"name":"2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127296358","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Interactive Visualizations to Support Randomized Clinical Trial Monitoring 交互式可视化支持随机临床试验监测

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00028

M. F. Rey, C. Freitas

{"title":"Interactive Visualizations to Support Randomized Clinical Trial Monitoring","authors":"M. F. Rey, C. Freitas","doi":"10.1109/sibgrapi54419.2021.00028","DOIUrl":"https://doi.org/10.1109/sibgrapi54419.2021.00028","url":null,"abstract":"Despite current technological advances, interactive tools to facilitate the analysis of data collected during clinical trials are still not widely available. Such a scenario makes researchers rely on time-consuming extractions from databases and subsequent application of analytical methods by statisticians to obtain results from which they can get insights. Moreover, during clinical trials, researchers need to keep track of subjects’ progress by monitoring their participation and the quality of the data collected at specific phases of the trial. We have developed a visualization-based interface that assists the epidemiologists of a randomized clinical trial focused on the effects of lifestyle intervention in developing type 2 diabetes for patients with Gestational Diabetes Mellitus (GDM). Coaches are responsible for the intervention, and research assistants collect data from hundreds of questionnaires and clinical exams. We adopted user-centered design principles, which allowed continuous improvements to the visualizations and interactive features during a year-long development process. Besides typical selection and filtering features, the visualizations we have designed allow the research team to monitor each participant’s progress and perform analyses that facilitate findings in and between subjects’ histories. Two formal evaluations were also performed with experts and non-experts, where the visualization-based interface proved to be intuitive and useful for assisting coaching activities, monitoring the progress of data collection, and performing analyses.","PeriodicalId":197423,"journal":{"name":"2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117124222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Message from Program Chairs 节目主持人的信息

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00005

引用次数: 1

Machine Learning Bias in Computer Vision: Why do I have to care? 计算机视觉中的机器学习偏差:为什么我必须关心?

2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2021-10-01 DOI: 10.1109/sibgrapi54419.2021.00010

Camila Laranjeira, V. F. Mota, J. A. D. Santos

引用次数: 1