Marta Duarte, V. Coch, Jovania Dias, S. Botelho, N. L. Filho, Paulo L. J. Drews-Jr
{"title":"Thermographic Non-Invasive Inspection Modelling of Fertilizer Pipelines Using Neural Networks","authors":"Marta Duarte, V. Coch, Jovania Dias, S. Botelho, N. L. Filho, Paulo L. J. Drews-Jr","doi":"10.1109/SIBGRAPI51738.2020.00045","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00045","url":null,"abstract":"Industry pipeline fault, like blockage can create major problems for engineers and financial loss for the company. The blockage detection is necessary for smooth functioning of an industry and safety of the environment. This work presents a model for non-invasive inspection of pipes. It proposes the use of a neural network to identify the obstruction stage in fertilizer industry, using external thermal images obtained from the pipelines. A dataset capable of mapping the external thermal behavior in profile of the internal deposit is developed. The Multilayer Perceptron neural network was able to learn the thermal pixel mapping in a deposit profile, obtaining satisfactory results.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129648958","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}
{"title":"Weakly Supervised Character Detection for License Plate Recognition","authors":"Luis Felipe de Araujo Zeni, C. Jung","doi":"10.1109/SIBGRAPI51738.2020.00037","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00037","url":null,"abstract":"Automatic Licence Plate Recognition (ALPR) is an essential task in the context of intelligent transportation systems. In a typical ALPR pipeline, the last stage receives as input a cropped license plate region and outputs the string with the plate characters. This paper presents a Weakly Supervised Character Detection (WSCD) approach that requires only string-level an-notations (as in generic text recognition methods) but is able to detect characters individually (as in detection-based methods, which require character-level annotations). The proposed method is evaluated in five distinct datasets and present very competitive results against other state-of-the-art methods.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"138 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132513666","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}
Germain Garcia Zanabria, Erick Gomez Nieto, Jaqueline Silveira, Jorge Poco, M. Nery, S. Adorno, L. G. Nonato
{"title":"Mirante: A visualization tool for analyzing urban crimes","authors":"Germain Garcia Zanabria, Erick Gomez Nieto, Jaqueline Silveira, Jorge Poco, M. Nery, S. Adorno, L. G. Nonato","doi":"10.1109/SIBGRAPI51738.2020.00028","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00028","url":null,"abstract":"Visualization assisted crime analysis tools used by public security agencies are usually designed to explore large urban areas, relying on grid-based heatmaps to reveal spatial crime distribution in whole districts, regions, and neighborhoods. Therefore, those tools can hardly identify micro-scale patterns closely related to crime opportunity, whose understanding is fundamental to the planning of preventive actions. Enabling a combined analysis of spatial patterns and their evolution over time is another challenge faced by most crime analysis tools. In this paper, we present Mirante, a crime mapping visualization system that allows spatiotemporal analysis of crime patterns in a street-level scale. In contrast to conventional tools, Mirante builds upon street-level heatmaps and other visualization resources that enable spatial and temporal pattern analysis, uncovering fine-scale crime hotspots, seasonality, and dynamics over time. Mirante has been developed in close collaboration with domain experts, following rigid requirements as scalability and versatile to be implemented in large and medium-sized cities. We demonstrate the usefulness of Mirante throughout case studies run by domain experts using real data sets from cities with different characteristics. With the help of Mirante, the experts were capable of diagnosing how crime evolves in specific regions of the cities while still being able to raise hypotheses about why certain types of crime show up.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127118392","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}
{"title":"SelfieArt: Interactive Multi-Style Transfer for Selfies and Videos with Soft Transitions","authors":"L. N. Alegre, M. M. O. Neto","doi":"10.1109/SIBGRAPI51738.2020.00011","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00011","url":null,"abstract":"We introduce SelfieArt, an interactive technique for performing multi-style transfer for portraits and videos. Our method provides a simple and intuitive way of producing exquisite artistic results that combine multiple styles in a harmonious fashion. It uses face parsing and a multi-style transfer model to apply different styles to the various semantic segments. This is achieved using parameterized soft masks, allowing users to adjust the smoothness of the transitions between stylized regions in real-time. We demonstrate the effectiveness of our solution on a large set of images and videos. Given its flexibility, speed, and quality of results, our solution can be a valuable tool for creative exploration, allowing anyone to transform photographs and drawings in world-class artistic results.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132898212","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}
Hermes H. Ferreira, E. S. Gastal, L. Schnorr, P. Navaux
{"title":"Performance and error analysis of recursive edge-aware Gaussian filters on GPUs","authors":"Hermes H. Ferreira, E. S. Gastal, L. Schnorr, P. Navaux","doi":"10.1109/SIBGRAPI51738.2020.00020","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00020","url":null,"abstract":"We present a schematic for image edge-aware Gaus-sian GPU filtering which has linear complexity on the number of pixels of the image. It allows us to reduce the execution time as we increase the number of Streaming Multiprocessors (SMs) on the GPU. We make use of a domain transformation and use a complex-valued recursive formulation of the Gaussian filter. The algorithm partitions the image in disjoint regions, where we compute in parallel the filtering operations, avoiding communication between regions. Our implementation leads to a real-time solution using a modern GPU. With the RTX 2080 Ti, we achieved an execution time of less than 10 milliseconds for 2 filtering iterations on high-resolution RGB images of dimensions $2048times 2048$.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132021817","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}
{"title":"MyFood: A Food Segmentation and Classification System to Aid Nutritional Monitoring","authors":"Charles N. C. Freitas, F. Cordeiro, V. Macário","doi":"10.1109/SIBGRAPI51738.2020.00039","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00039","url":null,"abstract":"The absence of food monitoring has contributed significantly to the increase in the population’s weight. Due to the lack of time and busy routines, most people do not control and record what is consumed in their diet. Some solutions have been proposed in computer vision to recognize food images, but few are specialized in nutritional monitoring. This work presents the development of an intelligent system that classifies and segments food presented in images to help the automatic monitoring of user diet and nutritional intake. This work shows a comparative study of state-of-the-art methods for image classification and segmentation, applied to food recognition. In our methodology, we compare the FCN, ENet, SegNet, DeepLabV3+, and Mask RCNN algorithms. We build a dataset composed of the most consumed Brazilian food types, containing nine classes and a total of 1250 images. The models were evaluated using the following metrics: Intersection over Union, Sensitivity, Specificity, Balanced Precision, and Positive Predefined Value. We also propose an system integrated into a mobile application that automatically recognizes and estimates the nutrients in a meal, assisting people with better nutritional monitoring. The proposed solution showed better results than the existing ones in the market. The dataset is publicly available at the following link http://doi.ory10.5281/zenodo.4041488.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130396327","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}
{"title":"From explanations to feature selection: assessing SHAP values as feature selection mechanism","authors":"W. E. Marcilio, D. M. Eler","doi":"10.1109/SIBGRAPI51738.2020.00053","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00053","url":null,"abstract":"Explainability has become one of the most discussed topics in machine learning research in recent years, and although a lot of methodologies that try to provide explanations to black-box models have been proposed to address such an issue, little discussion has been made on the pre-processing steps involving the pipeline of development of machine learning solutions, such as feature selection. In this work, we evaluate a game-theoretic approach used to explain the output of any machine learning model, SHAP, as a feature selection mechanism. In the experiments, we show that besides being able to explain the decisions of a model, it achieves better results than three commonly used feature selection algorithms.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125406706","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}
L. C. S. Júnior, Maricéia B. S. Pádua, Leonardo M. Ogusuku, M. Albertini, Renato Pimentel, A. Backes
{"title":"Wild boar Classification by Using a Fusion of Texture Analysis Methods","authors":"L. C. S. Júnior, Maricéia B. S. Pádua, Leonardo M. Ogusuku, M. Albertini, Renato Pimentel, A. Backes","doi":"10.1109/SIBGRAPI51738.2020.00048","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00048","url":null,"abstract":"In several countries the wild boar is an invasive species that remains widespread mainly due its adaptability and uncontrolled reproduction. The monitoring of populations of this pest by camera traps is a promising technique. These devices capture a large variety of animals, including native species that share similarities with the wild boar, therefore the precise identification of the animal species is essential to control the wild boars. To address this problem, we evaluated 18 different texture analysis methods on their ability to discriminate between two native “bush pigs” species of the Brazilian fauna, collared peccary (Pecari tajacu) and white-lipped peccary (Tayassu pecari), and one invasive species, wild boar (Sus scrofa). Results show that species identification is a difficult task due to the similarities among classes, being the distinction between native and invasive species an easier task, and that combining of texture methods improves the accuracy while it diminishes the number of false positives (i.e., native species classified as invasive).","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"273 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121411986","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}
Flavio Fontanella, A. Sá, Moacyr Alvim H. B. da Silva
{"title":"Visual Analysis of Forecasts of Football Match Scores","authors":"Flavio Fontanella, A. Sá, Moacyr Alvim H. B. da Silva","doi":"10.1109/SIBGRAPI51738.2020.00027","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00027","url":null,"abstract":"The present work describes the application of information visualization techniques to better understand the behavior of a forecasting model designed to predict football match scores based on past confrontations. We have run the forecasts during the 2019 season of Brazilian National Championship and, once it was over, we gathered data to observe the performance of the forecasts, based on a set of evaluators we propose. The main contribution of this paper is the introduction of visual devices, each attached to one of these evaluators, designed to enhance interpretation of the performance of forecasting models in general and sports forecasts in particular. We also present other visualizations that help taking notice of particular features of the championship and the forecasts.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130655087","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}
{"title":"Towards a Legion of Virtual Humans: Steering Behaviors and Organic Visualization","authors":"A. Antonitsch, S. Musse, L. H. Figueiredo","doi":"10.1109/SIBGRAPI51738.2020.00013","DOIUrl":"https://doi.org/10.1109/SIBGRAPI51738.2020.00013","url":null,"abstract":"Many studies have been done to control virtual humans and crowds. Navigation, split, merge, and collision avoidance are examples of what has been done in such areas when simulating crowds of individuals. This paper explores the macroscopic crowds’ concept and seeks to solve two main issues: macroscopic crowd control and visualization. In particular, we are interested in providing steering behaviors applied to macroscopic models of crowds, in this case, called Legions of people, which are an abstraction for a vast amount of people. To provide the steering behaviors, we propose a multi-level control of such structures that can represent less or more people, in an emergent way. We also propose a new organic visualization of macroscopic and huge crowds based on blobby models.","PeriodicalId":277256,"journal":{"name":"2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130755881","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}