Papers by Dalcimar Casanova
Artificial dataset for clustering algorithms(Complete)
This file contains a number of randomly generated datasets. The properties of each dataset are in... more This file contains a number of randomly generated datasets. The properties of each dataset are indicated in the name of each respective file: 'C' indicates the number of classes, 'F' indicates the number of features, 'Ne' indicates the number of objects contained in each class, 'A' is related to the average separation between classes and 'R' is an index used to differentiate distinct random trials. So, for instance, the file C2F10N2Ne5A1.2R0 is a dataset containing 2 classes, 10 features, 5 objects for each class and having a typical separation between classes of 1.2. The methodology used for generating the datasets is described in the accompanying reference.<br>
INFOCOMP Journal of Computer Science, 2015
A aproximacao poligonal de contornos e uma representacao simplificada da sua essencia utilizando ... more A aproximacao poligonal de contornos e uma representacao simplificada da sua essencia utilizando o menor numero possivel de segmentos poligonais. Neste artigo e apresentado um novo metodo de estimativa da aproximacao poligonal baseado na teoria das Redes Complexas. O metodo realiza inicialmente a modelagem da curva em uma rede regular e a transforma em uma rede complexa Pequeno-Mundo. Por meio da analise das propriedades desta rede, em especial o caminho geodesico, e calculada a aproximacao poligonal. O artigo apresenta experimentos realizados com contornos, que demonstram as principais caracteristicas do metodo bem como sua funcionalidade. O metodo proposto e comparado com a aproximacao tradicional baseada no calculo da curvatura.
Anais do Computer on the Beach, 2018
Processos de manufatura que envolvem classificacao de itens por meio de inspecao visual sao frequ... more Processos de manufatura que envolvem classificacao de itens por meio de inspecao visual sao frequentemente realizados por especialistas humanos, que estao sujeitos a falhas por cansaco e desatencao. Este trabalho propoem a utilizacao de uma tecnica recente de analise de textura baseada em fractais para classificacao de itens de manufatura. Para validacao da proposta, realizou-se um estudo de caso envolvendo a classificacao de 22 especies de madeira. Os resultados mostraram uma taxa de classificacao correta de 99,86% utilizando validacao cruzada por 10-fold e classificador LDA.
Polygonal approximation of a contour is a simplified representation of its essence using the smal... more Polygonal approximation of a contour is a simplified representation of its essence using the small number of possible polygonal segments. In this article a novel method of estimating a polygonal approximation based on Complex Networks theory is presented. The method performs initially the modeling of the curve in a regular network and after transforms this network in a Small-World Complex Network. By analysis of the network properties, in special, the geodesic path, it is calculated the polygonal approximation. The article presents the experiments performed on contours, which demonstrate the main characteristics of the method, as also its functionality. The proposed method is compared with traditional approximation based on curvature.
The main purpose of this project is to encourage discussion about the use of technologies for the... more The main purpose of this project is to encourage discussion about the use of technologies for the generation of information in the media context. Therefore, discussions will be held on current and highly relevant topics for the areas of Information and Communication, also considering its interdisciplinarity with Marketing, Journalism, Public Relations, Library Science, Business Management, and Computer Science. Discussions of great scientific and social relevance are addressed, such as disinformation, public policies, data opening, among others. The 1st Workshop on Media, Information and Data Science (WMIDS 2020) proved to be an environment of opportunity for great debates, with the participation of researchers of the highest level, united in an effort for the progress of science in Brazil. Throughout the event, it was possible to note the ability of science and its researchers to reinvent themselves and adapt at a time when social isolation, due to the COVID-19 pandemic, could become an obstacle or a counterpoint to the development of knowledge. All of this is the result of the joint effort of people who believe that knowledge is one of the main tools for human development, in all its specificities.
Resumo—The use of computer systems in sports has increased significantly in the last decade. Cons... more Resumo—The use of computer systems in sports has increased significantly in the last decade. Consequently, systems have been developed to help each athlete or team quantify their performance, such as distances traveled, speeds attained, and positions where each athlete was on the court or field. In this work, a method based on computer vision is proposed to analyse futsal matches. Videos were acquired using a single camera with a wide-angle lens, which facilitates the installation and calibration process in different matches and arenas. The approach is illustrated through video recordings of Pato Futsal team, from which the athletes were detected, their positions projected from pixels to real world coordinates and their trajectories estimated. The generated data visualization aims to help coaches in their physical and tactical analysis.
Analogy-based Effort Estimation: A Systematic Mapping of Literature
Many research initiatives have been developed in the field of Software Engineering, including the... more Many research initiatives have been developed in the field of Software Engineering, including the area of ​​software estimation. Software effort estimation techniques based on analogy are applied from historical data of projects, obtained in the early stages of software development. In this context, this paper presents a systematic mapping of the literature, aim to elicit the state of the art on analogy-based software effort estimation techniques , indicating challenges and research opportunities. The mapping was done for the period from 2007 to 2017 and was conducted separately for each of the selected sources. The articles found were reviewed according to previously established research and selection criteria, according to the study objectives. Note that the model of estimation by analogy has received more attention and is presented as a promising and feasible technique in relation to the others. The techniques of Adaptive Neuro-fuzzy Inferences (ANFIS), Collaborative Filtering...
Existem vários mĂ©todos aplicáveis Ă análise de sinais ultrassĂ´nicos que possibilitam a caracteriz... more Existem vários mĂ©todos aplicáveis Ă análise de sinais ultrassĂ´nicos que possibilitam a caracterização e localização de defeitos em objetos. Dentre tais mĂ©todos, destacam-se os que usam reconstrução de imagens, pois facilitam a visualização dos resultados. Nesse contexto, diversos algoritmos, com abordagens diferentes, objetivam melhorar a qualidade das imagens reconstruĂdas . Uma abordagem proposta em muitas aplicações e que começou a ser introduzida na reconstrução de imagens de ultrassom Ă© a de aprendizagem profunda (deep learning). Este artigo apresenta um mĂ©todo de estimação das coordenadas para a localização geomĂ©trica de defeitos dentro de uma regiĂŁo de interesse, a partir de uma imagem reconstruĂda por meio de uma rede neural profunda (Deep Neural Network - DNN). Para tanto, dados simulados com refletores acĂşsticos infinitesimais (representando pequenos defeitos) foram usados para treinar e validar uma rede neural do tipo autoencoder convolucional. Os resultados demonstram qu...
Flexible control of Discrete Event Systems using environment simulation and Reinforcement Learning
Applied Soft Computing
Intervening Factors in Pavement Roughness Assessment with Smartphones: Quantifying the Effects and Proposing Mitigation
Journal of Transportation Engineering, Part B: Pavements
AbstractResearchers have linked pavements roughness to acceleration signals provided by smartphon... more AbstractResearchers have linked pavements roughness to acceleration signals provided by smartphones, due to their simple handling and low cost, which might facilitate continuous data collection tha...
Proposal and study of statistical features for string similarity computation and classification
International Journal of Data Mining, Modelling and Management
Adaptations of features commonly applied in the field of visual computing, co-occurrence matrix (... more Adaptations of features commonly applied in the field of visual computing, co-occurrence matrix (COM) and run-length matrix (RLM), are proposed for the similarity computation of strings in general (words, phrases, codes and texts). The proposed features are not sensitive to language related information. These are purely statistical and can be used in any context with any language or grammatical structure. Other statistical measures that are commonly employed in the field such as longest common subsequence, maximal consecutive longest common subsequence, mutual information and edit distances are evaluated and compared. In the first synthetic set of experiments, the COM and RLM features outperform the remaining state-of-the-art statistical features. In 3 out of 4 cases, the RLM and COM features were statistically more significant than the second best group based on distances (P-value < 0.001). When it comes to a real text plagiarism dataset, the RLM features obtained the best results.
Texture analysis using fractal descriptors estimated by the mutual interference of color channels
Information Sciences, 2016
Fractal descriptors are used to describe color textures.Spatial structure and color distribution ... more Fractal descriptors are used to describe color textures.Spatial structure and color distribution are analyzed without separating the image into channels.The volumes of the dilated structure express the level of details at each scale based on the intensity/color distribution.The descriptors combine the efficiency of fractals in describing complex structures with the richness of color analysis.The method outperformed other approaches whose effectiveness is widely attested in studies on texture analysis methods. This work presents a method for color texture analysis based on fractal geometry. The method is based on its predecessor 4 and consists of mapping each color channel onto a surface and dilating such surface by spheres with a variable radius. The descriptors are obtained from the relation between the volumes of the dilated surfaces and the dilation radii. The dilation process creates a mutual interference among the color channels. The proposed descriptors measure the degree of such interference as well as the complexity of pixel intensity arrangements. This combination provides a robust and precise texture description. The efficiency of the method is assessed in a classification task of well-known texture data sets and the results demonstrate that it outperforms the best approaches described in the literature.
The leaves are one of the most important main sources used for plant identication. Because of thi... more The leaves are one of the most important main sources used for plant identication. Because of this the ImageCLEF 2011 proposed a challenge based on leaf analysis for plant identication. This paper reports the experiment results of the IFSC/USP team in participating of this task. The main goal is investigate the performance of Complex Network method for feature extraction and classication of plant species. The achieved results are promising and can help the botanists in the future.
Automatic Classification of Multiple Objects in Automotive Assembly Line
2019 24th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)
In automotive manufacturing, assembly tasks depend on the correct identification and selection of... more In automotive manufacturing, assembly tasks depend on the correct identification and selection of workpieces, that may belong to more than one type of vehicle to be produced on the same manufacturing line. Usually, those tasks are conducted essentially by humans, which recently have been complemented by artificial perception provided by computer vision systems (CVSs). Despite their relevance, the accuracy of CVSs depend mostly on the environment control, providing appropriate lighting, enclosure and stops for images to be collected. This makes the solution expensive and overrides part of its benefits. This paper proposes a deep learning-based alternative to detect and classify multiple objects in automotive assembly line. Results show that the detection system has acceptable accuracy, it does not require any interventions on the production line, and it keeps its cycle time. The approach is illustrated by two examples of object detection over a real automotive assembly plant.
Estimating and tuning adaptive action plans for the control of smart interconnected poultry houses
Expert Systems with Applications
Deep Learning Models for Visual Inspection on Automotive Assembling Line
International Journal of Advanced Engineering Research and Science
In automotive manufacturing, assembly tasks depend on visual inspection to ensure product and pro... more In automotive manufacturing, assembly tasks depend on visual inspection to ensure product and process quality, for example, scratches identification on machined surfaces or correct part identification and selection, which may belong to more than one type of vehicle to be produced on the same manufacturing line. Typically, these tasks are essentially human-led, which have recently been supplemented by the artificial perception provided by computer vision systems (CVSs). Despite their relevance, the accuracy of CVSs depends mostly on the environment control, providing appropriate lighting, enclosure, and stops for images to be collected. These problems makes the solution expensive and overrides part of its benefits, mainly when it interferes with the operating cycle time. Thus, this paper proposes the use of deep learning-based methodologies to assist in the visual inspection task, generating little influence on the original manufacturing environment and exploring it as an end-to-end tool to ease CVSs setup. The approach has illustrated by four proofs of concept in a real automotive assembly line based on models for object detection, semantic segmentation, and anomaly detection.
Assessing classification complexity of datasets using fractals
International Journal of Computational Science and Engineering
A framework for modelling, control and supervision of poultry farming
International Journal of Production Research
Generating action plans for poultry management using artificial neural networks
Computers and Electronics in Agriculture
A Gaussian pyramid approach to Bouligand–Minkowski fractal descriptors
Information Sciences
Uploads
Papers by Dalcimar Casanova