检索结果-内蒙古大学图书馆

Foreword to the Special Section on sibgrapi 2017

COMPUTERS & graphics-UK 2017年 68卷 A1-A6页

作者： Torchelsen, Rafael Panozzo, Daniele Rua Gomes Carneiro 1 BR-96010610 Pelotas RS Brazil NYU Courant Inst Math Sci 60 5th Ave5th Floor New York NY 10011 USA

the conference on graphics, patterns and images is an international venue annually promoted by the Brazilian Computer Society (SBC). the proceedings of the event have been published by IEEE Computer Society Press since 1997. In addition, sibgrapi 2017 includes two pre-conference events whose proceedings are published as Special Sections of the Elsevier Computers & graphics Journal and of the IEEE Geoscience and Remote Sensing Letters, and a post-conference publication as a Special Issue of the Elsevier Journal of Visual Communication and Image Representation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Method for Detecting the Fovea in Fundus images of the Eye

A Novel Method for Detecting the Fovea in Fundus Images of t...

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Pedro De Moura Garcia Diego Luis Kreutz Daniel Welfer Universidade Federal do Pampa Bage RS BR Universidade Federal do Pampa (UNIPAMPA) Alegrete Rio Grande do Sul Brasil Universidade Federal de Santa Maria (UFSM) Santa Maria Brasil

ISBN: (纸本)9781509035694;9781509035687

In this paper we propose a novel method for finding the fovea in colored images of the eye fundus. We use an image correlation coefficient to find the fovea, which is calculated from a set of template images of the fovea. the results, using the DIARETDB1 database, indicate that our method detects the fovea region with an accuracy of 82,02%.

关键词： Optical imaging Blood vessels Biomedical imaging Diabetes Correlation Visualization Retina

来源：评论

学校读者我要写书评

暂无评论

Improving Mass Detection in Mammography images: A Study of Weakly Supervised Learning and Class Activation Map Methods

Improving Mass Detection in Mammography Images: A Study of W...

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Vicente Sampaio Filipe R. Cordeiro Department of Computing Universidade Federal Rural de Pernambuco Recife Brazil Department of Computing Visual Computing Lab Universidade Federal Rural de Pernambuco Recife Brazil

In recent years, weakly supervised models have aided in mass detection using mammography images, decreasing the need for pixel-level annotations. However, most existing models in the literature rely on Class Activation Maps (CAM) as the activation method, overlooking the potential benefits of exploring other activation techniques. this work presents a study that explores and compares different activation maps in conjunction with state-of-the-art methods for weakly supervised training in mammography images. Specifically, we investigate CAM, GradCAM, GradCAM++, XGradCAM, and LayerCAM methods within the framework of the GMIC model for mass detection in mammography images. the evaluation is conducted on the VinDr-Mammo dataset, utilizing the metrics Accuracy, True Positive Rate (TPR), False Negative Rate (FNR), and False Positive Per Image (FPPI). Results show that using different strategies of activation maps during training and test stages leads to an improvement of the model. With this strategy, we improve the results of the GMIC method, decreasing the FPPI value and increasing TPR.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Context-Aware Approach for Filtering Empty images in Camera Trap Data Using Siamese Network

A Context-Aware Approach for Filtering Empty Images in Camer...

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Luiz Alencar Fagner Cunha Eulanda M. dos Santos ICOMP - Institute of Computing Federal University of Amazonas Manaus Amazonas

this paper presents a method based on a Siamese convolutional neural network (CNN) for filtering empty images captured by camera traps. the proposed method takes into account information of the environment surrounding the camera by comparing captured images with empty reference images obtained regularly from the same capture locations. Reference images are expected to highlight local vegetation features such as rocks, mountains and lakes. By calculating the similarity between the two images, the Siamese network determines whether or not the captured image contains an animal. We present a protocol to provide image pairs to train the models, as well as the data augmentation techniques employed to enhance the training procedure. three different CNN models are used as backbones for the Siamese network: MobileNetV2, ResNet50, and EfficientNetBO. In addition, experiments are conducted on three popular camera trap datasets: Snapshot Serengeti, Caltech and WCS. the results demonstrate the effectiveness of the proposed method due to the information of the capture location considered, and its potential for wildlife monitoring applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Revisiting Age Estimation on Panoramic Dental images

Revisiting Age Estimation on Panoramic Dental Images

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Julian Liang Patricia Cury Luciano Oliveira Intelligent Vision Research Lab Institute of Computing University Federal of Bahia Department of Periodontics Federal University of Bahia

Forensic dentistry has traditionally relied on bone or dental indicators, primarily utilizing dental radiographs, for age estimation. However, limited research has been conducted on automatic age estimation on panoramic images, needing a reeval-uation of the existing methodologies to assess the performance of computer-based methods. this study proposes to revisit the analysis of age estimation methods using panoramic dental radio-graphs. We have curated the largest publicly available dataset of panoramic dental images, encompassing diverse dental conditions and age ranges. Specifically, our study focuses on evaluating three distinct classes of deep-learning architectures: ViT, ConvNeXt-V2, and EfficientNets, employing a comprehensive to assess their performances that better favor reproducibility. By comparing our approach with existing studies in the literature, we offer valuable insights for forensic investigations in the field of age estimation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learning to Classify Seismic images with Deep Optimum-Path Forest

Learning to Classify Seismic Images with Deep Optimum-Path F...

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Luis Afonso Alexandre Vidal Michelle Kuroda Alexandre Xavier Falcão João P. Papa Department of Computing Federal University of São Carlos São Carlos SP Brazil Institute of Geology University of Campinas Campinas SP Brazil Institute of Computing University of Campinas Campinas SP Brazil Department of Computing São Paulo State University Bauru SP Brazil

Due to the lack of labeled information, clustering techniques have been paramount in the last years once more. In this paper, inspired by the deep learning phenomenon, we presented a multi-scale approach to obtain more refined cluster representations of the Optimum-Path Forest (OPF) classifier, which has obtained promising results in a number of works in the literature. Here, we propose to fill a gap in OPF-based works by using a deep-driven representation of the feature space. Additionally, we validated the work in the context of high resolution seismic images aiming at petroleum exploration, as well as in general-purpose applications. Quantitative and qualitative analysis are conducted in order to assess the robustness of the proposed approach.

关键词： Prototypes Context Optimization Machine learning Image resolution Petroleum Labeling

来源：评论

学校读者我要写书评

暂无评论

Automatic Detection of 2D Human Postures Based on Single images

Automatic Detection of 2D Human Postures Based on Single Ima...

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Humberto Souto Soraia Musse Graduate Programme in Computer Science Virtual Human Laboratory Pontifícia Universidade Católica do Rio Grande do Sul Porto Alegre Brazil

Estimating human pose in static images is a challenging task due to the high dimensional state space, presence of image clutter and ambiguities of image observations. In this paper we propose a method to automatically detect human poses in a single image, based on a 2D model combined with anthropometric data. Furthermore, we use artificial neural networks to detect high level information about the human posture. Experimental results showed that the proposed technique performs well in non trivial images.

关键词： Humans Image segmentation Training Joints Legged locomotion Face Skin

来源：评论

学校读者我要写书评

暂无评论

Fast Inverse Halftoning Algorithm for Ordered Dithered images

Fast Inverse Halftoning Algorithm for Ordered Dithered Image...

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Pedro Freitas Mylene Farias Aleteia Araujo Department of Computer Science University of Brasília (UnB) Brasília Brazil

In this paper, we present a simple and fast inverse half toning algorithm, targeted at reconstructing half toned images generated using dispersed-dot ordered dithering algorithms. the proposed algorithm uses a simple set of linear filters combined with a stochastic model in order to predict the best intensity values for the binary image pixels. the algorithm produces images with a better perceptual quality than the available algorithms in the literature, preserving most of the fine details of the original gray-level image. It has a high performance, which can be further improved with the use of parallelization techniques.

关键词： Image reconstruction Prediction algorithms PSNR Measurement Clustering algorithms Hurricanes Signal processing algorithms

来源：评论

学校读者我要写书评

暂无评论

Does the Input Image Spatial Resolution Generate Different Synthetic images? A Comparative Study of Facial Expression Synthesis Performance

Does the Input Image Spatial Resolution Generate Different S...

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Rafael Luiz Testa Ariane Machado-Lima Fátima L. S. Nunes School of Arts Sciences and Humanities University of São Paulo São Paulo Brazil

Facial expression synthesis has gained significant attention in the image synthesis field. Generative Adversarial Network (GAN) models have recently gained popularity due to the high-quality synthetic images they produce. However, these models require complex network architectures that can take days to train, even with high-performance graphics Processing Units (GPUs). Many efforts have been made to accelerate and compress such models, but little attention has been paid to the resolution of the images. this study aims to assess the impact of input/output spatial resolution on the resources needed for training a facial expression synthesis model, as well as on the quality of the results. Our results indicate that the produced images and videos had similar quality results measured through objective measures for the spatial resolution of 128 × 128, 256 × 256, and 480 × 480. Furthermore, we found that lower-resolution images could significantly reduce the time required to generate new facial expressions without compromising quality, as measured by objective measures.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancement of Underwater images in Low-to-High Turbidity Rivers

Enhancement of Underwater Images in Low-to-High Turbidity Ri...

引用

Brazilian Symposium on Computer graphics and Image Processing (sibgrapi)

作者： Daily Daleno de O. Rodrigues Wagner F. de Barros José P. de Queiroz-Neto Anderson G. Fontoura José Reginaldo H. Carvalho Instituto Federal do Amazonas (IFAM) Manaus Brazil Instituto Federal de Educacao Ciencia e Tecnologia do Norte de Minas Gerais Montes Claros MG BR Instituto Federal de Educacao Ciencia e Tecnologia do Amazonas Manaus AM BR Instituto de Computação (IComp) Universidade Federal do Amazonas (UFAM) Manaus Brazil

ISBN: (纸本)9781509035694;9781509035687

the demand for efficient enhancement methods of underwater images of the rivers in the Amazon region is increasing. However, most of those in the region present moderate turbidity and low luminosity. this work aims to improve these images by non-linear filtering techniques, which promote the minimization of light interaction characteristics with the environment, loss of the contrast and colors. the proposed method is compared with two others techniques that requires a unique image as input. the results of the proposed method is promising, with better visual quality considering a wide range of experiments with simulation data and real outdoor scenes.

关键词： Image color analysis Lighting Cameras Mathematical model Rivers Histograms Frequency-domain analysis

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：