the proceedings contain 60 papers. the topics discussed include: monocular visual odometry with cyclic estimation;extending the differential image foresting transform to root-based path-cost functions with application...
ISBN:
(纸本)9781538622193
the proceedings contain 60 papers. the topics discussed include: monocular visual odometry with cyclic estimation;extending the differential image foresting transform to root-based path-cost functions with application to superpixel segmentation;building structured lighting applications using low-cost cameras;least-squares morphing of dynamic meshes;repairing non-manifold boundaries of segmented simplicial meshes;activity recognition based on a magnitude-orientation stream network;real-time Brazilian license plate detection and recognition using deep convolutional neural networks;fine-tuning infinity restricted Boltzmann machines;and detecting computer generated images with deep convolutional neural networks.
3D object tracking allows augmented reality applications to add virtual content to the real world in a coherent way. Withthe popularization of RGB-D sensors, new 6 degree-of-freedom tracking techniques that use featu...
详细信息
3D object tracking allows augmented reality applications to add virtual content to the real world in a coherent way. Withthe popularization of RGB-D sensors, new 6 degree-of-freedom tracking techniques that use features extracted from 3D point clouds have been developed, providing more accurate results. this work proposes the use of particle swarm optimization to handle multiple pose hypotheses during top-down tracking from RGB-D images. A fitness function based on 3D point coordinates, color, and normal information was designed, which is able to handle partial object occlusions by applying a threshold to the Euclidean distance between 3D points. the best particle found for a given frame is kept to compose the set of particles of the next swarm and its components are used as an initial pose to define the boundaries of the search space for tracking in the next frame. In addition, we have taken advantage of GPU processing to reduce the running time. Experiments with a publicly available dataset showed that the use of GPU allowed fast object tracking and that the proposed method presents more accurate tracking results in comparison to state-of-the-art optimization-based techniques, especially in situations where objects are partially occluded. (C) 2018 Published by Elsevier Ltd.
the proceedings contain 4 papers. the topics discussed include: introduction to research in magnetic resonance imaging;an overview of max-tree principles, algorithms and applications;tensor fields for multilinear imag...
ISBN:
(纸本)9781509044344
the proceedings contain 4 papers. the topics discussed include: introduction to research in magnetic resonance imaging;an overview of max-tree principles, algorithms and applications;tensor fields for multilinear image representation and statistical learning models applications;and image operator learning and applications.
We present two new methods for creating stereoscopic models displayed over horizontal screens based on image processing. Both methods use Computer Vision: the first is constructed by the modification of the Harley'...
详细信息
ISBN:
(纸本)9781728106045
We present two new methods for creating stereoscopic models displayed over horizontal screens based on image processing. Both methods use Computer Vision: the first is constructed by the modification of the Harley's rectification algorithm, and the second uses the Fundamental theorem of Projective Geometry. Different from the solutions available in the literature, we do not use a calibration pattern, and we do not use unconstrained nonlinear optimization, resulting in simple and efficient algorithms.
the proceedings contain 60 papers. the topics discussed include: a levels-of-precision approach for simulating multiple physics-based soft tissues;retinal vessel segmentation using parallel grayscale skeletonization a...
ISBN:
(纸本)9781509035687
the proceedings contain 60 papers. the topics discussed include: a levels-of-precision approach for simulating multiple physics-based soft tissues;retinal vessel segmentation using parallel grayscale skeletonization algorithm and mathematical morphology;enhancing the visualization of the microvasculature of extrahepatic bile ducts obtained from confocal microscopy images;information theory-based detection of noisy bit planes in medical images;reducing the number of points on the convex hull calculation using the polar space subdivision in E2;an efficient global point cloud descriptor for object recognition and pose estimation;personalized visual simulation and objective validation of low-order aberrations of the human eye;transmitting what matters: task-oriented video composition and compression;on the use of calibration for pedestrian detection in on-board vehicular cameras;a data augmentation methodology to improve age estimation using convolutional neural networks;single sample face recognition from video via stacked supervised auto-encoder;a novel method for detecting the fovea in fundus images of the eye;efficient object recognition using sampling of keypoint triples and keygraph structure;interactive editing of C1-continuous 2D spline deformations using ECLES method;method based on triangulation for sensor deployment on 3D surfaces;using change-sets to achieve a bounded undo and make tutorials in 3D version control systems;a genetically programmable hybrid virtual reconfigurable architecture for image filtering applications;and a data fusion architecture proposal for visually impaired people.
this work presents a new method to transform omnidirectional images based on a combination of Moebius transformations in the complex plane and weighting functions that restrict the action of these mappings to regions ...
详细信息
this work presents a new method to transform omnidirectional images based on a combination of Moebius transformations in the complex plane and weighting functions that restrict the action of these mappings to regions of interest. the transformations are calculated based. on the specification of the image of three points and the weighting functions are designed to achieve specific goals such as local zoom or straight line rectification. Since no optimization or numerical methods are involved, our implementation of the proposed method can be upgraded to reach real-time performance. We provide a user interface and present many results that illustrate the potential of the proposed technique. (C) 2017 Elsevier Ltd. All rights reserved.
A thermal image is obtained by a camera that is sensitive to the thermal variation of the environment. this sensitivity in most cases affects the quality of the image obtained. therefore, it is very important to impro...
详细信息
ISBN:
(纸本)9781538622193
A thermal image is obtained by a camera that is sensitive to the thermal variation of the environment. this sensitivity in most cases affects the quality of the image obtained. therefore, it is very important to improve the quality of thermal images in terms of contrast and details. there are different contrast enhancement techniques which introduce minor distortions, preserving the proper brightness as well as the details of the image. this work presents a top-hat transform for the improvement of grayscale aerial thermal images. the proposed method is based in the use of increasing structuring elements of similar geometry within the fundamental operations of mathematical morphology. the evaluation of the performance of the current proposed method was made with several metrics for images in grayscale. the experimental results show that the proposed method improves several thermal images which were tested, enhancing the contrast, preserving the richness of the details as well as the mean brightness and introducing less distortion in the images. the results show better aerial thermal images with less distortion and adequate brightness, which is very useful for further processes over the images.
Computer graphics techniques for image generation are living an era where, day after day, the quality of produced content is impressing even the more skeptical viewer. Although it is a great advance for industries lik...
详细信息
ISBN:
(纸本)9781538622193
Computer graphics techniques for image generation are living an era where, day after day, the quality of produced content is impressing even the more skeptical viewer. Although it is a great advance for industries like games and movies, it can become a real problem when the application of such techniques is applied for the production of fake images. In this paper we propose a new approach for computer generated images detection using a deep convolutional neural network model based on ResNet-50 and transfer learning concepts. Unlike the state-of-the-art approaches, the proposed method is able to classify images between computer generated or photo generated directly from the raw image data with no need for any pre-processing or hand-crafted feature extraction whatsoever. Experiments on a public dataset comprising 9700 images show an accuracy higher than 94%, which is comparable to the literature reported results, without the drawback of laborious and manual step of specialized features extraction and selection.
Autoimmune diseases are the third cause of mortality in the world. the identification of anti-nuclear antibody (ANA) via Immunofluorescence (IIF) test in human epithelial type-2 cells (HEp-2) is a conventional method ...
详细信息
ISBN:
(纸本)9781538622193
Autoimmune diseases are the third cause of mortality in the world. the identification of anti-nuclear antibody (ANA) via Immunofluorescence (IIF) test in human epithelial type-2 cells (HEp-2) is a conventional method to support the diagnosis of such diseases. In the present work, three popular Convolutional Neural Networks (CNNs) are evaluated for this task: LeNet-5, AlexNet, and GoogLeNet. We also assess the impact of six different pre-processing strategies on the performance of these CNNs. Additionally, data augmentation based on the rotation of the training set images after the pre-processing strategies was evaluated. Our work is the first to consider AlexNet and GoogLeNet models for the proposed analysis and classification of HEp-2 cells images, besides the LeNet-5. Experimental results allow to conclude that neither pre-processing strategies were essential to improve accuracy values of the CNNs. However, when data augmentation is considered, contrast enhancement followed by data centralization is significant in order to achieve good results. Additionally, our results were compared with results from other state-of-art papers. Our best results were achieved by GoogLeNet architecture trained withimages with no preprocessing and no data augmentation, resulting in 98.17% of accuracy, which outperforms the results presented in other works in literature.
the logging and further analysis of borehole images is a major step in the interpretation of geological events. Natural fractures and beddings are features whose identification is commonly performed using acoustic and...
详细信息
ISBN:
(纸本)9781538622193
the logging and further analysis of borehole images is a major step in the interpretation of geological events. Natural fractures and beddings are features whose identification is commonly performed using acoustic and electrical borehole imaging tools. Such identification is a tedious task and is made visually by geologists, who must be experts on classification. the correct identification of planar features, represented as sinusoids into an image projection, depends on the quality of the images. Due to the distortions and noises of the images, known as artifacts, the automatic features detection is not trivial through conventional image processing methods. Since the identification process has to ensure that the marked events are true with minimal inconsistencies, we propose a pioneering approach to improving the quality of the results by applying deep neural networks to confirm or exclude candidate features extracted by a regular Hough transform. this is the first approach in literature to improve the quality of geological auto-detected marks by applying deep learning techniques for borehole images where our implementation is able to exclude most of the false positive marks.
暂无评论