The Experts below are selected from a list of 39162 Experts worldwide ranked by ideXlab platform

Guang-hai Liu - One of the best experts on this subject based on the ideXlab platform.

  • Image Retrieval Using the Fused Perceptual Color Histogram.
    Computational intelligence and neuroscience, 2020
    Co-Authors: Guang-hai Liu, Zhao Wei
    Abstract:

    Extracting visual features for image retrieval by mimicking human cognition remains a challenge. Opponent Color and HSV Color spaces can mimic human visual perception well. In this paper, we improve and extend the CDH method using a multi-stage model to extract and represent an image in a way that mimics human perception. Our main contributions are as follows: (1) a visual feature descriptor is proposed to represent an image. It has the advantages of a histogram-based method and is consistent with visual perception factors such as spatial layout, intensity, edge orientation, and the opponent Colors. (2) We improve the distance formula of CDHs; it can effectively adjust the similarity between images according to two parameters. The proposed method provides efficient performance in similar image retrieval rather than instance retrieval. Experiments with four benchmark datasets demonstrate that the proposed method can describe Color, texture, and spatial features and performs significantly better than the Color Volume histogram, Color difference histogram, local binary pattern histogram, and multi-texton histogram, and some SURF-based approaches.

  • Image Retrieval Based on a Multi-Integration Features Model
    Mathematical Problems in Engineering, 2020
    Co-Authors: Kai Chu, Guang-hai Liu
    Abstract:

    Feature integration theory can be regarded as a perception theory, but the extraction of visual features using such a theory within the CBIR framework is a challenging problem. To address this problem, we extract the Color and edge features based on a multi-integration features model and use these for image retrieval. A novel and highly simple but efficient visual feature descriptor, namely, a multi-integration features histogram, is proposed for image representation and content-based image retrieval. First, a Color image is converted from the RGB to the HSV Color space, and the Color features and Color differences are extracted. Then, the Color differences are calculated to extract the edge features using a set of simple integration processes. Finally, combining the Color, edge, and spatial layout features allows representing the image content. Experiments show that our method produces results comparable to existing and well-known methods on three datasets that contain 25,000 natural images. The performances are significantly better than that of the BOW histogram, local binary pattern histogram, histogram of oriented gradient, and multi-texton histogram, with performances similar to the Color Volume histogram.

  • Content-Based Image Retrieval Using Color Volume Histograms
    International Journal of Pattern Recognition and Artificial Intelligence, 2019
    Co-Authors: Ji-zhao Hua, Guang-hai Liu, Shu-xiang Song
    Abstract:

    Human visual perception has a close relationship with the HSV Color space, which can be represented as a cylinder. The question of how visual features are extracted using such an attribute is impor...

  • Exploiting Color Volume and Color Difference for Salient Region Detection
    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society, 2018
    Co-Authors: Guang-hai Liu, Jingyu Yang
    Abstract:

    Foreground and background cues can assist humans in quickly understanding visual scenes. In computer vision, however, it is difficult to detect salient objects when they touch the image boundary. Hence, detecting salient objects robustly under such circumstances without sacrificing precision and recall can be challenging. In this paper, we propose a novel model for salient region detection, namely, the foreground-center-background (FCB) saliency model. Its main highlights as follows. First, we use regional Color Volume as the foreground, together with perceptually uniform Color differences within regions to detect salient regions. This can highlight salient objects robustly, even when they touched the image boundary, without greatly sacrificing precision and recall. Second, we employ center saliency to detect salient regions together with foreground and background cues, which improves saliency detection performance. Finally, we propose a novel and simple yet efficient method that combines foreground, center, and background saliency. Experimental validation with three well-known benchmark data sets indicates that the FCB model outperforms several state-of-the-art methods in terms of precision, recall, F-measure, and particularly, the mean absolute error. Salient regions are brighter than those of some existing state-of-the-art methods.

  • Salient regions detection using convolutional neural networks and Color Volume
    IOP Conference Series: Materials Science and Engineering, 2018
    Co-Authors: Guang-hai Liu, Yingkun Hou
    Abstract:

    Convolutional neural network is an important technique in machine learning, pattern recognition and image processing. In order to reduce the computational burden and extend the classical LeNet-5 model to the field of saliency detection, we propose a simple and novel computing model based on LeNet-5 network. In the proposed model, hue, saturation and intensity are utilized to extract depth cues, and then we integrate depth cues and Color Volume to saliency detection following the basic structure of the feature integration theory. Experimental results show that the proposed computing model outperforms some existing state-of-the-art methods on MSRA1000 and ECSSD datasets.

Byeong Seok Shin - One of the best experts on this subject based on the ideXlab platform.

  • Interactive high-quality visualization of Color Volume datasets using GPU-based refinements of segmentation data.
    Journal of X-ray science and technology, 2016
    Co-Authors: Byeonghun Lee, Koojoo Kwon, Byeong Seok Shin
    Abstract:

    Data sets containing Colored anatomical images of the human body, such as Visible Human or Visible Korean, show realistic internal organ structures. However, imperfect segmentations of these Color images, which are typically generated manually or semi-automatically, produces poor-quality rendering results. We propose an interactive high-quality visualization method using GPU-based refinements to aid in the study of anatomical structures. In order to represent the boundaries of a region-of-interest (ROI) smoothly, we apply Gaussian filtering to the opacity values of the Color Volume. Morphological grayscale erosion operations are performed to reduce the region size, which is expanded by Gaussian filtering. Pseudo-Coloring and Color blending are also applied to the Color Volume in order to give more informative rendering results. We implement these operations on GPUs to speed up the refinements. As a result, our method delivered high-quality result images with smooth boundaries and provided considerably faster refinements. The speed of these refinements is sufficient to be used with interactive renderings as the ROI changes, especially compared to CPU-based methods. Moreover, the pseudo-Coloring methods used presented anatomical structures clearly.

  • GPU-Based Fast Refinements for High-Quality Color Volume Rendering
    Advances in Parallel and Distributed Computing and Ubiquitous Services, 2016
    Co-Authors: Byeonghun Lee, Koojoo Kwon, Byeong Seok Shin
    Abstract:

    Color Volume datasets of the human body, such as Visible Human or Visible Korean, describe realistic anatomical structures. However, imperfect segmentation of these Color Volume datasets, which are typically generated manually or semi-automatically, produces poor-quality rendering results. We propose an interactive high-quality visualization method using GPU-based refinements to support the study of anatomical structures. To smoothly represent the boundaries of a region-of-interest (ROI), we apply Gaussian filtering to the opacity values of the Color Volume. Morphological grayscale erosion operations are performed to shrink the boundaries, which are expanded by the Gaussian filtering. We implement these operations on GPUs for the sake of fast refinements. As a result, our method delivered high-quality result images with smooth boundaries providing considerably faster refinements, sufficient for interactive renderings as the ROI changes, compared to CPU-based method.

  • Unfolding for Color Volume Datasets Based on Segmented Contours
    IFMBE Proceedings, 2009
    Co-Authors: Yihwa Kang, Sukhyun Lim, Byeong Seok Shin
    Abstract:

    Unfolding is a rendering method to visualize organs at a glance by virtually incising them. Although conventional methods exploit gray-scale Volume datasets such as CT or MR images, we use the Visible Korean Human (VKH) dataset preserving actual Color. VKH dataset consists of anatomical and segmented images. Segmented images store the boundary of organs. In rendering stage, we perform the radial Volume ray casting along with the central path of a target organ. If a ray reaches to nontransparent regions by referring the segmented Volumes, the Color composition is begun. As a result, we can produce high-quality unfolding results. Since our approach can be applied to virtual dissection including actual organs Colors, it is helpful for the anatomy studies.

  • unfolding for Color Volume dataset using the difference of segmented contours
    Journal of Korean Society of Medical Informatics, 2008
    Co-Authors: Yihwa Kang, Byeong Seok Shin, Dong Sun Shin
    Abstract:

    Objective: Unfolding is a rendering method to visualize organs at a glance by virtually incising them. Although conventional methods exploit gray-scale Volume datasets such as CT or MR images, we use the Visible Korean Human dataset preserving actual Color. This can be helpful for the study of anatomical knowledge. Segmented images of Visible Korean Human dataset store the boundary of organs. Since medical experts manually perform the segmentation from anatomical Color images, it is very time-consuming. In general, therefore, some images selectively sampled with interval from entire Color images are segmented. When we generate a segment Volume dataset with the selected images, final results are deteriorated due to lack of segmentation information for missed images. In this paper, we solve this problem by generating intermediate images without performing a manual segmentation. Methods: Firstly, after comparing differences of organ’s contours in between two consecutive segmented images, we represent the differences as a user-defined value in the intermediate images. This procedure is repeated for all pairs of manually segmented images to reconstruct entire Volume data consist of manually segmented images and their intermediate images. In rendering stage, we perform the radial Volume ray casting along with the central path of target organ. If a ray reaches to a region having the user-defined values, we advance over the region without compositions to the boundary of that region. Then the Color composition is begun by performing backtracking, since the advanced region is regarded to the thickness of it. Results: As a result, we can produce high quality unfolding images for the stomach, colon, bronchus, and artery of the Visible Korea Human dataset. Conclusion: Since our approach can be applied to virtual dissection including actual human Colors, it is helpful for the endoscopy and anatomy studies.

  • ICAT - Visualization of segmented Color Volume data using GPU
    Advances in Artificial Reality and Tele-Existence, 2006
    Co-Authors: Koojoo Kwon, Byeong Seok Shin
    Abstract:

    Recently, several Color Volume data such as Visible Human became available for generating a realistic image. These dataset are commonly operated on CPU, however, the rendering time is time-consuming task on CPU. GPU-based Volume rendering method can visualize Color Volume data more easily and quickly because it provides 3D texture including RGB channel. In this paper, we present the GPU-based visualization method of segmented Color Volume data. During the rendering stage, we need two Volume datasets, Color and segmented Volume. However, the segmented Volume requires additional memory. In our method, we use only one 3D texture in GPU. We encode three kinds of values in the 3D texture, Color, segmented index and tagged values. Segmented index means the index value of internal organ. And the tagged values are the information of region of interest. We can visualize fast the Color image of real human body without additional memory.

Sérgio M. C. Nascimento - One of the best experts on this subject based on the ideXlab platform.

  • CCIW - Estimating the Colors of Paintings
    Lecture Notes in Computer Science, 2015
    Co-Authors: Sérgio M. C. Nascimento, João M. M. Linhares, Catarina João, Kinjiro Amano, Cristina Montagner, Maria João Melo, Márcia Vilarigues
    Abstract:

    Observers can adjust the spectrum of illumination on paintings for optimal viewing experience. But can they adjust the Colors of paintings for the best visual impression? In an experiment carried out on a calibrated Color moni- tor images of four abstract paintings obtained from hyperspectral data were shown to observers that were unfamiliar with the paintings. The Color Volume of the images could be manipulated by rotating the Volume around the axis through the average (a*, b*) point for each painting in CIELAB Color space. The task of the observers was to adjust the angle of rotation to produce the best subjective impression from the paintings. It was found that the distribution of angles selected for data pooled across paintings and observers could be de- scribed by a Gaussian function centered at 10o, i.e. very close to the original Colors of the paintings. This result suggest that painters are able to predict well what compositions of Colors observers prefer.This work was supported by the Centro de Física of Minho University, by FEDER through the COMPETE Program and by the Portuguese Foundation for Science and Technology (FCT) in the framework of the projects PTDC/MHC-PCN/4731/2012 and PTDC/EAT- EAT/113612/2009, by and the COST-Action TD1201, Colour and Space in Cultural Heritage (COSCH) through Short Term Scientific Missions (STSM): “Hyperspectral imaging on historical manuscripts and natural scenes” (COST-STSM-TD1201- 010813-032699, 2013). Cristina Mon- tagner was supported by the grant SFRH/BD/66488/2009

  • Effects of high-Color-discrimination capability spectra on Color-deficient vision
    Journal of the Optical Society of America. A Optics image science and vision, 2013
    Co-Authors: Esther Perales, João M. M. Linhares, Osamu Masuda, Francisco M. Martínez-verdú, Sérgio M. C. Nascimento
    Abstract:

    Light sources with three spectral bands in specific spectral positions are known to have high-Color-discrimination capability. W. A. Thornton hypothesized that they may also enhance Color discrimination for Color-deficient observers. This hypothesis was tested here by comparing the Rosch–MacAdam Color Volume for Color-deficient observers rendered by three of these singular spectra, two reported previously and one derived in this paper by maximization of the Rosch–MacAdam Color solid. It was found that all illuminants tested enhance discriminability for deuteranomalous observers, but their impact on other congenital deficiencies was variable. The best illuminant was the one derived here, as it was clearly advantageous for the two red–green anomalies and for tritanopes and almost neutral for red–green dichromats. We conclude that three-band spectra with high-Color-discrimination capability for normal observers do not necessarily produce comparable enhancements for Color-deficient observers, but suitable spectral optimization clearly enhances the vision of the Color deficient.

  • Color diversity index: the effect of chromatic adaptation
    International Conference on Applications of Optics and Photonics, 2011
    Co-Authors: João M. M. Linhares, Sérgio M. C. Nascimento
    Abstract:

    Common descriptors of light quality fail to predict the chromatic diversity produced by the same illuminant in different contexts. The aim of this paper was to study the influence of the chromatic adaptation in the context of the development of the Color diversity index, a new index capable of predicting illuminant-induced variations in several types of images. The spectral reflectance obtained from hyperspectral images of natural, indoor and artistic paintings, and the spectral reflectance of 1264 Munsell surfaces were converted into the CIELAB Color space for each of the 55 CIE illuminants and 5 light sources tested. The influence of the CAT02 chromatic adaptation was estimated for each illuminant and for each scene. The CIELAB Volume was estimated by the convex hull method and the number of discernible Colors was estimated by segmenting the CIELAB Color Volume into unitary cubes and by counting the number of non-empty cubes. High correlation was found between the CIELAB Volume occupied by the Munsell surfaces and the number of discernible Colors and the CILEAB Color Volume of the Colors in all images analyzed. The effects of the chromatic adaptation were marginal and did not change the overall result. These results indicate that the efficiency of the new illuminant chromatic diversity index is not influenced by chromatic adaptation

  • CGIV/MCS - Chromatic diversity index - an approach based on natural scenes
    2010
    Co-Authors: João M. M. Linhares, Paulo Daniel Araújo Pinto, Sérgio M. C. Nascimento
    Abstract:

    Common descriptors of light quality fail to predict the chromatic diversity produced by the same illuminant in different contexts such as images of natural scenes. The aim of this paper was to introduce a new index, capable of predicting illuminantinduced variations in the chromatic diversity off natural scenes. The spectral reflectance of each pixel of 50 images of natural scenes obtained using a hyperspectral imaging and the spectral reflectance of 1264 Munsell surfaces were converted into the CIELAB Color space for each of the 55 illuminants and 5 light sources. The CIELAB Volume was estimated by the convex hull method. The number of discernible Colors was estimated by segmenting the CIELAB Color Volume into unitary cubes and by counting the number of non-empty cubes. High correlation was found between the CIELAB Volume occupied by the Munsell surfaces, the number of discernible Colors and CILEAB Color Volume of the Colors of natural scenes. These results seem to indicate that a new illuminant chromatic diversity index based on natural scenes could be defined using the CIELAB Volume of the Munsell surfaces.

  • The number of discernible Colors in natural scenes
    Journal of the Optical Society of America. A Optics image science and vision, 2008
    Co-Authors: João M. M. Linhares, Paulo Daniel Araújo Pinto, Sérgio M. C. Nascimento
    Abstract:

    The number of Colors discernible by normal trichromats has been estimated for the idealized object-Color solid. How well these estimates apply to natural scenes is an open question, as it is unknown how much their Colors approach the theoretical limits. The aim of this work was to estimate the number of discernible Colors based on a database of hyperspectral images of 50 natural scenes. The Color Volume of each scene was computed in the CIELAB Color space and was analyzed using the CIEDE2000 Color-difference formula. It was found that the Color Volume of the set of natural scenes was about 30% of the theoretical maximum for the full object-Color solid, and it corresponded to a number of about 2.3 million discernible Colors. Moreover, when the lightness dimension was ignored, only about 26,000 (1%) could be perceived as different Colors. These results suggest that natural stimuli may be more constrained than expected from the analysis of the theoretical limits.

Robert F Labadie - One of the best experts on this subject based on the ideXlab platform.

C G Kealey - One of the best experts on this subject based on the ideXlab platform.

  • genetic parameter estimates for scrotal circumference and semen characteristics of line 1 hereford bulls
    Journal of Animal Science, 2006
    Co-Authors: C G Kealey, M D Macneil, M W Tess, T W Geary, R A Bellows
    Abstract:

    The objectives of this study were to estimate heritability for scrotal circumference (SC) and semen traits and their genetic correlations (r[subscript g]) with birth weight (BRW). Semen traits were recorded for Line 1 Hereford bulls (n = 841), born in 1963 or from 1967 to 2000, that were selected for use at Fort Keogh (Miles City, MT) or for sale. Semen was collected by electroejaculation when bulls were a mean age of 446 d. Phenotypes were BRW, SC, ejaculate Volume, subjective scores for ejaculate Color, swirl, sperm concentration and motility, and percentages of sperm classified as normal and live or having abnormal heads, abnormal midpieces, proximal cytoplasmic droplets (primary abnormalities), bent tails, coiled tails, or distal cytoplasmic droplets (secondary abnormalities). Percentages of primary and secondary also were calculated. Data were analyzed using multiple-trait derivative-free REML. Models included fixed effects for contemporary group, age of dam, age of bull, inbreeding of the bull and his dam, and random animal and residual effects. Random maternal and permanent maternal environmental effects were also included in the model for BRW. Estimates of heritability for BRW, SC, semen Color, Volume, concentration, swirl, motility, and percentages of normal, live, abnormal heads, abnormal midpieces, proximal cytoplasmic droplets, bent tails, coiled tails, distal cytoplasmic droplets, and primary and secondary abnormalities were 0.34, 0.57, 0.15, 0.09, 0.16, 0.21, 0.22, 0.35, 0.22, 0.00 0.16, 0.37, 0.00 0.34 0.00, 0.30, and 0.33, respectively. Estimates of r[subscript g] for SC with Color, Volume, concentration, swirl, motility, and percentages of live, normal, and primary and secondary abnormalities were 0.73, 0.20, 0.77, 0.40, 0.34, 0.63, 0.33, -0.36, and -0.45, respectively. Estimates of r[subscript g] for BRW with SC, Color, Volume, concentration, swirl, motility, and percentages live, normal, and primary and secondary abnormalities were 0.28, 0.60, 0.08, 0.58, 0.44, 0.21, 0.34, 0.20, -0.02, and -0.16, respectively. If selection pressure was applied to increase SC, all of the phenotypes evaluated would be expected to improve. Predicted correlated responses in semen characteristics per genetic SD of selection applied to SC were 0.87 genetic SD or less. If selection pressure was applied to reduce BRW, the correlated responses would generally be smaller but antagonistic to improving all of the phenotypes evaluated. Predicted correlated responses in SC and semen characteristics per genetic SD of selection applied to BRW were less than 0.35 genetic SD.

  • Estimation of genetic parameters of yearling scrotal circumference and semen characteristics in line 1 hereford bulls
    2004
    Co-Authors: C G Kealey
    Abstract:

    Objectives of this research were to estimate heritabilities of scrotal circumference and semen traits, and genetic correlations among these traits and birth weight. Line 1 Hereford bulls (n = 841), born in 1963 or from 1967 to 2000, were selected for use by USDA-ARS at Miles City, Montana or for sale. Semen was collected by electro-ejaculation when the bulls were approximately one year of age (mean = 446d) and all samples were evaluated by one person. Traits analyzed were scrotal circumference, Color, Volume, concentration, swirl, motility, and percents normal, live, abnormal heads, abnormal mid-pieces, proximal distal droplets, bent tails, coiled tails, distal proximal droplets, and primary and secondary abnormalities. Data were analyzed using MTDF-REML. Models included fixed effects for contemporary group, age of dam, age of bull at evaluation, inbreeding of the bull and his dam, and random animal, maternal, permanent maternal environmental, and residual effects. Heritability estimates for scrotal circumference, Color, Volume, concentration, swirl, motility, and percents normal, live, abnormal mid-pieces, proximal distal droplets, coiled tails, and primary and secondary abnormalities were 0.57, 0.15, 0.09, 0.16, 0.21, 0.22, 0.23, 0.34, 0.17, 0.34, 0.30 0.34, and 0.29, respectively. Estimates of genetic correlations between birth weight and scrotal circumference, Color, Volume, concentration, swirl, motility, and percents normal, live, abnormal mid-pieces, proximal distal droplets, coiled tails, and primary and secondary abnormalities were 0.36, 0.60, 0.07, 0.58, 0.44, 0.21, 0.20, 0.34, -0.03, -0.52, -0.20, -0.25, and 0.05, respectively. The moderate estimates of heritability for many of the traits indicate potential for favorable selection response. Positive genetic correlations between birth weight and majority of the traits suggest selection to reduce birth weight may compromise semen traits. However, for most traits the expected correlated responses are small. Desirable genetic correlations among scrotal circumference and semen traits suggest selection for one trait would not compromise the other semen traits. Expected correlated responses in semen traits to selection for increased scrotal circumference appear favorable.