Search results for "Computer Science::Computer Vision and Pattern Recognition"

showing 10 items of 193 documents

Optical flow estimation from multichannel spherical image decomposition

2011

The problem of optical flow estimation is largely discussed in computer vision domain for perspective images. It was also proven that, in terms of optical flow analysis from these images, we have difficulty distinguishing between some motion fields obtained with little camera motion. The omnidirectional cameras provided images with large filed of view. These images contain global information about motion and allow to remove the ambiguity present in perspective case. Nevertheless, these images contain significant radial distortions that is necessary to take into account when treating these images to estimate the motion. In this paper, we shall describe new way to compute efficient optical fl…

business.industryPerspective (graphical)ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONOptical flowPhysics::OpticsMotion (geometry)Spherical imageImage (mathematics)WaveletComputer Science::Computer Vision and Pattern RecognitionSignal ProcessingComputer visionComputer Vision and Pattern RecognitionArtificial intelligenceDecomposition method (constraint satisfaction)businessOmnidirectional antennaSoftwareMathematicsComputer Vision and Image Understanding
researchProduct

A statistical model for magnitudes and angles of wavelet frame coefficients and its application to texture retrieval

2014

Abstract This paper presents a texture descriptor based on wavelet frame transforms. At each position in the image, and for each resolution level, we consider both vertical and horizontal wavelet detail coefficients as the components of a bivariate random vector. The magnitudes and angles of these vectors are computed. At each level the empirical histogram of magnitudes is modeled by a Generalized Gamma distribution, and the empirical histogram of angles is modeled by a different version of the von Mises distribution that accounts for histograms with 2 modes. Each texture is characterized by few parameters. A new distance is presented (based on the Kullback–Leibler divergence) that allows g…

business.industryTexture DescriptorGeneralized gamma distributionComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONPattern recognitionWaveletImage textureArtificial IntelligenceComputer Science::Computer Vision and Pattern RecognitionHistogramSignal Processingvon Mises distributionComputer Vision and Pattern RecognitionArtificial intelligenceDivergence (statistics)businessImage retrievalSoftwareMathematicsPattern Recognition
researchProduct

Curvelet-based method for orientation estimation of particles from optical images

2014

A method based on the curvelet transform is introduced to estimate the orientation distribution from two-dimensional images of small anisotropic particles. Orientation of fibers in paper is considered as a particular application of the method. Theoretical aspects of the suitability of this method are discussed and its efficiency is demonstrated with simulated and real images of fibrous systems. Comparison is made with two traditionally used methods of orientation analysis, and the new curvelet-based method is shown to perform better than these tradi- tional methods. peerReviewed

curveletmultiscaleComputer Science::Computer Vision and Pattern Recognitionanisotropicorientationfiber
researchProduct

Modified total variation regularization using fuzzy complement for image denoising

2015

In this paper, we propose a denoising algorithm based on the Total Variation (TV) model. Specifically, we associate to the regularization term of the Rodin-Osher-Fatimi (ROF) functional a small weight whenever denoising is performed in edge and texture regions, which means less regularization and more details preservation. On the other hand, a large weight is associated if the region being filtered is smooth which means noise will be well suppressed. The weight computation is inspired from the fuzzy edge complement. Experiments on well-known images and comparison with state of the art denoising algorithms demonstrate that the proposed method not only presents good denoising performance but …

fuzzy complementbusiness.industryNoise reductionPattern recognitionTotal variation denoisingNon-local meansRegularization (mathematics)Fuzzy logicElectronic mailtotal variationComputer Science::Computer Vision and Pattern RecognitiondenoisingComputer visionVideo denoisingArtificial intelligenceNoise (video)edge detectorbusinessMathematics2015 International Conference on Image and Vision Computing New Zealand (IVCNZ)
researchProduct

Ocular counterrolling. Some practical considerations of a new evaluation method for diagnostic purposes.

1986

Ocular counterrolling (OCR) data taken from the literature (12 publications) were used to test the best fit (least-square fit) of these measurements with respect to three mathematical models: a sine relation between OCR and the lateral tilt stimulus, a complex cosine-square relation, and a logarithmic relation between OCR gain and tilt. The latter proved to be the best fitting function. On the basis of this model, we attempted to define a physiological transfer function between OCR gain and tilt, which could serve as a reference of normal population, assuming healthy subjects for the investigations applied. Comparison of this physiological range with pathological data demonstrated marked di…

medicine.medical_specialtyBest fittingMathematical modelLogarithmEye MovementsComputer sciencebusiness.industryNormal populationPattern recognitionGeneral MedicineVestibular Function TestsTransfer functionModels BiologicalSurgeryOtolithic MembraneOtorhinolaryngologyComputer Science::Computer Vision and Pattern RecognitionEvaluation methodsmedicineCurve fittingHumansArtificial intelligencebusinessOcular counterrollingActa oto-laryngologica
researchProduct

Laser illumination designs for snapshot multi-spectral-line imaging

2019

For multi-spectral imaging, both acquisition time of the spectral image set and the spectral bandwidth of each image have to be minimized. Ultimate performance can be achieved if the set of monochromatic (single-wavelength) spectral images is obtained with a single snapshot — a technique provisionally called "snapshot multi-spectral-line imaging" or SMSLI. Using contemporary RGB colour cameras, up to three spectral line images can be extracted from a snapshot image data cube at specific illumination that comprises only three spectral lines, each of them positioned within one of the detection bands (R, G or B) [1]. Techniques able to provide more spectral line images are under development, a…

medicine.medical_specialtyComputer sciencebusiness.industryData_MISCELLANEOUSMultispectral imageComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONLaser01 natural sciencesSpectral linelaw.inventionSpectral imaging010309 opticsData cubeOpticslawComputer Science::Computer Vision and Pattern Recognition0103 physical sciencesmedicineRGB color modelSnapshot (computer storage)Monochromatic colorbusiness
researchProduct

Single snapshot RGB multispectral imaging at fixed wavelengths: proof of concept

2014

A concept of single snapshot multispectral imaging by standard RGB image sensors under spectrally-specific illumination comprising a fixed number of narrow spectral lines is discussed and experimentally validated. The limiting conditions, RGB band spectral crosstalk corrections and potential applications for parametric mapping of skin are regarded, along with the preliminary results of the proof-of-concept measurements.

medicine.medical_specialtyComputer sciencebusiness.industryMultispectral imageSpectral lineSpectral imagingWavelengthOpticsProof of conceptComputer Science::Computer Vision and Pattern RecognitionmedicineRGB color modelSnapshot (computer storage)businessParametric statisticsSPIE Proceedings
researchProduct

HDR Imaging Pipeline for Spectral Filter Array Cameras

2017

Multispectral single shot imaging systems can benefit computer vision applications in needs of a compact and affordable imaging system. Spectral filter arrays technology meets the requirement, but can lead to artifacts due to inhomogeneous intensity levels between spectral channels due to filter manufacturing constraints, illumination and object properties. One solution to solve this problem is to use high dynamic range imaging techniques on these sensors. We define a spectral imaging pipeline that incorporates high dynamic range, demosaicing and color image visualization. Qualitative evaluation is based on real images captured with a prototype of spectral filter array sensor in the visible…

medicine.medical_specialtyDemosaicingColor imageComputer sciencebusiness.industryMultispectral imageComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION020206 networking & telecommunications02 engineering and technologyReal imageSpectral imagingHigh-dynamic-range imagingFilter (video)Computer Science::Computer Vision and Pattern Recognition0202 electrical engineering electronic engineering information engineeringmedicine020201 artificial intelligence & image processingComputer visionArtificial intelligencebusinessHigh dynamic range
researchProduct

Development of a Protocol for CCD Calibration: Application to a Multispectral Imaging System

2005

In this paper we describe in detail a method for calibrating a CCD-based camera. The calibration aims to remove both temporal and systematic noises introduced by the sensor, electronics, and optics after which we can correct the non-linearity of its response. For the non-linearity correction we use a simple and powerful approach consisting on a complementary approach between a polynomial fitting and an LUT based algorithm. The proposed methodology is accurate in the sense that it takes into account individual characteristics of each pixel. In each pixel, systematic noises are measured through acquiring offset images, thermal images, and FlatField images. A rigorous protocol for acquiring th…

medicine.medical_specialtyPixelComputer sciencebusiness.industryMultispectral imageComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONSpectral imagingComputer Science::Computer Vision and Pattern RecognitionmedicineCalibrationComputer visionArtificial intelligencebusinessOptical filterInternational Journal of Robotics and Automation
researchProduct

Multi-spectral skin imaging by a consumer photo-camera

2010

The possibilities to perform multi-band spectral imaging by means of a consumer color camera without external filters have been studied. Images at up to 6 spectral bands may be extracted from a single color image after appropriate signal processing. The proposed technique was tested in pilot measurements of in-vivo skin hemoglobin maps and laser-excited autofluorescence images.

medicine.medical_specialtySignal processingMaterials sciencebusiness.industryColor imagePhysics::Medical PhysicsHyperspectral imagingSpectral bandsLaserSpectral imaginglaw.inventionImaging spectroscopyAutofluorescenceOpticslawComputer Science::Computer Vision and Pattern RecognitionmedicinebusinessSPIE Proceedings
researchProduct