Search results for "ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION"

Efficient Dense Disparity Map Reconstruction using Sparse Measurements

2018

International audience; In this paper, we propose a new stereo matching algorithm able to reconstruct efficiently a dense disparity maps from few sparse disparity measurements. The algorithm is initialized by sampling the reference image using the Simple Linear Iterative Clustering (SLIC) superpixel method. Then, a sparse disparity map is generated only for the obtained boundary pixels. The reconstruction of the entire disparity map is obtained through the scanline propagation method. Outliers were effectively removed using an adaptive vertical median filter. Experimental results were conducted on the standard and the new Middlebury datasets show that the proposed method produces high-quali…

Hierarchical Parallelization of an H.264/AVC Video Encoder

2006

Last generation video encoding standards increase computing demands in order to reach the limits on compression efficiency. This is particularly the case of H.264/AVC specification that is gaining interest in industry. We are interested in applying parallel processing to H.264 encoders in order to fulfill the computation requirements imposed by stressing applications like video on demand, videoconference, live broadcast, etc. Given a delivered video quality and bit rate, the main complexity parameters are image resolution, frame rate and latency. These parameters can still be pushed forward in such a way that special purpose hardware solutions are not available. Parallel processing based on…

GESTALT-INSPIRED FEATURES EXTRACTION FOR OBJECT CATEGORY RECOGNITION

2013

International audience; We propose a methodology inspired by Gestalt laws to ex- tract and combine features and we test it on the object cat- egory recognition problem. Gestalt is a psycho-visual the- ory of Perceptual Organization that aims to explain how vi- sual information is organized by our brain. We interpreted its laws of homogeneity and continuation in link with shape and color to devise new features beyond the classical proxim- ity and similarity laws. The shape of the object is analyzed based on its skeleton (good continuation) and as a measure of homogeneity, we propose self-similarity enclosed within shape computed at super-pixel level. Furthermore, we pro- pose a framework to …

Image Content Enhancement Through Salient Regions Segmentation for People With Color Vision Deficiencies

2019

Color vision deficiencies affect visual perception of colors and, more generally, color images. Several sciences such as genetics, biology, medicine, and computer vision are involved in studying and analyzing vision deficiencies. As we know from visual saliency findings, human visual system tends to fix some specific points and regions of the image in the first seconds of observation summing up the most important and meaningful parts of the scene. In this article, we provide some studies about human visual system behavior differences between normal and color vision-deficient visual systems. We eye-tracked the human fixations in first 3 seconds of observation of color images to build real f…

2014

Codebook is an effective image representation method. By clustering in local image descriptors, a codebook is shown to be a distinctive image feature and widely applied in object classification. In almost all existing works on codebooks, the building of the visual vocabulary follows a basic routine, that is, extracting local image descriptors and clustering with a user-designated number of clusters. The problem with this routine lies in that building a codebook for each single dataset is not efficient. In order to deal with this problem, we investigate the influence of vocabulary sizes on classification performance and vocabulary universality with the kNN classifier. Experimental results in…

A Spatial Pyramidal Decomposition Method for ear representation using local dual cross patterns

2019

International audience; In recent years, several scientific works are oriented to develop optimal ear representation, for ear recognition, which is discriminant, compact, and easyto-implement to ensure the best performance in terms of accuracy, computation cost, and storage requirement. In this manner, this paper presents a novel ear representation based on texture analysis framework, which relies mainly on Dual Cross Pattern (DCP) descriptor and Spatial Pyramid Histogram (SPH) method. The features are extracted using DCP descriptor to capture the textural structure then, the SPH of horizontal ear decomposition is applied to obtain the local information. The feature vector representations o…

Analysis of the depth of field of integral imaging displays based on wave optics.

2013

In this paper, we analyze the depth of field (DOF) of integral imaging displays based on wave optics. With considering the diffraction effect, we analyze the intensity distribution of light with multiple micro-lenses and derive a DOF calculation formula for integral imaging display system. We study the variations of DOF values with different system parameters. Experimental results are provided to verify the accuracy of the theoretical analysis. The analyses and experimental results presented in this paper could be beneficial for better understanding and designing of integral imaging displays.

Image transmission through dynamic scattering media by single-pixel photodetection

2014

Smart control of light propagation through highly scattering media is a much desired goal with major technological implications. Since interaction of light with highly scattering media results in partial or complete depletion of ballistic photons, it is in principle impossible to transmit images through distances longer than the extinction length. Nevertheless, different methods for image transmission, focusing, and imaging through scattering media by means of wavefront control have been published over the past few years. In this paper we show that single-pixel optical systems, based on compressive detection, can also overcome the fundamental limitation imposed by multiple scattering to suc…

Superresolved common-path phase-shifting digital inline holographic microscopy using a spatial light modulator.

2012

Common-path phase-shifting lensless holographic microscopy has been recently proposed as a novel approach capable of high numerical aperture imaging in a lensless digital inline holographic microscopy layout [Opt. Lett.35, 3919 (2010)]. Here we present proof-of-concept validation for improving the resolution limit imposed by diffraction in such a setup. This is accomplished by shifting the phase lens displayed at the spatial light modulator, which moves the illumination point source to different off-axis positions. For each off-axis position, a set of inline phase-shifted holograms are recorded by the digital sensor and stored at the computer’s memory for later digital postprocessing. As a …

Quantitative Phase Imaging in Microscopy Using a Spatial Light Modulator

2010

In this chapter, we present a new method capable of recovery of the quantitative phase information of microscopic samples. Essentially, a spatial light modulator (SLM) and digital image processing are the basics to extract the sample’s phase distribution. The SLM produces a set of misfocused images of the input sample at the CCD plane by displaying a set of lenses with different power at the SLM device. The recorded images are then numerically processed to retrieve phase information. Computations are based on the wave propagation equation and lead to a complex amplitude image containing information of both amplitude and phase distributions of the input sample diffracted wave front. The prop…