Cristian Perra

Image blockiness evaluation based on Sobel operator

This paper presents a novel no-reference image blockiness index which exploits the properties of the Sobel operator. The blockiness index is a combination of two measures. The first quantifies the luminance variation of block boundaries pixels. The second quantifies the luminance variation of remaining pixels. Experimental results illustrating the performance of the proposed method show improvements over similar recently published methods.

Data formats for high efficiency coding of Lytro-Illum light fields

Light fields captured by the Lytro-Illum camera are the first to appear in the consumer market, capable of providing refocused pictures at acceptable spatial resolution and quality. Since this is partially due to sampling of a huge number of light rays, efficient compression methods are required to store and exchange light field data. This paper presents a performance study of HEVC-compatible coding of Lytro-Illum light fields using different data formats for standard coding. The efficiency of 5 different light field data formats are evaluated using a data set of 12 light field images and the standard HEVC coding configurations of Still-Image Profile, All-Intra, Low Delay B and P and Random Access. Unexpectedly, the results show that relative performance is not consistent across all coding configurations, raising new research questions regarding standard coding of Lytro-Illum light fields using HEVC. Most importantly, the proposed data formats greatly increase HEVC performance.

Lossless plenoptic image compression using adaptive block differential prediction

Plenoptic images are obtained from the projection of the light crossing a matrix of microlens arrays which replicates the scene from different direction into a camera device sensor. Plenoptic images have a different structure with respect to regular digital images, and novel algorithms for data compression are currently under research. This paper proposes an algorithm for the compression of plenoptic images. The micro images composing a plenoptic image are processed by an adaptive prediction tool, aiming at reducing data correlation before entropy coding takes place. The algorithm is compared with state-of-the-art image compression algorithms, namely, JPEG 2000 and JPEG XR. Obtained results demonstrate that the proposed algorithm improves the coding efficiency.

High efficiency coding of light field images based on tiling and pseudo-temporal data arrangement

Digital light fields are obtained by sampling of huge number of light rays, capturing the colour intensity of a whole visual scene from different view angles. The large amount of data that is necessary to represent a light field requires highly efficient compression methods to store and exchange such type of signals. This paper presents a light field coding scheme based on a low-complexity preprocessing approach that generates a pseudo-video sequence suitable for standard compression using HEVC. The efficiency of the proposed scheme is evaluated using a data set of twelve light field images and low-delay B HEVC coding configuration. The R-D performance obtained by the proposed scheme is consistently better than JPEG, revealing significantly higher gains, particularly for higher compression ratios.

An image browsing application based on JPEG XR

A method for browsing large compressed images over the Internet is presented. Images are compressed using a very recent coded file format, referred to as JPEG XR, designed primarily for storage of continuous-tone photographic images and currently at the beginning of the ISO/IEC standardization process. In the proposed method, region of interest specified by the user are progressively downloaded and displayed with an approach aiming to minimize the transferred information between the request and the image presentation. Experimental analysis show that the proposed method is an effective way for browsing large JPEG XR coded images.

TV white spaces exploitation for multimedia signal distribution

The new spectrum regulation policies for dynamic spectrum access, especially those concerning the use of the white spaces in the Digital Terrestrial Television (DTT) bands, arise the need for fast and reliable signal identification and classification methods. In this paper we present a two-stage identification method for signals in the white spaces, using combined energy detection and feature detection. The band of interest is divided by means of the Discrete Wavelet Packet Transformation (DWPT) in sub-bands where the signal power is calculated. Modulation classifiers taking into account the statistical parameters of the signal in the wavelet domain are used as features for identifying the modulation schemes, in this case specifically for the DVB-T broadcast standard. Finally, a signal transmission architecture based on Motion JPEG XR has been implemented in order to explore and evaluate a practical application of indoor signal distribution over white-spaces.

Interactive Image Viewing in Mobile Devices Based on JPEG XR

Services for high definition image browsing on mobile devices require a careful design since the user experience is heavily depending on the network bandwidth, processing delay, display resolution, image quality. Modern applications require coding technologies providing tools for resolution and quality scalability, for accessing spatial regions of interest (ROI), for reducing the domain of the coding algorithm decomposing large images into tiles. Some state-of-the-art technologies satisfying these requirements are the JPEG2000 and the JPEG XR. This paper presents the design of an interactive high resolution image viewing architecture for mobile devices based on JPEG XR. Display resolution, resolution scalability, image tiling are investigated in order to optimize the coding parameters with the objective to improve the user experience. Experimental tests are performed on a set of large images and comparisons against accessing the images without parameter optimization are reported.

High-frequency error recovery in JPEG XR coded images

Image data compressed with JPEG XR when transmitted over error-prone channels is severely distorted. In fact, due to the adaptive coding strategies used by the codec, even a single bit error causes a mismatch in the alignment of the reading position from the bitstream, leading to completely different images at the decoder side. We propose an extension to the JPEG XR algorithm, consisting in an error recovery process enabling the decoder to realign itself to the right bitstream position and to correctly decode the most part of the image. We have performed several experiments using different encoder parameter and different error probabilities while image distortion is measured by PSNR objective metric. The simplicity of the proposed algorithm adds very little computational overhead and seems very promising as confirmed by objective image quality results in experimental tests.

On the coding of plenoptic raw images

Plenoptic images are obtained from cameras with internal microlens arrays. Nearby pixels in a plenoptic image are representing the real world from different spatial and angular position and the general assumption of spatial correlation exploited for the development of former image coding algorithms loses most of its foundations. In this work, the problem of coding plenoptic images is analyzed in order to better understand the compromise between raw signal quantization and quality loss. Moreover, some image coding standards have been analyzed as tools for the compression of raw plenoptic images. Results have been measured both in terms of PSNR and SSIM of final reconstructed images.

A low computational complexity blockiness estimation based on spatial analysis

Measuring blockiness distortion in images and video is fundamental in the field of image and video quality assessment. A no-reference method for image blockiness estimation in the spatial domain is presented in this paper. Based on the Sobel operator, a low computational complexity algorithm is proposed for measuring blockiness distortion. Simulation results show that the blockiness estimation obtained by the proposed algorithm is comparable with algorithms demanding higher computational complexity.