Distortion | EPFL Graph Search

In signal processing, distortion is the alteration of the original shape (or other characteristic) of a signal. In communications and electronics it means the alteration of the waveform of an information-bearing signal, such as an audio signal representing sound or a video signal representing images, in an electronic device or communication channel. Distortion is usually unwanted, and so engineers strive to eliminate or minimize it. In some situations, however, distortion may be desirable. For example, in noise reduction systems like the Dolby system, an audio signal is deliberately distorted in ways that emphasize aspects of the signal that are subject to electrical noise, then it is symmetrically "undistorted" after passing through a noisy communication channel, reducing the noise in the received signal. Distortion is also used as a musical effect, particularly with electric guitars. The addition of noise or other outside signals (hum, interference) is not considered distortio

On the Use of Masking Models for Image and Audio Watermarking

Justin Picard, Arnaud Robert

In most watermarking systems, masking models, inherited from data compression algorithms, are used to preserve fidelity by controlling the perceived distortion resulting from adding the watermark to the original signal. So far, little attention has been paid to the consequences of using such models on a key design parameter: the robustness of the watermark to intentional attacks. The goal of this paper is to demonstrate that by considering fidelity alone, key information on the location and strength of the watermark may become available to an attacker; the latter can exploit such knowledge to build an effective mask attack. First, defining a theoretical framework in which analytical expressions for masking and watermarking are laid, a relation between the decrease of the detection statistic and the introduced perceptual distortion is found for the mask attack. The latter is compared to the Wiener filter attack. Then, considering masking models widely used in watermarking, experiments on both simulated and real data (audio and images) demonstrate how knowledge on the mask enables to greatly reduce the detection statistic, even for small perceptual distortion costs. The critical tradeoff between robustness and distortion is further discussed, and conclusions on the use of masking models in watermarking drawn.

2005

Advanced OFDM systems for terrestrial multimedia links

Renzo Posega

Recently, there has been considerable discussion about new wireless technologies and standards able to achieve high data rates. Due to the recent advances of digital signal processing and Very Large Scale Integration (VLSI) technologies, the initial obstacles encountered for the implementation of Orthogonal Frequency Division Multiplexing (OFDM) modulation schemes, such as massive complex multiplications and high speed memory accesses, do not exist anymore. OFDM offers strong multipath protection due to the insertion of the guard interval; in particular, the OFDM-based DVB-T standard had proved to offer excellent performance for the broadcasting of multimedia streams with bitrates over ten megabits per second in difficult terrestrial propagation channels, for fixed and portable applications. Nevertheless, for mobile scenarios, improving the receiver design is not enough to achieve error-free transmission especially in presence of deep shadow and multipath fading and some modifications of the standard can be envisaged. To address long and medium range applications like live mobile wireless television production, some further modifications are required to adapt the modulated bandwidth and fully exploit channels up to 24MHz wide. For these reasons, an extended OFDM system is proposed that offers variable bandwidth, improved protection to shadow and multipath fading and enhanced robustness thanks to the insertion of deep time-interleaving coupled with a powerful turbo codes concatenated error correction scheme. The system parameters and the receiver architecture have been described in C++ and verified with extensive simulations. In particular, the study of the receiver algorithms was aimed to achieve the optimal tradeoff between performances and complexity. Moreover, the modulation/demodulation chain has been implemented in VHDL and a prototype system has been manufactured. Ongoing field trials are demonstrating the ability of the proposed system to successfully overcome the impairments due to mobile terrestrial channels, like multipath and shadow fading. For short range applications, Time-Division Multiplexing (TDM) is an efficient way to share the radio resource between multiple terminals. The main modulation parameters for a TDM system are discussed and it is shown that the 802.16a TDM OFDM physical layer fulfills the application requirements; some practical examples are given. A pre-distortion method is proposed that exploit the reciprocity of the radio channel to perform a partial channel inversion achieving improved performances with no modifications of existing receivers.

EPFL2005

(Geometry Aware) Deep Learning-based Omnidirectional Image Compression

Yamin Sepehri

Omnidirectional images are the spherical visual signals that provide a wide, 360◦, view of a scene from a specific position. Such images are becoming increasingly popular in fields like virtual reality and robotics. Compared to conventional 2D images, the storage and badwidth requirements of omnidirectional signals are much higher, due to the specific nature of them. Thus, there is a need for image compression schemes to reduce the dedicated storage space of omnidirectional images. Image compression algorithms can be broadly classified into two groups: lossless and lossy. Lossless schemes are able to reconstruct the exact original data but they cannot reduce the size beyond a specific criteria. Lossy methods are generally better solutions if they do not add a high visual distortion to the reconstructed image, as long as they provide a decent compression rate. If a planar, lossy image compression scheme is applied on omnidirectional images, some problems show up. It is possible to apply a planar compression scheme on a projected version of a 360◦image; however, in these projection schemes (such as equirectangular projection) the sampling rate is different in the poles and the center. Consequently, the filters of the planar compression schemes that do not consider this difference ends in suboptimal result and distortions in the reconstructed images. Recently, with the success of deep neural networks in many image processing tasks, researchers began to use them for the image compression as well. In this study, we propose a deep learning-based method for the compression of omnidirectional images by combining some state of the art approaches from the deep learning-based image compression schemes and some special convolutional layers that take into account the geometry of the omnidirectional image. In comparison to the available methods, it is the first method that can be applied directly on the equirectangularly projected version of omnidirectional images and considers the geometry in the scheme and the layers themselves. To propose this method, different geometry-aware convolutional layers have been tried. We exploited various methods of downsampling and upsampling, such as spherical pooling layers, strided or transposed convolutions, bilinear interpolation, and pixel shuffle. In the end, a method is proposed that benefits from specific spherical convolutional layers which contain sampling methods considering the geometry of omnidirectional images. The sampling positions differ in the different heights of the image based on the nature of the projected omnidirectional image. Additionally, as it benefits from an iterative training method that calculates the residual between the output and input and feeds it again to the network as input of the next iteration, it can provide different compression rates with just one pass of training. Finally, it benefits from a novel method of patching that is well-aligned with the spherical convolution layers and helps the method to run efficiently without a need for a high computational power. The model was compared with a similar architecture without spherical convolutions and spherical patching and showed some improvements. The architecture has been optimized and improved and it has the potential to compete with popular image compression schemes such as JPEG especially in terms of reconstructing the colors.

2020