Luigi Albani

Minimum-Error Splitting Algorithm for a Dual Layer LCD Display—Part I: Background and Theory

A dual layer high dynamic range liquid crystal display (LCD) can be built by stacking two panels one on top of the other. In this way, the dynamic range is theoretically squared and the bit depth is also increased. However, in order to minimize the parallax and reconstruction errors, dedicated splitting algorithms are needed to generate the two images which drive the panels. In this paper, we present an algorithm, based on variational techniques, which seeks the joint minimization of both errors. We propose a simplified visible difference metric that exploits some limitations of the human visual system and can be easily incorporated into an optimization algorithm. The image splitting task is formulated as a quadratic programming problem, which can be efficiently solved by means of appropriate numerical methods. Preliminary tests on medical images showed that the algorithm has good performances and appears robust with respect to the parameter adjustment.

Picture enhancement in video and block-coded image sequences

We propose a sharpness enhancement technique for video and decoded image sequences. A non-linear function generates a correction signal that is added to the luminance edges. Spatio-temporal information is used to enhance details, avoiding blocking artefacts and noise.

Minimum-Error Splitting Algorithm for a Dual Layer LCD Display—Part II: Implementation and Results

A dual layer liquid crystal display (LCD) is able to achieve a high dynamic range by stacking two liquid crystal panels one on top of the other over an enhanced backlight unit. However, the finite distance between the two panels inevitably introduces a parallax error when the display is observed off-axis, and the dynamic range limitations of the individual panels introduce a reconstruction error near sharp edges in the input image. In Part I, we have formulated the image splitting as a constrained optimization problem in which a joint minimization of the parallax error and the visibility of the reconstruction error is performed.

A rational unsharp masking method for TV applications

In this paper, we present an unsharp masking-based approach for edge enhancing for TV application. The proposed scheme enhances the true details, limits the overshoot near sharp edges and attenuates the temporal artifacts which are visible in an image sequence. Simulations show that the processed sequence presents sharp edges which make it more pleasant to the human eye. Moreover, the amount of noise and nonhomogeneities in the sequence are clearly reduced.

Does Veiling Glare in the Human Eye Hinder Detection in High-Dynamic-Range Displays?

Ever since Stiles and Holladay (1929), veiling glare (VG) in the human visual system has been known to hinder the visibility of subtle targets. In this work, we quantitatively study how veiling glare affects contrast detection tasks using a dual-layer high-dynamic-range (HDR) display and empirically model the VG effect on thresholds. We used a binary decision for the presence of a Gaussian target in the center of the display on white noise backgrounds. The VG source was realized using a ring pattern with varying parameters. Detection thresholds were estimated using a double-random staircase technique including signal absent trials. In addition, divergence of the subjects fixation from the target in the center was tracked in real-time and used to provide auditory feedback to minimize adaptation effects. Our results are interpreted in terms of illuminance and angular distance between source and target. Sensitivity was lower for smaller angular distances and for larger source intensities. Results from three subjects were used to formulate a bivariate model of VG effect for contrast thresholds similar to Stiles and Holladay. The model can be used to suggest optimal, content-dependent, HDR presentation modes for medical images.

Joint Kalman-based noise filtering and motion compensated video coding for low bit rate videoconferencing

The problem of noise filtering for video coding applications is considered in this paper. In a previous paper by S. Olivieri and L. Albini (see Proc. ICIP99, Kobe, Japan, Oct. 1999) an advantageous technique to combine a motion compensated (MC) noise filter with the coding loop of an MC hybrid coding scheme was illustrated, and a rate-distortion (RD) optimization framework was proposed. Here, we describe an alternative structure for denoising image sequences to be coded, which is based on Kalman filtering. Experimental results reveal that this technique provides performances as good as those of the RD-optimized noise filter of S. Olivieri and L. Albini, while the computational complexity has been greatly reduced.

Clinical evaluation of a medical high dynamic range display

PURPOSEnRecent new medical displays do have higher contrast and higher luminance but do not have a High Dynamic Range (HDR). HDR implies a minimum luminance value close to zero. A medical HDR display prototype based on two Liquid Crystal layers has been developed. The goal of this study is to evaluate the potential clinical benefit of such display in comparison with a low dynamic range (LDR) display.nnnMETHODSnThe study evaluated the clinical performance of the displays in a search and detection task. Eight radiologists read chest x-ray images some of which contained simulated lung nodules. The study used a JAFROC (Jacknife Free Receiver Operating Characteristic) approach for analyzing FROC data. The calculated figure of merit (FoM) is the probability that a lesion is rated higher than all rated nonlesions on all images. Time per case and accuracy for locating the center of the nodules were also compared. The nodules were simulated using Sameis model. 214 CR and DR images [half were healthy images (chest nodule-free) and half diseased images] were used resulting in a total number of nodules equal to 199 with 25 images with 1 nodule, 51 images with 2 nodules, and 24 images with 3 nodules. A dedicated software interface was designed for visualizing the images for each session. For the JAFROC1 statistical analysis, the study is done per nodule category: all nodules, difficult nodules, and very difficult nodules.nnnRESULTSnFor all nodules, the averaged FoMHDR is slightly higher than FoMLDR with 0.09% of difference. For the difficult nodules, the averaged FoMHDR is slightly higher than FoMLDR with 1.38% of difference. The averaged FoMHDR is slightly higher than FoMLDR with 0.71% of difference. For the true positive fraction (TPF), both displays (the HDR and the LDR ones) have similar TPF for all nodules, but looking at difficult and very difficult nodules, there are more TP for the HDR display. The true positive fraction has been also computed in function of the local average luminance around the nodules. For the lowest luminance range, there is more than 30% in favor of the HDR display. For the highest luminance range, there is less than 6% in favor of the LDR display.nnnCONCLUSIONSnThis study shows the potential benefit of using a HDR display in radiology.

Effect of Veiling Glare on Detectability in High-Dynamic-Range Medical Images

We describe a methodology for predicting the detectability of subtle targets in dark regions of high-dynamic-range (HDR) images in the presence of veiling glare in the human eye. The method relies on predictions of contrast detection thresholds for the human visual system within a HDR image based on psychophysics measurements and modeling of the HDR display device characteristics. We present experimental results used to construct the model and discuss an image-dependent empirical veiling glare model and the validation of the model predictions with test patterns, natural scenes, and medical images. The model predictions are compared to a previously reported model (HDR-VDP2) for predicting HDR image quality accounting for glare effects.

HDR medical display based on dual layer LCD

There is increased interest in the visualization community to experiment the benefit of HDR presentation. Current developments in HDR displays are geared towards projecting more realistic images than conventional (non-HDR) displays. The dynamic range of the natural world is approximately 14 orders of magnitude while conventional displays are limited to at most 3 orders of magnitude in luminance. A high dynamic range based on Dual Layer liquid crystal display (LCD) is built by stacking two panels one on top of the other. In this way, the dynamic range is theoretically squared and the bit depth is also increased. However, in order to minimize the parallax and reconstruction errors, dedicated splitting algorithms are needed to generate the two images which drive the panels. Moreover, to cope with the reduce transmittance of this Dual Layer LCD concept, a high brightness backlight is required and new LED technology enable a reliable implementation.

GOTHIC: glare optimizer tool for high-dynamic-range images and content with implementation in video

Introduction: The luminance range of the sun to the night sky is approximately 14 orders of magnitude. Current display technology can present approximately three orders of magnitude, however this number is increasing as High-Dynamic-Range (HDR) technology develops to further emulate reality [Seetzen et al. 2004]. Another benefit to HDR technology is the increased bit-depth enabling the display of more information. However, a major limitation in the perception of added bit-depth is veiling glare. The increased luminance range in HDR displays have the ability to produce glare sources that can reduce the visible contrast in neighboring dark areas. This effect is especially undesirable in the visualization of scientific data and in medical images. The HDR presentation must be optimized so that the benefits of a wide luminance range are not diminished by glare in the human visual system. One important question is, what is the largest luminance range that avoids these veiling glare effects while presenting the most bit-depth? We have found that the answer is highly dependent on the spatial and luminance distribution in the image. Many models have been proposed to estimate the veiling glare in a given image. A well known model is High-Dynamic-Range Visual Difference Predictor 2 (HDR-VDP-2) [Mantiuk et al. 2011], a calibrated method able to determine the visibility of differences in HDR images. Building on a number of previous metrics of visible difference, this model operates in a broad range of viewing conditions, from scotopic to photopic vision. More importantly, HDR-VDP-2 can be used to represent the effects of visual glare in signal detection. The inputs of the HDR-VDP-2 are a luminance map of an image, a reference image, and an image with the target. The software outputs the probability of target detection accounting for various visual effects including veiling glare.