Filippo Naccari

Natural scenes classification for color enhancement

An innovative solution for automatic color rendition of natural digital images is presented. It is based on an adaptive color correction, driven by a natural scene classifier designed over a wide database of natural digital images used as reference. The perceived quality of the processed images is globally closer to the expected color rendition for a great part of observers, without evident artifacts. A visual comparison with two wide diffused automatic and semi-automatic color enhancement tools is also presented.

Color space transformations for digital photography exploiting information about the illuminant estimation process

The color reproduction accuracy is a key factor to the overall perceived image quality in digital photography. In this framework, both the illuminant estimation process and the color correction matrix concur in the formation of the overall perceived image quality. To the best of our knowledge, the two processes have always been studied separately, thus ignoring the interactions between them. We investigate here these interactions, showing how the color correction transform amplifies the illuminant estimation errors. We demonstrate that incorporating knowledge about the illuminant estimation behavior in the optimization of the color correction matrix makes it possible to alleviate the error amplification. Different strategies to improve color accuracy under both perfect and imperfect white point estimations are investigated, and the experimental results obtained with a digital camera simulator are reported.

Color correction pipeline optimization for digital cameras

Abstract. The processing pipeline of a digital camera converts the RAW image acquired by the sensor to a representation of the original scene that should be as faithful as possible. There are mainly two modules responsible for the color-rendering accuracy of a digital camera: the former is the illuminant estimation and correction module, and the latter is the color matrix transformation aimed to adapt the color response of the sensor to a standard color space. These two modules together form what may be called the color correction pipeline. We design and test new color correction pipelines that exploit different illuminant estimation and correction algorithms that are tuned and automatically selected on the basis of the image content. Since the illuminant estimation is an ill-posed problem, illuminant correction is not error-free. An adaptive color matrix transformation module is optimized, taking into account the behavior of the first module in order to alleviate the amplification of color errors. The proposed pipelines are tested on a publicly available dataset of RAW images. Experimental results show that exploiting the cross-talks between the modules of the pipeline can lead to a higher color-rendition accuracy.

Coding techniques for CFA data images

In this paper we present a comparison between different approaches to CFA (colour filter array) image encoding. We show different performance offered by a new algorithm based on a vector quantization technique, JPEG-LS, a low complexity encoding standard and classical JPEG. We also show the effects of CFA image encoding on the colour reconstructed images by a typical image generation pipeline. A discussion about the computational complexity and memory requirement of the different encoding approaches is also presented.

Hardware/software solution for high precision defect correction in digital image sensors

High resolution image sensors are now standard in imaging devices such as mobile phones with camera functionality. Resolution improvement in very small sensors is obtained by decreasing the pixel size but, in CMOS sensors, the likelihood of defective pixels also augments. Hence, sophisticated processing is necessary for achieving high quality images despite of noise and defects. This paper presents a hardware/software solution for high precision correction of defective pixels in an image sensor. The method maintains an always up-to-date map of the defective pixels and also allows detection of new defects as they show up during the lifetime of the sensor. The reliability of the map is assured by tracking the history of pixels defectiveness. The map is updated automatically and in real time without user intervention.

Aesthetic scoring of digital portraits for consumer applications

Computational Aesthetics applied on digital photography is becoming an interesting issue in different frameworks (e.g., photo album summarization, imaging acquisition devices). Although it is widely believed and can often be experimentally demonstrated that aesthetics is mainly subjective, we aim to find some formal or mathematical explanations of aesthetics in photographs. We propose a scoring function to give an aesthetic evaluation of digital portraits and group pictures, taking into account faces aspect ratio, their perceptual goodness in terms of lighting of the skin and their position. Also well-known composition rules (e.g., rule of thirds) are considered especially for single portrait. Both subjective and quantitatively experiments have confirmed the effectiveness of the proposed methodology.

Multidomain pixel analysis for illuminant estimation and compensation

The illuminant estimation has an important role in many domain applications such as digital still cameras and mobile phones, where the final image quality could be heavily affected by a poor compensation of the ambient illumination effects. In this paper we present an algorithm, not dependent on the acquiring device, for illuminant estimation and compensation directly in the color filter array (CFA) domain of digital still cameras. The algorithm proposed takes into account both chromaticity and intensity information of the image data, and performs the illuminant compensation by a diagonal transform. It works by combining a spatial segmentation process with empirical designed weighting functions aimed to select the scene objects containing more information for the light chromaticity estimation. This algorithm has been designed exploiting an experimental framework developed by the authors and it has been evaluated on a database of real scene images acquired in different, carefully controlled, illuminant conditions. The results show that a combined multi domain pixel analysis leads to an improvement of the performance when compared to single domain pixel analysis.

Natural images classification by multiple pixel analysis

This work describes an algorithm aimed at performing an automatic natural scene image classification. It is based on a combined analysis of chromatic and spatial features of the pixels, for the recognition of landscape and portrait scenes. Quantitative results are also presented and show the effectiveness of the combined analysis proposed. Various image processing applications can easily take advantage of the proposed solution.

Adaptive low cost algorithm for video stabilization

Video stabilization is a technique used to compensate user hand shaking. It avoids grabbing the unintentional motion in a video sequence, which causes unpleasant effects for the final user. In this paper we present a very simple but effective low power consumption solution, suitable for cheap and small video cameras, running at 31 fps for a VGA sequence with a simple ARM926EJ-S. The proposed solution is robust to common difficult conditions, like noise perturbations, illumination changes, motion blurring and rolling shutter distortions.

Adaptive Low Cost Algorithm for Video Stabilization

Video stabilization is a technique used to compensate user hand shaking. It avoids grabbing the unintentional motion in a video sequence, which causes unpleasant effects for the final user. In this paper we present a very simple but effective low power consumption solution, suitable for cheap and small video cameras, which is robust to common difficult conditions, like noise perturbations, illumination changes, motion blurring and rolling shutter distortions.