Alain Horé

Is there a relationship between peak-signal-to-noise ratio and structural similarity index measure?

In this study, the authors analyse two well-known image quality metrics, peak-signal-to-noise ratio (PSNR) as well as structural similarity index measure (SSIM), and the authors derive an analytical relationship between them which works for some kinds of common image degradations such as Gaussian blur, additive Gaussian noise, Jpeg and Jpeg2000 compressions. The analytical relationship brings more clarity on the interpretation of PSNR and SSIM values, explains some differences found between these quality measures in the literature and confirms some experimental observations regarding these measures. A series of tests realised on images from the Kodak database give a better understanding of the performance of SSIM and PSNR in assessing image quality.

A New Image Scaling Algorithm Based on the Sampling Theorem of Papoulis and Application to Color Images

We present in this paper a new image scaling algorithm which is based on the generalized sampling theorem of Papoulis. The main idea consists in using the first and second derivatives of an image in the scaling process. The derivatives contain information about edges and discontinuities that should be preserved during resizing. The sampling theorem of Papoulis is used to combine this information. We compare our algorithm with nine of the most common scaling algorithms and two measures of quality are used: the standard deviation for evaluation of the blur, and the curvature for evaluation of the aliasing. The results presented here show that our algorithm gives the best images with very few aliasing, good contrast, good edge preserving and few blur. We also present how our algorithm applies to color images.

Reducing aliasing in images: a PDE-based diffusion revisited

In this paper, we introduce a new diffusion algorithm that can be used for reducing aliasing on both step edges and lines. It derives from the diffusion model of Perona and Malik, and works as an adaptive level-curve method in which diffusion is carried out in the normal direction of the gradient for step edges, while the eigenvalues of the Hessian matrix are used for lines. To get sharp images, we use high-pass filters to preserve as much as possible the high frequency content while diffusing. Experimental tests using grayscale and colour images show that our algorithm efficiently reduces aliasing.

A new image scaling algorithm based on the sampling theorem of papoulis

We present in this paper a new image scaling algorithm which is based on the generalized sampling theorem of Papoulis. The main idea consists in using the first and second derivatives of an image in the scaling process. The derivatives contain information about edges and discontinuities that should be preserved during resizing. The sampling theorem of Papoulis is used to combine this information. We compare our algorithm with nine of the most common scaling algorithms and two measures of quality are used: the standard deviation for evaluation of the blur, and the curvature for evaluation of the aliasing. The results presented here show that our algorithm gives the best images with very few aliasing, good contrast, good edge preserving and few blur. We also present how our algorithm applies to color images.

An Edge-Sensing Universal Demosaicing Algorithm

In this paper, we introduce an edge detection algorithm for mosaiced images which can be used to enhance generic demosaicing algorithms. The algorithm is based on pixels color differences in the horizontal, vertical and diagonal directions. By using our edge-detection technique to enhance the universal demosaicing algorithm of Lukac et al., experimental results show that the presence of color shifts and artefacts in demosaiced images is reduced. This is confirmed in regard to both subjective and objective evaluation.

Improving image acquisition: a fish-inspired solution

In this paper, we study the rendering of images with a new mosaic/color filter array (CFA) called the Burtoni mosaic. This mosaic is derived from the retina of the African cichlid fish Astatotilapia burtoni. To evaluate the effect of the Burtoni mosaic on the quality of the rendered images, we use two quality measures in the Fourier domain which are the resolution error and the aliasing error. In our model, no demosaicing algorithm is used, which makes it independent of such algorithms. We also use 11 semantic sets of color images in order to highlight the images classes that are well fitted for the Burtoni mosaic in the process of image acquisition. We have compared the Burtoni mosaic with the Bayer CFA and with an optimal CFA proposed by Hao et al. Experiments have shown that the Burtoni mosaic gives the best performances for images of 9 semantic sets which are the high frequency, aerial, indoor, face, aquatic, bright, dark, step and line classes.

Enhancement of the Quality of Images through Complex Mosaic Configurations

In this paper, we study the rendering of images with a new mosaic called the Burtoni mosaic. This mosaic, which is derived from the retina of the African cichlid fish Astatotilapia Burtoni, is parameterized by the sizes of the red, green and blue pixels, and by the distance between the pixels. To evaluate the effect of the Burtoni mosaic on the quality of a images, we use two signal processing quality measures (resolution and aliasing errors) and one perceptual quality measure (the structural similarity index measure, SSIM). We show that the use of the Burtoni mosaic results in less aliasing and better resolution than the usual vertical stripes or TV pattern. Also, the images obtained using the Burtoni mosaic are closer to the original images, in terms of the SSIM, than those obtained with the popular Bayer and TV mosaics. This result holds for both text and natural images.

A Simple Scaling Algorithm Based on Areas Pixels

In this paper, we propose a new scaling algorithm which performs the scaling up/down transform using an area pixel model rather than a point pixel model. The proposed algorithm uses a variable number of pixels of an original image to calculate a pixel of the scaled image. Our algorithm has good characteristics such as fine-edge and good smoothness. Different quality parameters (standard deviation, root mean square error, aliasing, edge map) are used to compare our algorithm with four image scaling algorithms: nearest neighbour, bilinear, bicubic and winscale algorithms. The results show that our algorithm generally produces images of better quality with few aliasing, few blur and high contrast.

A New Super-Resolution Algorithm Based on Areas Pixels and the Sampling Theorem of Papoulis

In several application areas such as art, medicine, broadcasting and e-commerce, high-resolution images are needed. Super-resolution is the algorithmic process of increasing the resolution of an image given a set of displaced low-resolution, noisy and degraded images. In this paper, we present a new super-resolution algorithm based on the generalized sampling theorem of Papoulis and wavelet decomposition. Our algorithm uses an area-pixel model rather than a point-pixel model. The sampling theorem is used for merging a set of low-resolution images into a high-resolution image, and the wavelet decomposition is used for enhancing the image through efficient noise removing and high-frequency enhancement. The proposed algorithm is non-iterative and not time-consuming. We have tested our algorithm on multiple images and used the peak-to-noise ratio, the structural similarity index and the relative error as quality measures. The results show that our algorithm gives images of good quality.

A generic demosaicing algorithm based on a diffusion model

In this paper, a diffusion-based generic demosaicing algorithm is proposed which can be used for various sensor images captured by digital cameras equipped with various RGB color filter arrays. This algorithm improves our previous edge-sensing generic demosaicing algorithm by enhancing the computation of the green band. In fact, since the green band plays a major and crucial role in the performance of the edge-sensing generic demosaicing algorithm, a diffusion-based model is used for reducing the errors generated when computing the green band. A series of tests has been made on images of the Kodak database, and our diffusion-based demosaicing algorithm performs better than the edge-sensing generic demosaicing algorithm in regard to both subjective and objective evaluation.