Dissecting and Reassembling Color Correction Algorithms for Image Stitching

Abstract

This paper introduces a new compositional framework for classifying color correction methods according to their two main computational units. The framework was used to dissect fifteen among the best color correction algorithms and the computational units so derived, with the addition of four new units specifically designed for this work, were then reassembled in a combinatorial way to originate about one hundred distinct color correction methods, most of which never considered before. The above color correction methods were tested on three different existing datasets, including both real and artificial color transformations, plus a novel dataset of real image pairs categorized according to the kind of color alterations induced by specific acquisition setups. Differently from previous evaluations, special emphasis was given to effectiveness in real world applications, such as image mosaicing and stitching, where robustness with respect to strong image misalignments and light scattering effects is required. Experimental evidence is provided for the first time in terms of the most recent perceptual image quality metrics, which are known to be the closest to human judgment. Comparative results show that combinations of the new computational units are the most effective for real stitching scenarios, regardless of the specific source of color alteration. On the other hand, in the case of accurate image alignment and artificial color alterations, the best performing methods either use one of the new computational units, or are made up of fresh combinations of existing units.

L2 Divergence for robust colour transfer

“…In this section we present further qualitative and quantitative comparisons between our technique and these methods. To quantitatively assess recolouring results, three metricspeak signal to noise ratio (PSNR), structural similarity index (SSIM) and colour image difference (CID) -are often used when considering palette and target images of the same content for which correspondences are easily available (Lissner et al, 2013;Oliveira et al, 2015;Hwang et al, 2014;Bellavia and Colombo, 2018). Alternatively user studies have also been used to assess the perceptual visual quality of the recolouring (Hristova et al, 2015).…”

Entropy-Based Combined Metric for Automatic Objective Quality Assessment of Stitched Panoramic Images

“…Color correction and balancing in the image and video stitching has also been investigated in the paper [13], whereas the mosaicking performance has been examined by Paalanen et al [14]. A classification of color correction methods for image stitching can be made using the framework proposed by Bellavia and Colombo [15] who utilize well-known Feature Similarity (FSIM) metric [16] together with the improved Color Image Difference (iCID) measure [17] to assess the quality. Another idea, useful for the analysis of color inconsistency, has been proposed by Niu et al [18] and is based on the calculation of the color contrast similarity and the color value difference.…”

Restoration and Enhancement of Historical Stereo Photos

“…Moreover, discoloration of the support due to aging can be present. In order to improve the final result, the state-of-the-art color correction method named Gradient Preserving Spline with Linear Color Propagation (GPS/LCP) presented in [ 28 ] is employed to correct the illumination of according to . Specifically, the color map , with is used to obtain the color corrected image according to where, in the error free ideal case, it must hold that (see Figure 2 l).…”