Image Color Transfer Approach by Analogy with Taylor Expansion

Image Color Transfer Approach by Analogy with Taylor Expansion

Hongbo Liu, Ye Ji, Aboul Ella Hassanien
Copyright: © 2013 |Pages: 12
DOI: 10.4018/ijsda.2013040103
(Individual Articles)
No Current Special Offers


The Taylor expansion has shown in many fields to be an extremely powerful tool. In this paper, the authors investigated image features and their relationships by analogy with Taylor expansion. The kind of expansion could be helpful for analyzing image feature and engraftment, such as transferring color between images. By analogy with Taylor expansion, the image color transfer algorithm is designed by the first and second-order information. The luminance histogram represents the first-order information of image, and the co-occurrence matrix represents the second-order information of image. Some results illustrate the proposed algorithm is effective. In this study, each polynomial in the Taylor analogy expansion of images is considered as one of image features which help in re-understanding images and its features. By using the proposed technique, the features of image, such as color, texture, dimension, time series, would be not isolated but mutual relational based on image expansion.
Article Preview

In previous published work, Reinhard et al. introduced a method (Reinhard, Adhikhmin, Gooch, & Shirley, 2001), in which colors from a source image were transferred to a second colored image using a simple but surprisingly successful procedure. The basic method matched the three-dimensional distribution of color values between the images and then transforms the color distribution of the destination image to match the distribution of the source image. Welsh et al. transferred color to grayscale images (Welsh, Ashikhmin, & Mueller, 2002), in which the greyscale image was represented by a one dimensional distribution, hence only the luminance channels can be matched between the two images. We pay attention to not only luminance but also texture factor on rendering technique between pictures. Texture analysis and synthesis has had a long history in psychology, statistics and computer vision. Gibson pointed out the importance of texture for visual perception (Engelbrecht, 2000). Julesz had done some pioneering work on texture discrimination that paved the way for the development of the field (Julesz, 1962). Eferos et al. studied image quilting for texture synthesis and transfer (Efros & Leung, 1999; Efros & Freeman, 2001). Nealen and Alexa described improved methods, hybrid texture synthesis technique for texture synthesis and show how these methods can also be used for texture transfer (Nealen & Alexa, 2003). Hertzmann et al. described a new framework for processing images by example, called “image analogies” which generalizes texture synthesis for the case of two corresponding image pairs (Hertzmann, Jacobs, Oliver, Curless, & Salesin, 2003). Pitie et al. propose a kind color transfer technique for the image processing (Pitie, Kokaram, & Dahyot, 2007). Chung and Wen develop an algorithm for colorizing a grayscale image. In their approach, a block-based vector quantization of luminance mapping technique are used to automatically colorize the grayscale image to improve the quality of the colorized grayscale image (Chung & Chen, 2009). Quang et al. propose a reproducing kernel Hilbert space framework for image and video colorization (Quang, Kang, & Le, 2010). Pouli and Reinhard present a kind of histogram reshaping technique which allows significantly better control and transfers the color palette between images of arbitrary dynamic range (Pouli & Reinhard, 2010). Image analogies are applied in several disparate areas, including machine learning, texture synthesis, nonphotorealistic rendering, and image-based rendering (Ruderman, Cronin, & Chiao, 1998; Ji, Liu, Wang, & Tang, 2004; Ji & Chen, 2008).

Complete Article List

Search this Journal:
Volume 12: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 11: 5 Issues (2022)
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing