Abstract
Color is an effective cue for identifying regions of interest or objects for a wide range of applications in computer vision and digital image processing research. However, color information in recorded image data, typically represented in RGB format, is not always an intrinsic property of an object itself, but rather it also depends on the illumination condition and sensor characteristic. When these factors are not properly taken into consideration, the performance of color analysis system can deteriorate significantly. This chapter investigates two common methodologies to attain reliable color description of recorded image data, color constancy and color invariance. Comprehensive overview of existing techniques are presented. Further, fundamental physical models of light reflection, and a color image formation process in typical imaging devices are discussed, which provide important underlying concepts for various color constancy and invariance algorithms. Finally, two experiments are demonstrated to evaluate the performance of representative color constancy and invariance algorithms.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The choice of canonical illuminant is arbitrary, but for image reproduction application, it is commonly defined as an illuminant for which the camera sensor is balanced [4].
- 2.
Unlike optically homogeneous materials (e.g. metals, glasses, and crystals) which have a constant refraction index throughout the material, inhomogeneous materials (e.g. paints, ceramics, and plastics) are composed of a vehicle with many embedded colorant particles that differs optically from the vehicle. The DRM limits its discussion to optically inhomogeneous materials.
- 3.
The bi-directional reflectance distribution function (BRDF) describes that surface reflectance of a material \(S(\varvec{\theta },\lambda )\) is a function of wavelength \(\lambda \) and imaging geometry \(\varvec{\theta }\), where \(\varvec{\theta }\) contains information related to the directions of the incident and reflected radiance in the local coordinate system. According to the Lambertian surface model, the BRDF is a constant function of the imaging geometry so that \(S(\varvec{\theta },\lambda ) \rightarrow S(\lambda )\).
- 4.
Here, NIR hypothesis is used that the spectral reflectance distribution of the specular reflection term in the DRM is similar to the spectral energy distribution of the scene illuminant.
- 5.
where its horizontal axis represents the inverse-intensity and its vertical axis represents the image chromaticity \(\sigma _c\).
- 6.
The log-relative-chromaticity value of a pixel is defined as the logarithm of the ratio between the chromaticity of the given pixel and the chromaticity of the scene illuminant.
- 7.
The instability of the method when \(I_B = 0\) can be addressed by exploiting histogram based method rather than using input RGB values directly [18].
- 8.
The memory color refers to a group of colors that has the most perceptive impact on the human visual system, such as skin color, blue sky, and green foliage [71].
- 9.
A plane divides 3-dimensional space into two half-spaces. A half-space can be specified by a linear inequality.
- 10.
Here, geometric terms \(w_d\) and \(w_s\) of DRM in (3) are restated by specifying their dependence on the surface normal, the direction of the light source, and the direction of the viewer.
- 11.
Here, we assume the integrated white light condition (i.e. the area under the sensor spectral sensitivity function is approximately the same for all three channels)[35] such that \(\int _\omega \rho _R(\lambda )d\lambda = \int _\omega \rho _G(\lambda )d\lambda = \int _\omega \rho _B(\lambda )d\lambda = \rho \).
- 12.
Although there exist many different ways to compute HSI representation, here, we use a definition given in [55].
- 13.
Shadow edges indicate the shadow of the illuminated object. Highlight edges are generated by the specular property of the object surface. Material edges indicate the discontinuity due to a change of surface material
- 14.
It is well known that Apple systems are calibrated to gamma value of 1.8, whereas most other systems are calibrated to gamma value of 2.2. It implies that images of same scene may appear differently dependent on the target systems.
- 15.
Since the hue is dependent on \(w_d(\mathbf n ,\mathbf s )\) and \(w_s(\mathbf n ,\mathbf s ,\mathbf v )\), it is no longer invariant to surface orientation and illumination direction with the nonlinear response. It is still invariant to illumination intensity as \(E\) term can be cancelled out.
References
Agarwal V, Abidi BR, Koschan A, Abidi MA (2006) An overview of color constancy algorithms. J Pattern Recognit Res 1(1):42–54
Arandjelović O (2012) Colour invariants under a non-linear photometric camera model and their application to face recognition from video. Pattern Recognit 45(7):2499–2509
Barnard K (2000) Improvements to gamut mapping colour constancy algorithms. In: Proceedings of the European conference on computer vision-Part I. Springer, London, pp 390–403
Barnard K, Cardei V, Funt B (2002a) A comparison of computational color constancy algorithms. I: Methodology and experiments with synthesized data. IEEE Trans Image Process 11(9):972–984
Barnard K, Martin L, Funt B, Coath A (2002b) A data set for color research. Color Res Appl 27(3):147–151
Bianco S, Schettini R (2012) Color constancy using faces. IEEE Conf. on Computer Vision and Pattern Recognition, In, pp 65–72
Bianco S, Ciocca G, Cusano C, Schettini R (2008a) Improving color constancy using indoor-outdoor image classification. IEEE Trans Image Process 17(12):2381–2392
Bianco S, Gasparini F, Schettini R (2008b) Consensus-based framework for illuminant chromaticity estimation. J Electron Imaging 17(2):1–9
Bianco S, Ciocca G, Cusano C, Schettini R (2010) Automatic color constancy algorithm selection and combination. Pattern Recognit 43(3):695–705
Bleier M, Riess C, Beigpour S, Eibenberger E, Angelopoulou E, Troger T, Kaup A (2011) Color constancy and non-uniform illumination: can existing algorithms work? In: IEEE international conference on computer vision workshops, pp 774–781
Brainard DH, Freeman WT (1997) Bayesian color constancy. J Opt Soc Am A 14(7):1393–1411
Buchsbaum G (1980) A spatial processor model for object colour perception. J Franklin Inst 310(1):1–26
Cardei VC, Funt BV (1999) Committee-based color constancy. In: Color imaging conference, pp 311–313
Ciurea F, Funt BV (2003) A large image database for color constancy research. In: Color imaging conference, pp 160–164
Delahunt PB, Brainard DH (2004) Does human color constancy incorporate the statistical regularity of natural daylight. J Vis 4(2):57–81
Drew M, Joze H, Finlayson G (2012) Specularity, the Zeta-image, and information-theoretic illuminant estimation. In: European conference on computer vision. Workshops and demonstrations, vol 7584, pp 411–420
Ebner M (2007) Color constancy. Wiley-IS &T series in imaging science and technology. Wiley, West Sussex
Ebner M (2009) Color constancy based on local space average color. Mach Vision Appl 20(5):283–301
Finlayson G (1996) Color in perspective. IEEE Trans Pattern Anal Mach Intell 18(10):1034–1038
Finlayson G, Xu R (2003a) Illuminant and gamma comprehensive normalisation in log RGB space. Pattern Recogn Lett 24(11):1679–1690
Finlayson G, Hordley S, Xu R (2005) Convex programming colour constancy with a diagonal-offset model. In: IEEE international conference on image processing, vol 3, pp III - 948–951
Finlayson GD, Hordley SD (2000) Improving gamut mapping color constancy. IEEE Trans Image Process 9(10):1774–1783
Finlayson GD, Schaefer G (2001a) Hue that is invariant to brightness and gamma. In: BMVC
Finlayson GD, Schaefer G (2001b) Solving for colour constancy using a constrained dichromatic reflection model. Int J Comput Vision 42(3):127–144
Finlayson GD, Trezzi E (2004) Shades of Gray and Colour Constancy. In: Twelfth color imaging conference: color science and engineering systems, technologies, and applications, pp 37–41
Finlayson GD, Xu R (2003b) Convex programming color constancy. In: IEEE workshop on color and photometric methods in computer vision
Finlayson GD, Hordley SD, Hubel PM (2001) Color by correlation: A simple, unifying framework for color constancy. IEEE Trans Pattern Anal Mach Intell 23(11):1209–1221
Forsyth DA (1990) A novel algorithm for color constancy. Int J Comput Vision 5(1):5–36
Foster DH, Amano K, Nascimento SMC, Foster MJ (2006) Frequency of metamerism in natural scenes. J Opt Soc Am A 23(10):2359–2372
Funt B, Shi L (2010a) The effect of exposure on MaxRGB color constancy. In: Proceedings of the SPIE, vol 7527, pp 75,270Y–1–7
Funt B, Shi L (2010b) The rehabilitation of MaxRGB. In: Proceedings of IS &T color imaging conference, pp 256G–259G
Gehler P, Rother C, Blake A, Minka T, Sharp T (2008) Bayesian color constancy revisited. In: IEEE conference on computer vision and pattern recognition, pp 1–8
Geusebroek JM, Smeulders AWM (2005) A six-stimulus theory for stochastic texture. Int J Comput Vision 62(1–2):7–16
Geusebroek JM, Burghouts GJ, Smeulders AWM (2005) The Amsterdam library of object images. Int J Comput Vision 61(1):103–112
Gevers T, Smeulders AWM (1999) Color-based object recognition. Pattern Recognit 32(3):453–464
Gevers T, Stokman H (2003) Classifying color edges in video into shadow-geometry, highlight, or material transitions. IEEE Trans Multimedia 5(2):237–243
Gijsenij A, Gevers T (2007) Color constancy by local averaging. In: International conference on image analysis and processing workshops, pp 171–174
Gijsenij A, Gevers T (2011) Color constancy using natural image statistics and scene semantics. IEEE Trans Pattern Anal Mach Intell 33(4):687–698
Gijsenij A, Gevers T, Lucassen M (2009) A perceptual analysis of distance measures for color constancy algorithms. J Opt Soc Am A 26(10):2243–2256
Gijsenij A, Gevers T, van de Weijer J (2010) Generalized gamut mapping using image derivative structures for color constancy. Int J Comput Vision 86(2–3):127–139
Gijsenij A, Gevers T, van de Weijer J (2011) Computational color constancy: survey and experiments. IEEE Trans Image Process 20(9):2475–2489
Gijsenij A, Gevers T, van de Weijer J (2012a) Improving color constancy by photometric edge weighting. IEEE Trans Pattern Anal Mach Intell 34(5):918–929
Gijsenij A, Lu R, Gevers T (2012b) Color constancy for multiple light sources. IEEE Trans Image Process 21(2):697–707
Hordley SD (2006) Scene illuminant estimation: past, present, and future. Color Res Appl 31(4):303–314
Hordley SD, Finlayson GD (2006) Reevaluation of color constancy algorithm performance. J Opt Soc Am A 23(5):1008–1020
Kakumanu P, Makrogiannis S, Bourbakis N (2007) A survey of skin-color modeling and detection methods. Pattern Recognit 40(3):1106–1122
Kanan C, Flores A, Cottrell GW (2010) Color constancy algorithms for object and face recognition. In: Proceedings of the international conference on advances in visual computing—Volume Part I. Springer, Berlin, pp 199–210
Khan R, Hanbury A, St+ttinger J, Bais A, (2012) Color based skin classification. Pattern Recogn Lett 33(2):157–163
Land EH (1977) The Retinex theory of color vision. Scientific Am 237(6):108–128
Lee HC (1986) Method for computing the scene-illuminant chromaticity from specular highlights. J Opt Soc Am A 3(10):1694–1699
Lee HC, Breneman E, Schulte C (1990) Modeling light reflection for computer color vision. IEEE Trans Pattern Anal Mach Intell 12(4):402–409
Li B, Xu D, Lang C (2009) Colour constancy based on texture similarity for natural images. Coloration Technol 125(6):328–333
MacAdam D (1970) Sources of color science. MIT Press, Cambridge
Mallick S, Zickler T, Kriegman D, Belhumeur P (2005) Beyond lambert: reconstructing specular surfaces using color. In: IEEE conference on computer vision and pattern recognition, vol 2, pp 619–626
Perez F, Koch C (1994) Toward color image segmentation in analog vlsi: algorithm and hardware. Int J Comput Vision 12:17–42
Poynton C (2003) Digital video and HDTV: algorithms and interfaces. Morgan Kaufmann series in computer graphics and geometric Mo. Morgan Kaufmann, San Francisco
Riess C, Eibenberger E, Angelopoulou E (2011) Illuminant color estimation for real-world mixed-illuminant scenes. In: IEEE international conference on computer vision workshops, pp 782–789
Rosenberg CR, Minka TP, Ladsariya A (2003) Bayesian color constancy with Non-Gaussian models. In: Neural information processing systems. MIT Press, Cambridge
van de Sande K, Gevers T, Snoek C (2010) Evaluating color descriptors for object and scene recognition. IEEE Trans Pattern Anal Mach Intell 32(9):1582–1596
Schaefer G, Hordley S, Finlayson G (2005) A combined physical and statistical approach to colour constancy. In: IEEE conference on computer vision and pattern recognition, vol 1, pp 148–153
Shafer SA (1992) Using color to separate reflection components. In: Healey GE, Shafer SA, Wolff LB (eds) Color. Jones and Bartlett Publishers, Inc., Boston, pp 43–51
Swain MJ, Ballard DH (1991) Color indexing. Int J Comput Vision 7(1):11–32
Tan RT, Nishino K, Ikeuchi K (2004) Color constancy through inverse-intensity chromaticity space. J Opt Soc Am A 21(3):321–334
Tominaga S, Wandell BA (1989) Standard surface-reflectance model and illuminant estimation. J Opt Soc Am A 6(4):576–584
Vazquez-Corral J, Vanrell M, Baldrich R, Tous F (2012) Color constancy by category correlation. IEEE Trans Image Process 21(4):1997–2007
Viola PA, Jones MJ (2004) Robust real-time face detection. Int J Comput Vision 57(2):137–154
van de Weijer J, Gevers T, Geusebroek JM (2005) Edge and corner detection by photometric quasi-invariants. IEEE Trans Pattern Anal Mach Intell 27(4):625–630
van de Weijer J, Gevers T, Bagdanov A (2006) Boosting color saliency in image feature detection. IEEE Trans Pattern Anal Mach Intell 28(1):150–156
van de Weijer J, Gevers T, Gijsenij A (2007a) Edge-based color constancy. IEEE Trans Image Process 16(9):2207–2214
van de Weijer J, Schmid C, Verbeek J (2007b) Using high-level visual information for color constancy. In: IEEE international conference on computer vision, pp 1–8
Yendrikhovskij SN, Blommaert FJJ, de Ridder H (1999) Color reproduction and the naturalness constraint. Color Res Appl 24(1):52–67
Zickler T, Mallick SP, Kriegman DJ, Belhumeur PN (2008) Color subspaces as photometric invariants. Int J Comput Vision 79(1):13–30
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Lee, D., Plataniotis, K.N. (2014). A Taxonomy of Color Constancy and Invariance Algorithm. In: Celebi, M., Smolka, B. (eds) Advances in Low-Level Color Image Processing. Lecture Notes in Computational Vision and Biomechanics, vol 11. Springer, Dordrecht. https://doi.org/10.1007/978-94-007-7584-8_3
Download citation
DOI: https://doi.org/10.1007/978-94-007-7584-8_3
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-007-7583-1
Online ISBN: 978-94-007-7584-8
eBook Packages: EngineeringEngineering (R0)