Editing Color constancy (section)

==Retinex theory==
The "Land effect" is the capacity to see full color images solely by looking at superimposed images of black and white transparencies of the same scene, one taken through a red filter and the other taken through a green filter, and illuminated by red and white light, respectively (or even by two different yellow wavelengths). The effect was discovered by [[Edwin H. Land]], who was attempting to reconstruct [[James Clerk Maxwell]]'s early experiments in full-colored images. Land saw that, even when only yellow light illuminated the superimposed images, the visual system would still perceive a full (if muted) range of color. Land described this effect in a 1959 article in ''[[Scientific American]].''<ref>{{Cite journal|url=http://www.psy.vanderbilt.edu/courses/psy236/ColorVision/Land1959.pdf|title=Experiments in Color Vision|last=Land|first=Edwin|date=May 1959|journal=Scientific American|volume=200|issue=5|pages=84–94 passim|doi=10.1038/scientificamerican0559-84|pmid=13646648|bibcode=1959SciAm.200e..84L}}</ref><ref name=wendy /> In 1977, Land wrote another ''Scientific American'' article that described a generalized Land effect, leading to formulation of his "Retinex Theory" to explain what he believed was main basis of human color vision.<ref>{{Cite journal |last=Land |first=Edwin |date=December 1977 |title=The Retinex Theory of Color Vision |url=https://www.jstor.org/stable/24953876 |journal=Scientific American |volume=237 |issue=6 |pages=108–128 |bibcode=1977SciAm.237f.108L |doi=10.1038/scientificamerican1277-108 |jstor=24953876 |pmid=929159 |archive-url=https://web.archive.org/web/20240311155053/https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=d34b21123f86b8515e1150472efbb42147654a18 |archive-date=2024-03-11|url-access=subscription }}</ref> The word "retinex" is a [[blend word|blend]] of "[[retina]]" and "[[Cerebral cortex|cortex]]", suggesting that both the eye and the brain are involved in the processing.

The generalized Land effect can be experimentally demonstrated as follows. A display called a "Mondrian" (after [[Piet Mondrian]] whose paintings are similar) consisting of numerous colored patches is shown to a person. The display is illuminated by three white lights, one projected through a red filter, one projected through a green filter, and one projected through a blue filter. The person is asked to adjust the intensity of the lights so that a particular patch in the display appears white. The experimenter then measures the intensities of red, green, and blue light reflected from this white-appearing patch. Then the experimenter asks the person to identify the color of a neighboring patch, which, for example, appears green. Then the experimenter adjusts the lights so that the intensities of red, blue, and green light reflected from the green patch are the same as were originally measured from the white patch. The person shows color constancy in that the green patch continues to appear green, the white patch continues to appear white, and all the remaining patches continue to have their original colors.

Land, with John McCann, also developed a computer program designed to imitate the retinex processes thought to be taking place in human physiology.<ref>J. McCann, S.P. McKee & T. Taylor, "[https://mccannimaging.com/Retinex/Land_McCann_Retinex_files/76MMT%20VisRes.pdf Quantitative Studies in Retinex Theory, A Comparison Between Theoretical Predictions and Observer Responses to Color Mondrian Experiments"], Vision Res. '''16''': 445–458, (1976)</ref> Color constancy is a desirable feature of [[computer vision]], and many algorithms have been developed for this purpose. These include several retinex algorithms.<ref>{{cite journal | doi = 10.1117/12.805474 | title=Fast implementation of color constancy algorithms | year=2009 | journal=Color Imaging XIV: Displaying, Processing, Hardcopy, and Applications | volume=7241 | pages=724106 | last1 = Morel | first1 = Jean-Michel | last2 = Petro | first2 = Ana B. | last3 = Sbert | first3 = Catalina| bibcode=2009SPIE.7241E..06M | editor4-first=Alessandro | editor4-last=Rizzi | editor3-first=Shoji | editor3-last=Tominaga | editor2-first=Gabriel G | editor2-last=Marcu | editor1-first=Reiner | editor1-last=Eschbach | citeseerx=10.1.1.550.4746 | s2cid=19950750 }}</ref><ref>{{cite journal |first1=R. |last1=Kimmel |first2=M. |last2=Elad |first3=D. |last3=Shaked |first4=R. |last4=Keshet |first5=I. |last5=Sobel |url=https://www.cs.technion.ac.il/~ron/PAPERS/retinex_ijcv2003.pdf |title=A Variational Framework for Retinex |journal=International Journal of Computer Vision |volume=52 |issue=1 |pages=7–23 |year=2003 |doi=10.1023/A:1022314423998|s2cid=14479403 }}</ref><ref>{{cite patent |invent1=Barghout, Lauren  |invent2=Lawrence Lee |title=Perceptual information processing system |country=US |status=Patent |number=20040059754A1}}</ref><ref>{{cite book | last=Barghout | first=Lauren | series=Communications in Computer and Information Science | volume=443 | title=Information Processing and Management of Uncertainty in Knowledge-Based Systems | chapter=Visual Taxometric Approach to Image Segmentation Using Fuzzy-Spatial Taxon Cut Yields Contextually Relevant Regions | publisher=Springer International Publishing | publication-place=Cham | year=2014 | isbn=978-3-319-08854-9 | issn=1865-0929 | doi=10.1007/978-3-319-08855-6_17 | pages=163–173}}</ref> These algorithms receive as input the red/green/blue values of each [[pixel]] of the image and attempt to estimate the reflectances of each point. One such algorithm operates as follows: the maximal red value ''r''<sub>max</sub> of all pixels is determined, and also the maximal green value ''g''<sub>''max''</sub> and the maximal blue value {{not a typo|''b''<sub>max</sub>}}. Assuming that the scene contains objects which reflect all red light, and (other) objects which reflect all green light and still others which reflect all blue light, one can then deduce that the illuminating light source is described by (''r''<sub>max</sub>, ''g''<sub>max</sub>, ''b''<sub>max</sub>). For each pixel with values (''r'', ''g'', ''b'') its reflectance is estimated as (''r''/''r''<sub>max</sub>, ''g''/''g''<sub>max</sub>, ''b''/''b''<sub>max</sub>). The original retinex algorithm proposed by Land and McCann uses a localized version of this principle.<ref>{{cite journal | last1 = Provenzi | first1 = Edoardo | last2 = De Carli | first2 = Luca | last3 = Rizzi | first3 = Alessandro | last4 = Marini | first4 = Daniele | year = 2005 | title = Mathematical definition and analysis of the Retinex algorithm | journal = JOSA A | volume = 22 | issue = 12| pages = 2613–2621 | doi=10.1364/josaa.22.002613| pmid = 16396021 | bibcode = 2005JOSAA..22.2613P }}</ref><ref>{{cite journal | last1 = Bertalmío | first1 = Marcelo | last2 = Caselles | first2 = Vicent | last3 = Provenzi | first3 = Edoardo | year = 2009 | title = Issues About Retinex Theory and Contrast Enhancement | journal =  International Journal of Computer Vision| volume = 83 | pages = 101–119 | doi=10.1007/s11263-009-0221-5| s2cid = 4613179 }}</ref>

Although retinex models are still widely used in computer vision, actual human color perception has been shown to be more complex.<ref>{{cite conference | last1=Hurlbert | first1=Anya C. | last2=Wolf | first2=Christopher J. L. | editor-first1=Bernice E. | editor-first2=Thrasyvoulos N. | editor-last1=Rogowitz | editor-last2=Pappas | title=Contribution of local and global cone-contrasts to color appearance: a Retinex-like model | series=Human Vision and Electronic Imaging VII | publisher=SPIE | date=2002-06-03 | volume=4662 | pages=286–297 | issn=0277-786X | doi=10.1117/12.469525 }}</ref>