wikimedia.org
Edwin Land was born on May 7, 1909, in Bridgeport, CT, to Harry and Martha Land. He was educated at Norwich Free Academy in Norwich, CT where he graduated with honors. He spent one year at Harvard University studying chemistry, dropped out and moved to New York City. He educated himself at the New York Public Library and experimented after hours at a Columbia University laboratory where he invented the Polaroid J sheet, an arrangement of microscopic crystals on a plastic sheet filtering out certain kinds of polarized light, used for example in sunglasses. He obtained a patent for it in 1929. He returned to Harvard where he concentrated on chemistry and physics. He and a partner formed a company that produced polarized sheets, named Polaroid Corporation in 1937. During World War II, the company developed new kinds of night vision goggles and a system that could reveal enemies in camouflage uniforms. When in New Mexico in 1944, Land’s three-year-old daughter Jennifer asked him why the camera he used could not immediately produce an image. This got Land to think about such possibilities, and early in 1947, the Polaroid Model 95 camera and the related Type 40 Land film producing black-and-white images began to be offered on the market. The system soon became a big success.
In 1963, Polaroid began to produce a color camera and film, the Polacolor system. A fully automatic color system, SX 70, was introduced in 1972. The success of the Polaroid systems lasted until the introduction of digital cameras. Land’s interest in the phenomena of color vision began already in the 1930s.
In the years after the War, Land continued to cooperate with the US Federal Government, being involved in the development of the U2 spying plane. In 1982, Land retired from the Polaroid Corporation and began to spend much time on the subject of color vision. He also established the Rowland Institute for Science in Cambridge, MA, associated with Harvard University, where research in many advanced fields was and is conducted.
Land was listed on 535 patents as inventor. He received some 20 honorary Ph.D. awards, including one from Harvard. He obtained the Presidential Medal of Freedom from the US Government in 1963 and became a member of the Royal Society in England. He obtained 15 awards from various organizations, including the Optical Society of America. Land died on March 1, 1991, in Cambridge, MA [1].
80.1 Retinex Color Vision Model
Already in 1959, Land published an article in Scientific American, titled “Experiments in color vision” [2]. There he described experiments that point to higher complexity of the color vision system in case of complex stimulus patterns than demonstrated for simple patterns: a test sample in a simple surround. He made black-and-white transparent slide images of complex scenes containing many different colored objects, one generated through a red filter and the other through a green one. Then he projected them overlapping onto a screen, with the light source for the one made through the red filter being long-wavelength light and the other one broadband colorless light. The conventional expectation is that the appearance of the projection is only in various levels of redness. However, surprisingly, the projected image had nearly normal appearance, with blue, green, yellow, and other perceived color present. He experimentally evaluated the appearances with many different combinations of projector lights resulting in varying outcomes but usually multi-colored appearances. As he commented, “In this experiment we are forced to the astonishing conclusion that the rays are not in themselves color-making. Rather they are bearers of information that the eye uses to assign appropriate colors to various objects in an image.” Using only two chromatic primaries (or one with an achromatic one) in this projection experiment of complex images does not result in reproduction of equal chromatic quality as three chromatic primaries, but its surprising results have as yet not been fully explained.
Image of land’s experimental setup for his Mondrian experiments
Land’s retinex theory resulted in criticism from other color researchers who showed that his effects were more limited than claimed and that a number of findings had been shown by previous workers. Two-primary color reproduction did not receive any practical application. The broad issue of mathematically modeling color appearance, initiated in 1906 by J. von Kries, continued to be and still is addressed, resulting in several appearance models [4]. An important current example of a color appearance model is the CIECAM02 model first published in 2004 [5].
In Fig. 80.1, two “Mondrian” images are displayed, each one having three projectors with their output individually controllable in regard to spectral power and luminance. On the left is a telescopic photometer with which the reflected light from each area of the images can be separately measured, with the results projected onto the scale above the images. Illuminating radiation can be adjusted for different color fields so that the reflected light is identical, even though the appearance of the fields varies broadly, such as white, red, blue, or green [3], (reproduced with permission).