Light, which is reflected from an object, varies with the type of illuminant used. Nevertheless, the color of an object appears to be approximately constant to a human observer. The ability to compute color constant descriptors from reflected light, is called color constancy. In order to solve the problem of color constancy, some assumptions have to be made. One frequently made assumption is that on average, the world is gray. We address the problem of color constancy and focus on the use of space average color for color constancy. Instead of computing global space average color we suggest to use local space average color as the illuminant frequently varies across an image. We discuss several different methods on how to compute local space average color. The performance of the different algorithms as well as related algorithms is evaluated on an object recognition task. Algorithms based on local space average color are simple, yet highly effective for the problem of color constancy. Such algorithms are particularly suited for object recognition tasks.
Objects retain their color in spite of changes in the wavelength and energy composition of the light they reflect. This phenomenon is called color constancy and plays an important role in computer vision research. We have devised a parallel algorithm for color constancy. The algorithm runs on a two dimensional grid of processors each of which can exchange information with its four neighboring processors. Each processor calculates local average color. This information is then used to estimate the reflectances of the object. The algorithm was tested on several images of everyday objects. The algorithm also works for scenes where the illuminant changes smoothly over the image.
Abstract. The human visual system is able to correctly determine the color of objects in view irrespective of the illuminant. This ability to compute color constant descriptors is known as color constancy. We have developed a parallel algorithm for color constancy. This algorithm is based on the computation of local space average color using a grid of processing elements. We have one processing element per image pixel. Each processing element has access to the data stored in neighboring elements. Local space average color is used to shift the color of the input pixel in the direction of the gray vector. The computations are executed inside the unit color cube. The color of the input pixel as well as local space average color is simply a vector inside this Euclidean space. We compute the component of local space average color which is orthogonal to the gray vector. This component is subtracted from the color of the input pixel to compute a color corrected image. Before performing the color correction step we can also normalize both colors. In this case, the resulting color is rescaled to the original intensity of the input color such that the image brightness remains unchanged.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.