Determining color at an object point from multple images providing conflicting color information

摘要

A set of images of an object are captured from a set of lenses positioned at different angles. A common point is determined, which has a color value that differs among the images. Color at the point is mathematically expressed as a set of equations. The equations comprise a recorded color value for the point, an ambient contribution, a diffuse contribution, and a specular contribution. The ambient and diffuse contributions are set as equal across the equations. The specular contribution is determined, such as by quantifying the light sources and solving for the specular contribution using a Phong lighting model equation. True color for the point, which is based on the ambient and diffuse contribution, is determined by solving a set of simultaneous equations once the specular contribution is known. An image is created where the point has the true color as determined above.

主权项

1. A method comprising:
a computing device determining a common point of an object present in the plurality of images, which has a color value that differs among a plurality of images showing the object; the computing device mathematically expressing color value of the common point as a set of equations, one of the equations for each of the plurality of images; the computing device making an ambient contribution and a diffuse contribution for color as defined within the set of equations approximate equal across the set of equations; the computing device quantifying a specular contribution at the common point for at least a subset of the plurality of images; the computing device substituting the quantified specular contribution value into a corresponding variable and solving the simulations equations for a value of the ambient and diffuse contribution at the point; and the computing device establishing a color of the point as the ambient and diffuse contribution, wherein the established color is a value different from any value recorded for the point within any of the images.