Abstract: A computing device and a method for correcting color distortion are provided, in particular for correcting purple flare. A digital image has an array of pixels each represented by a green (G), a red (R) and a blue (B) primary color, and for each pixel, the computing device establishes a minimum value between the R and B intensity values, and compares the minimum value and the G intensity value. The original R, G and B intensity values are maintained if the G intensity value is equal or greater than the minimum value. The R and B intensity values are attenuated by an adjustment value if the G intensity value is less than the minimum value. The adjustment value is calculated as the difference of the minimum value and the G intensity value.

Abstract: A computing device and a method for correcting color distortion are provided, in particular for correcting purple flare. A digital image has an array of pixels each represented by a green (G), a red (R) and a blue (B) primary color, and for each pixel, the computing device establishes a minimum value between the R and B intensity values, and compares the minimum value and the G intensity value. The original R, G and B intensity values are maintained if the G intensity value is equal or greater than the minimum value. The R and B intensity values are attenuated by an adjustment value if the G intensity value is less than the minimum value. The adjustment value is calculated as the difference of the minimum value and the G intensity value.

Abstract: Relative histograms compare occurrences in event channels for a multi-channel data set for determining comparisons therebetween. One or more relative histograms are formed using relating functions applied to channels, preferably including a quotient between a first reference channel and a second channel and between the reference channel and at least a third channel. Preferably the relating functions generate values where a ratio and its reciprocal are symmetrical about an identity value. More preferably a data set is presented in a two-dimensional histogram for establishing the deviation of significant counts from an adjustment point which is preferably at the identity value. In practice, the relative histograms can be applied to correct tint in the red, green and blue channels of color images.

Abstract: A method is provided for automatically adjusting the color of a digital image. In one aspect, an strength offset is determined to automatically remove gray from a distorted image. In an other aspect applied independently or after gray removal, scaling factors are automatically determined to adjust the image to conform to empirically determined optimal brightening of the image. Using a determination of color dot maximums and a function which ensures constraint within the dynamic range, scaling factors are established which correspond to the strength values of each dot maximum and which are applied to each of the dot's R,G and B to maintain ratios therebetween and thus maintain the true color. The function for establishing the scaling factors is interpolated from a nominal function and a target average strength for the image.

Abstract: A method is provided for enhancing a digital image without distortion of the color. The result is an adjusted image which preserves the essential color of each and every dot in the input digital image while varying the effective light gathering power—like a virtual flash. The image enhancement is performed in RGB color space and comprises determining the maximum strength of the R,G, and B of a dot's RGB triplet and similarly for all dots. The dot maximums are scaled through a scaling function which is constrained in domain and range to the system's dynamic range. The same scaling factor that is applied to a dot maximum is also applied to each of R,G and B in the triplet. Preferably, a continuous scaling function is provided which smoothly approaches the minimum and maximum of the system's dynamic range for providing an aesthetically pleasing enhancement while maintaining true color.