Nikon Corporation

Juping Yang

Yukio Nakagawa

Toru Fujii

Fuyuhiko Inoue

An image correction apparatus includes: an image obtaining unit that obtains image data from outside constituted of a plurality of pixels each having a value in a color component among a plurality of color components; an average value calculation unit that calculates average values of color component values corresponding to individual color components in the image data; and a correction unit that corrects the color component values of individual pixels to match the average values of the color component values corresponding to individual color components with at least one specific reference value.

Image correction apparatus and recording medium having image correction program

A histogram is prepared by calculating the pixel number ratio corresponding to each gradation in relation to the total number of pixels in image data obtained from the outside. A threshold value for the pixel number ratio corresponding to a given gradation that is determined through user specification is compared against the pixel number ratio corresponding to each gradation in the histogram. Based upon the results of the comparison, the histogram gradation width is corrected. For instance, the width representing the range of gradations with pixel number ratios smaller than the threshold value in the histogram is compressed. Then, the gradation width in the histogram is expanded to achieve the number of pre-compression gradations.

Image correction apparatus and recording medium for storing image correction program

Correlation values between a plurality of templates prepared beforehand and a picture-element data distribution as a pick-up result of an image are calculated, and the position of a maximum template whose correlation value is the maximum of the calculated correlation values is obtained. Subsequently, a curved surface function is calculated which fits the distribution of calculated correlation-values in positions near the maximum template's position, and the position of the picked-up image is calculated based on the curved surface function. As a result, the number of templates prepared beforehand to achieve desired accuracy in detecting a position and the number of times of calculating correlation-values can be reduced, and the image position can be quickly and accurately detected.

Growing community of inventors

Kawasaki, Japan

Juping Yang