The Color Calibration Of Image Or Video Sequence

Color values in an image are related to image radiance by a nonlinear function. That is, the color values of photos captured under different lighting, by different camera, or using different exposure time do not keep linear relationship. This causes problems for a number of algorithms in compute vision. This paper presents a new method for calibrating those photos to make color consistency. The main idea of this paper employs the brightness transfer function, which is gotten by the calculated response function and photo exposure time, to calibrate photos. Our method can process two-photo calibration. Moreover, we can use different calibration methods to deal with photos either with or without exposure time. At last, the brightness transfer function is optimized for keeping white balance, which makes each channel of calibrated photo neither over-exposure nor too dark. Our method can also process long video sequence calibration by reducing cumulative error.Based on the prior work, we propose a method to detect different lighting range in image, and calibrate the color in different regions using different brightness transfer functions. The application demonstrates our work on color calibration, and also improves traditional color calibration method, which just calibrate the photos using only one brightness transfer function.