Global image covery and virtual 3D globes are booming. The two companies Microsoft and Google are competing in presenting the biggest and actual database of satellite, aerial and streetside images. As it is inevitable to use images from different sources for these kinds of projects, the topic multisensory image fusion becomes more important. The color and radiometry of images is one essential topic in the matter of image fusion technologies. The main interest is to get well balanced image mosaics out of different kind of images without seeing any color shifts or problems.

This paper will give a conceptual overview how to color characterize an aerial camera and with this be able to use all the advantages of standardized color spaces. Besides a radiometric calibration for detecting vignetting and dead pixels as well as sensor thresholds the camera will be characterized. Camera characterization describes a methodology to find appropriate parameters to change the device dependent RGB color space of a digital camera, also known as color gamut, into a device independent color space. The resulting parameters (depending on the method this can also be a LUT) are then used to define an ICC (International Color Consortium) profile. An ICC profile is a standard to describe the color space transformation from a device into a device independent color space as XYZ or LAB or vice versa from a color independent color space into an output device color space, as CMYK for printers or sRGB for LCD screens.

The transformation of a device color gamut into a device independent color space opens the possibilities to use defined methods for changing the colors of an image in a standardized way. Methods like color transformations, white balancing or gamut mappings help to improve the quality of colors. The standardized transformation back into a color space of an output device guarantees an output with the same colors (so far the color gamut of the device is similar) as the input device has recorded.

When characterizing sensors (as mobile or aerial digital cameras) all images recorded with these sensors can be transformed into the device independent color space and with this opens new possibilities in adjusting multiple images from multiple sensors.  In addition, the transformation back to an output color space is specified as long as the output device is also calibrated and defined with an ICC-profile.

This paper will present possibilities for creating ICC profiles for input devices as digital cameras. Methods for adjusting images will be evaluated to apply on multiple images from multiple sensors. The final results will lead to adjusted color images, ready for mosaicing. The main aim is to present methods and concepts in color theory which can be applied in the field of sensor fusion and remote sensing to improve the workflow for color adjustment.  

K. Barnard and B. Funt. “Camera Characterization for Color Research” Color Research and Application 27 (2002), 152-163.

International Color Consortium Specification. Tech. Rep. ICC.1:2004-10, International Color Consortium, 2004. Available online (http://www.color.org)

M.H. Kim, J.  Kautz. „Characterization for High Dynamic Range Imaging.”  EUROGRAPHICS 2008, Volume 27 (2008), Number 2

E. Reinhard, E.A. Khan, A.O. Akyuz. „ Color Imaging: Fundamentals and Applications.”  Wellesley: AK Peters, Ltd., 2008.

G. Sharma, M. J. Vrhel, and H. J. Trussell. “Color imaging for multimedia.” Proceedings of the IEEE, 86(6):1088, Jun. 1998

M. J. Vrhel, H. J. Trussell: “Color Device Calibration: A Mathematical Formulation.” IEEE Transactions on Image Processing 8:12 (1999), 1796-1806.