Radiation Film Calibration In Three-Dimensional RGB Color Space
Abstract
Purpose
To propose a novel calibration equation and computational algorithms for film dosimetry using all three RGB color data.
Methods
Radiation changes the optical properties of radiochromic films. The changes are quantized by taking photographs. In the proposed method, a single spline calibration equation was obtained in the 3D RGB (red, green, and blue) color space from the pixel values of films irradiated to various known radiation dosages and scanned by a color flatbed scanner. To estimate the unknown dose delivered to a new film, we measured a set of RGB values that is point S in the 3D RGB space. Then, a corresponding point P, which was the closest to point S on the calibration curve, was identified. The dose at P was easily estimated using the calibration equation. To evaluate the method's feasibility, we used ingenious radiochromic films, which were scanned with a flatbed color scanner to obtain RGB color data of the irradiated films. Dose estimation accuracy was evaluated through computer simulation.
Results
The calibration curve of the test films exhibited a snake-like trajectory, indicating the true complexity of the spectral changes in film characteristics as a function of absorbed dose. A single dose was obtained with the new method for a given RGB value of a film irradiated by a fixed dose. In contrast, the standard calibration method yielded three distinct estimated doses for the three colors. The new method accurately estimated the dose across the entire calibration dose range, even when noise was added. However, the standard calibration method failed to estimate doses other than the calibration doses accurately.
Conclusion
A novel calibration equation formulated in the three-color 3D space was proposed to achieve more accurate film calibration. The method's performance is being tested for radiochromic film dosimetry.