White balance method for reducing the level of dark noise in interference images
The paper considers the solution of an urgent scientific and technical problem of reducing the level of dark noise in interference images. These pictures are recorded using optical cameras included in the laser interferometer, built according to the Fizeau system, and are used in this case to control the shape and quality of the lenses. For an effective analysis of the interference pattern by the operator, the observed image should be as high-contrast as possible. Lens contamination, as well as limited sensitivity and the inherent noise of the cameras used, lead to significant noise levels in the recorded images. There are a large number of different noise reduction methods, but most of them are not applicable in this case, which requires the development of a specialized algorithm that can be integrated into the camera driver and should be characterized by high performance and no frame loss when processing streaming data. The algorithm presented in the paper was applied for white balance, it was used to adjust the histogram of brightness of each color, while the bright points in the image remained bright and the dark ones remained dark, unlike other common algorithms. The paper considers a comparison of the proposed algorithm with other white balance methods based on the analysis of more than 2500 interferogram images. The effect of heating the optical camera matrix on the operation of all the algorithms considered was also analyzed. The main result of the work carried out is the development of an original white balance algorithm that is optimal in terms of speed and performance, demonstrating a reduction in the influence of dark noise, as well as an increase in image contrast, which significantly improves their visual perception during subsequent analysis of interferograms by the operator.
Authors: D. V. Dorofeev, D. K. Kostrin
Direction: Physics
Keywords: dark noise, interference image, optical camera, laser interferometer, lens, mathematical algorithm, histogram, white balance, contrast
View full article