Wavelet methods for noise reduction in the calibration of a CCD camera

What is it about?

A charge coupled camera is commonly used in computer vision applications. Accurate calibration of camera is of significance to get a good quality image. Noise due to uneven lighting conditions can significantly degrade the quality of noise. Noisy images further degrade the results of analysis. In this paper, thresholding methods using Bayes Shrink and Sure Shrink algorithms are investigated for denoising such images. The algorithms are applied to the images of calibration grid and the results are compared for both the methods. The resulting denoised images are assessed using SNR and MSE. It has been found that level dependent SURE soft algorithm results into best SNR and least value of SME.

Featured Image