A. Mansouri (Norway), J.Y. Hardeberg (France), F.S. Marzani, and P. Gouton (Norway)
Image Acquisition, Multispectral System, Deblurring, Optical Calibration.
In multispectral imagery, the effectiveness of the analysis and spectrum reconstruction depend heavily on the acquired channel images. A multispectral system based on interference filters suffers typically from some blurring on its channel-images. In this paper a new method for the optical calibration by deblurring the channel-images of such multispectral system is described. First, we analyze the cause of the optical blur and we model it, we further establish an acquisition protocol which offers the best compromise between the acquisition parameters and the global reduction of the blur. We then describe an algorithm to compute the optimal degradation parameter for each channel-image. We finally perform the restoration with the computed parameters using an adapted regularized Wiener filtering. This methodology is evaluated on real images; results are illustrated and commented.
Important Links:
Go Back
