An image restoration method based on cross-correlation

Abstract

Interference often appears between camera and object during imaging. In this paper, an image restoration method in video sequences based on cross-correlation is proposed. A submerged object with modulated light and interference is recorded by a high speed camera. The captured data and the modulated light are regarded as input signal and reference signal, respectively. A new pixel matrix is obtained and can be restored by filtering additive noise after executing a cross-correlation operation between the signals. Experimental results demonstrate that the method can eliminate interference and enhance image quality efficiently.