QuickBird remote sensing image denoising based on CNN

Abstract

Image denoising is the first pre-processing step in analyzing and understanding images, and is crucial to acquire the high-quality image products. QuickBird, launched in October 18, 2001, has spatial resolution with 0.61m in panchromatic mode, and can be used in various fields. In order to take the benefit of the high spatial resolution information of the QuickBird images, this paper proposed a method to remove the noise in QuickBird images using cellular neural netwok, CNN. Experimental results show that CNN-based approach performs effectively in removing the noise in QuickBird images. © Springer-Verlag Berlin Heidelberg 2012.