Stripe Noise Removal from Remote Sensing Images

Remote sensing images are in many domains used, including geographic, military, urban planning and environmental surveillance, but they are somewhat limiting their application due of additional stripe noise. Clear images from stripe pictures may be easily predicted in most existing stream noise reduction algorithms without considering the underlying characteristics of strip noise that cause the structure to be destroyed. Thus a new strategy was suggested in this study from the point of view of the image breakdown. The inherent qualities of strip noise and image properties are taken into consideration. The suggested methodology combines regularization, group regulation and television regularization in a framework for picture decomposition, into a (TV). The first two terms are used to execute stripe noise qualities through statistical analyses and regularization of the TV should evaluate the portions of the smooth structures of the stripe-free image. In addition, an effective alternating minimization methodology is proposed to solve the picture decomposition model.