Speed-up generalized morphological component analysis technology used in remote sensing image inpainting application
- PDF / 11,134,369 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 77 Downloads / 177 Views
ORIGINAL PAPER
Speed-up generalized morphological component analysis technology used in remote sensing image inpainting application Chong Yu & Xiong Chen
Received: 13 August 2013 / Accepted: 3 January 2014 # Saudi Society for Geosciences 2014
Abstract In this paper, we introduced a remote sensing image inpainting method based on speed-up generalized morphological component analysis (SGMCA). Due to its capability to represent and separate the morphological diversities, generalized morphological component analysis (GMCA) algorithm is a state-of-the-art image inpainting method. SGMCA algorithm introduced in this paper can accelerate the iterative process of GMCA algorithm. By adding some more assumptions to GMCA algorithm, SGMCA algorithm is proven as a much faster algorithm which can handle very large-scale problems. Several experiments illustrate that SGMCA algorithm can recover the remote sensing images with different patterns of missing pixels. It is even hard to distinguish the original remote sensing image from the recovered image through visual effect. The peak signal to noise ratio and structural similarity indices explain why the salient visual effect is obtained, and confirm the marvelous inpainting capability of SGMCA algorithm. Quantitative analysis on time consumption proves that SGMCA algorithm can greatly improve the iterative speed of GMCA algorithm, indeed. Keywords Remote sensing image . Inpainting . Speed-up generalized morphological component analysis . Time consumption
C. Yu (*) : X. Chen (*) School of Information Science and Technology, Fudan University, Shanghai 200433, China e-mail: [email protected] e-mail: [email protected] C. Yu e-mail: [email protected]
Introduction Remote sensing images provide an unparalleled data source for many applications such as region classification, environmental monitoring, weather forecasting, land surface mapping, etc. However, in some conditions, the so-called bad pixels will exist in the remote sensing images. The bad pixel often has a behavior that is statistically different from its surrounding pixels. The causes of these bad pixels are various, may include nonresponse of detector, offset variations of detectors, calibration errors, image damage, and so on. In general, the bad pixels can be classified into two categories, i.e., the warm pixel and the dead pixel (Shen and Zhang 2009). The warm pixel is the pixel which is brighter or darker than the healthy pixel to some extent. The dead pixel is the pixel whose measurement does not have any correlation with the true scene that is being measured. We can view all these bad pixels as the missing data of the original perfect remote sensing images. Due to the existence of the missing data severely degrades the quality of remote sensing images, the recovery of these missing pixels is an ever-increasing demanding application. In the digital image processing field, the process of retrieving the missing data in an image is known as image inpainting application (Bugeau et al. 2010). Image inpainting app
Data Loading...
