Terra MODIS band 5 Stripe noise detection and correction using MAP-based algorithm

• Published in: 2011 International Conference on Remote Sensing, Environment and Transportation Engineering pp. 8612 - 8615
• Main Authors: Rongbin Wang, Chao Zeng, Pingxiang Li, Huanfeng Shen
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2011
• Subjects: band 5; destriping; detection; Geology; maximum a posteriori (MAP); MODIS; Noise; Noise measurement; Pixel; Remote sensing; Satellites; stripe noise; Terra
• ISBN: 9781424491728; 142449172X
• DOI: 10.1109/RSETE.2011.5964181

Since 1 of the 20 detectors in Terra MODIS band 5 (1.230~1.250 μm) are noisy, there are sharp and repetitive stripes over the entire image. As for MODIS geolocated data, the stripes are irregular and sometimes uncontinuous, it brings a difficult problem to the image retrieving process. This paper presents a detection method to extract the stripe noise, and a maximum a posteriori (MAP) based algorithm to correct the contaminated pixels. In the MAP method, the likelihood probability density function (PDF) is proposed based on a linear image noise model, and a Huber-Markov model is employed as the prior PDF. The gradient descent optimization method is used to receive the destriped image. The proposed algorithm has been tested using a Terra MODIS band 5 geolocated image. The experimental results demonstrate that the proposed algorithm performs well.