Cloud is one of the most common influences in remote sensing imagery. Because of cloud interference, much
important and useful information covered by cloud cannot be recovered well. How to detect and remove cloud
is an important issue for wide application of remote sensing data. A novel and effective method is proposed in
this paper to detect cloud in remote sensing image using light transmittance. Light transmittance is employed
to detect the cloud and also determine its corresponding thickness distribution. First, a cloud optical model is
defined based on the airlight-albedo model. Second, preliminary cloud light transmittance is estimated using dark
channel prior and then refine the result through guided filtering algorithm. Finally, we use light transmittance
to detect cloud region by thresholding and obtain detailed information about the distribution of cloud thickness
through mapping light transmittance of the cloud region into a gray image. Our method has been tested on real
remote sensing images with clouds. Compared with the existing methods, experimental results have proved the
better efficiency of our method in cloud detection.
SAR image despeckling is an active research area in image processing due to its importance in improving the quality of image for object detection and classification.In this paper, a new approach is proposed for multiplicative noise in SAR image removal based on nonlocal sparse representation by dictionary learning and collaborative filtering. First, a image is divided into many patches, and then a cluster is formed by clustering log-similar image patches using Fuzzy C-means (FCM). For each cluster, an over-complete dictionary is computed using the K-SVD method that iteratively updates the dictionary and the sparse coefficients. The patches belonging to the same cluster are then reconstructed by a sparse combination of the corresponding dictionary atoms. The reconstructed patches are finally collaboratively aggregated to build the denoised image. The experimental results show that the proposed method achieves much better results than many state-of-the-art algorithms in terms of both objective evaluation index (PSNR and ENL) and subjective visual perception.