A Cloud Detection Method for Landsat 8 Images Based on PCANet

Abstract

Cloud detection for remote sensing images is often a necessary process, because cloud is widespread in optical remote sensing images and causes a lot of difficulty to many remote sensing activities, such as land cover monitoring, environmental monitoring and target recognizing. In this paper, a novel cloud detection method is proposed for multispectral remote sensing images from Landsat 8. Firstly, the color composite image of Bands 6, 3 and 2 is divided into superpixel sub-regions through Simple Linear Iterative Cluster (SLIC) method. Then, a two-step superpixel classification strategy is used to predict each superpixel as cloud or non-cloud. Thirdly, a fully connected Conditional Random Field (CRF) model is used to refine the cloud detection result, and accurate cloud borders are obtained. In the two-step superpixel classification strategy, the bright and thick cloud superpixels, as well as the obvious non-cloud superpixels, are firstly separated from potential cloud superpixels through a threshold function, which greatly speeds up the detection. The designed double-branch PCA Network (PCANet) architecture can extract the high-level information of cloud, then combined with a Support Vector Machine (SVM) classifier, the potential superpixels are correctly classified. Visual and quantitative comparison experiments are conducted on the Landsat 8 Cloud Cover Assessment (L8 CCA) dataset; the results indicate that our proposed method can accurately detect clouds under different conditions, which is more effective and robust than the compared state-of-the-art methods.

Cloud Detection: An Assessment Study from the ESA Round Robin Exercise for PROBA-V

“…All the above methods are pixel-wise, i.e., they treat pixels separately without taking account of spatial correlations among them or local features that are instead typical of images. Among the classification methods that use spatial features of images we mention [30] (Markov chains), [31] (Discriminant Analysis), [32] (relying on PCANet and SVM). We also mention the special case of Artificial Intelligence Deep Learning algorithms (e.g., [6,23]).…”

CloudFCN: Accurate and Robust Cloud Detection for Satellite Imagery with Deep Learning

“…Superpixels can be used to break scenes up into a set of regions that contain similar pixel values [42]. This allows regions to be treated like discrete objects, to be classified by a CNN [43,44] or bespoke architectures like PCANet [45] as cloud or non-cloud. However, this can cause issues, as clouds are very often amorphous and merge with one another, meaning the performance is highly sensitive to the initial superpixel construction itself, which cannot be updated during training.…”

Editorial to Special Issue “Multispectral Image Acquisition, Processing, and Analysis”

“…The second one, "A Cloud Detection Method for Landsat 8 Images Based on PCANet" by Y. Zi et al proposed a cloud detection method for Landsat 8 images based on several processing steps [10]. In a first step, the color composite images of Bands 6, 3, and 2 were divided into superpixel sub-regions through the Simple Linear Iterative Cluster (SLIC) method.…”