Utilizing Multilevel Features for Cloud Detection on Satellite Imagery

Abstract

Cloud detection, which is defined as the pixel-wise binary classification, is significant in satellite imagery processing. In current remote sensing literature, cloud detection methods are linked to the relationships of imagery bands or based on simple image feature analysis. These methods, which only focus on low-level features, are not robust enough on the images with difficult land covers, for clouds share similar image features such as color and texture with the land covers. To solve the problem, in this paper, we propose a novel deep learning method for cloud detection on satellite imagery by utilizing multilevel image features with two major processes. The first process is to obtain the cloud probability map from the designed deep convolutional neural network, which concatenates deep neural network features from low-level to high-level. The second part of the method is to get refined cloud masks through a composite image filter technique, where the specific filter captures multilevel features of cloud structures and the surroundings of the input imagery. In the experiments, the proposed method achieves 85.38% intersection over union of cloud in the testing set which contains 100 Gaofen-1 wide field of view images and obtains satisfactory visual cloud masks, especially for those hard images. The experimental results show that utilizing multilevel features by the combination of the network with feature concatenation and the particular filter tackles the cloud detection problem with improved cloud masks.

Cloud detection methodologies: variants and development—a review

“…Authors [4] used the potential of deep learning methods for cloud detection and introduced the concept of patches with a convolution neural network, which allows the classifier to learn features such as spectral content and shape attributes of the clouds. Authors [5] proposed deep learning for cloud detection using multilevel image features of satellite imagery. They [5] performed two major operations; (i) obtain cloud probability map and (ii) get refined cloud masks through image filter.…”

“…Authors [5] proposed deep learning for cloud detection using multilevel image features of satellite imagery. They [5] performed two major operations; (i) obtain cloud probability map and (ii) get refined cloud masks through image filter. The proposed methods are very effective methods for cloud detection.…”

“…The detail about methods and models is given in Table 2. Various researchers [2][3][4][5][6] used artificial neural net with many variations such as support vector machine, fusing multi-scale convolution features, deep learning, decision tree, Bayesian classification, random forest-based methods, object-based convolution neural network, etc. Machine learning methods are very flexible and less complex since they simulate decisions on training data, but are not consistent, because model training depends on input data.…”

A Cloud Detection Method Using Convolutional Neural Network Based on Gabor Transform and Attention Mechanism with Dark Channel Subnet for Remote Sensing Image

“…From the perspective of color characteristics, we have created a new subnet, named Dark channel subnet, and used it as an auxiliary input of the spatial attention module to further eliminate the redundant information in the low-level feature map that assists the decoder. Cloud detection in GF-1 WFV imagery is a challenging task because of the unfixed radiometric calibration parameters and insufficient spectral information [36][37][38]. Therefore, we have evaluated the proposed algorithm by applying it to the GF-1 WFV data set [39].…”

Convolutional Neural Networks for On-Board Cloud Screening

“…CNNs usually consist of thousands of filters with millions of learnable parameters which are trained to detect semantic representations useful for a particular task and over a large amount of annotated images. Regarding cloud detection, recent studies have been carried out to make use of CNNs also in performing such detection task [28][29][30][31]. For instance, in [28], to perform cloud detection, satellite images first undergo a simple linear iterative clustering process in which homogeneous pixels are clustered into superpixels; then, a four-layer CNN is employed to extract features and, finally, two fully connected layers predict the superpixels class.…”