Semi-supervised DNN regression on airborne hyperspectral imagery for improved spatial soil properties prediction

The presence of clouds prevents optical satellite imaging systems from obtaining useful Earth observation information and negatively affects the processing and application of optical satellite images. Therefore, the detection of clouds and their accompanying shadows is an essential step in preprocessing optical satellite images and has emerged as a popular research topic in recent decades due to the interest in image time series analysis and remote sensing data mining. This review first analyzes the trends of the field, summarizes the progress and achievements in the cloud and cloud shadow detection methods in terms of features, algorithms, and validation of results, and then discusses existing problems, and provides our prospects at the end. We aim at identifying the emerging research trends and opportunities, while providing guidance for selecting the most suitable methods for coping with cloud contaminated problems faced by optical satellite images, an extremely important issue for remote sensing of cloudy and rainy areas. In the future, expected improvements in accuracy and generalizability, the combination of physical models and deep learning, as well as artificial intelligence and online big data processing platforms will be able to further promote processing efficiency and facilitate applications of image time series. In addition, this review collects the latest open-source tools and datasets for cloud and cloud shadow detection and launches an online project (Open Satellite Image Cloud Detection Resources, i.e., OpenSICDR) to share the latest research outputs ( https://github.com/dr-lizhiwei/OpenSICDR ). 

Cloud and cloud shadow detection for optical satellite imagery: Features, algorithms, validation, and prospects

Cloud detection, as a crucial step, has always been a hot topic in the field of optical remote sensing image processing. In this paper, we propose a deep learning cloud detection Network that is based on the Gabor transform and Attention modules with Dark channel subnet (NGAD). This network is based on the encoder-decoder framework. The information on texture is an important feature that is often used in traditional cloud detection methods. The NGAD enhances the attention of the network towards important texture features in the remote sensing images through the proposed Gabor feature extraction module. The channel attention module that is based on the larger scale features and spatial attention module that is based on the dark channel subnet have been introduced in NGAD. The channel attention module highlights the important information in a feature map from the channel dimensions, weakens the useless information, and helps the network to filter this information. A dark channel subnet with spatial attention module has been designed in order to further reduce the influence of the redundant information in the extracted features. By introducing a “dark channel”, the information in the feature map is reconstructed from the spatial dimension. The NGAD is validated while using the Gaofen-1 WFV imagery in four spectral bands. The experimental results show that the overall accuracy of NGAD reaches 97.42% and the false alarm rate reaches 2.22%. The efficiency of cloud detection using NGAD exceeds the state-of-art image segmentation network model and remote sensing image cloud detection model.

https://www.mdpi.com/2072-4292/12/19/3261/pdf

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

Deep network based on up and down blocks using wavelet transform and successive multi-scale spatial attention for cloud detection

Cloud detection (CD) with deep learning (DL) algorithms has been greatly developed in the applications involving the predictions of extreme weather and climate. In this review, the different conventional CD methods based on threshold, time differentiation, machine learning, and the intelligent algorithms including convolution neural networks (CNN), simple linear iterative clustering (SLIC), and semantic segmentation algorithms (SSAs) are introduced in detail, and, especially, the majority of CD publications employing the advanced and prevalent DL algorithms during the last decade are summarized and analyzed. First, in terms of the detection for different types of clouds, we meticulously compare the labels, scenarios and volumes of three popular CD datasets and put forward further the constructive recommendations about the cloud images selection, multi-bands images preprocessing, and truth labels combination for creating similar datasets. Subsequently, the structures, detection accuracies, and operating speeds of several different CD network models comprising the fully convolutional neural networks (FCNs), U-Net, SegNet, pyramid scene parsing network (PSPNet), as well as the associated derivatives are conducted elaborately to explore the comprehensively optimal performance for CD. In addition, aiming at expanding the applications in the resource-limited space-borne environment, we conclude the mainstream compression strategies of a number of different lightweight networks. Finally, the various limitations constraining the performance of the existing state-of-the-art DL CD methods and the corresponding development tendency are presented, which, expectantly, could be referential for the following researches.

A review on deep learning techniques for cloud detection methodologies and challenges

Recent developments in hyperspectral satellites have dramatically promoted the wide application of large-scale quantitative remote sensing. As an essential part of preprocessing, cloud detection is of great significance for subsequent quantitative analysis. For Gaofen-5 (GF-5) data producers, the daily cloud detection of hundreds of scenes is a challenging task. Traditional cloud detection methods cannot meet the strict demands of large-scale data production, especially for GF-5 satellites, which have massive data volumes. Deep learning technology, however, is able to perform cloud detection efficiently for massive repositories of satellite data and can even dramatically speed up processing by utilizing thumbnails. Inspired by the outstanding learning capability of convolutional neural networks (CNNs) for feature extraction, we propose a new dual-branch CNN architecture for cloud segmentation for GF-5 preview RGB images, termed a multiscale fusion gated network (MFGNet), which introduces pyramid pooling attention and spatial attention to extract both shallow and deep information. In addition, a new gated multilevel feature fusion module is also employed to fuse features at different depths and scales to generate pixelwise cloud segmentation results. The proposed model is extensively trained on hundreds of globally distributed GF-5 satellite images and compared with current mainstream CNN-based detection networks. The experimental results indicate that our proposed method has a higher F1 score (0.94) and fewer parameters (7.83 M) than the compared methods.

An Effective Cloud Detection Method for Gaofen-5 Images via Deep Learning

The needs of high-precision earth observation have led to the development of high-resolution and high-dimensionality RS data and greatly promoted the standard for processing and application of airborne hyperspectral images The varying brightness gradients of the airborne images cause problems in generating “seamless” mosaic for hyperspectral surveys, which severely affect the radiometric consistencies for subsequent analyses We present a semiempirical method to generate seamless mosaicking of multi-strip airborne hyperspectral images and introduce the model principle as well as the calculation process in detail The experimental results based on HyMap images in Lop Nor area show that this method can efficiently remove the illumination gradient in both single image and between multi-scene images Moreover, the MNF-transformed images and spectrum from overlap were chosen to assess the model; the results show that the Hapke-based model can be used to improve the airborne hyperspectral mosaicking effect and have great potential to subsequent quantitative applications

Seamless Mosaicking of Multi-strip Airborne Hyperspectral Images Based on Hapke Model

The invention discloses an aerial hyperspectral multi-flight-strip image radiation distortion correction method and relates to the technical field of hyperspectral data radiation correction and application. The method is used for correcting a radiation distortion phenomenon generated between multi-flight-strip images caused by factors such as observation angles and light conditions; on the basis of traditional radiation correction, not only the radiation distortion caused by a sensor is corrected, but also a conversion coefficient of the radiation correction under a specific observation condition is obtained through the relation between a phase angle and a radiance value; according to the conversion coefficient, the radiation distortion data caused by the influence of the observation angle, the atmosphere and the illumination condition are corrected, thereby effectively improving the radiation quality of the aerial hyperspectral data. In addition, the method in the invention can carryout radiation distortion correction on the multi-flight-strip aviation data, thereby realizing seamless mosaic of large-area images and providing high-precision data for later quantitative application.

Aerial hyperspectral multi-flight-strip image radiation distortion correction method

The invention discloses a hyperspectral mineral anomaly information extraction method based on a singularity theory, and relates to the technical field of hyperspectral remote sensing mineral recognition. According to the method, firstly a characteristic absorption position of a target mineral on a spectral image is obtained; then, the characteristic absorption position on the image is compared with a characteristic absorption position on a standard spectrum; if the characteristic absorption position on the image is consistent with the characteristic absorption position on the standard spectrum, and the matching degree between hyperspectral image data and standard spectrum data reaches a set threshold value, pixels of the spectral image are considered to contain the target mineral, so thatmineral information is extracted; and finally, according to the extracted mineral information, a singularity index is calculated pixel by pixel on the whole spectral image, and a mineral anomaly information distribution diagram is obtained. By adopting the method provided by the invention, the extraction of mineral weak information is enhanced; and mineral anomaly information with "mutation" characteristics is more closely related to a mineralization relation compared with a large amount of background mineral information, so that the mineral information can be extracted more easily.

Peng He

Papers

An Effective Cloud Detection Method for Gaofen-5 Images via Deep Learning

Seamless Mosaicking of Multi-strip Airborne Hyperspectral Images Based on Hapke Model

Aerial hyperspectral multi-flight-strip image radiation distortion correction method

Hyperspectral mineral anomaly information extraction method based on singularity theory