In terms of raw data, the earth observation industry is undeniably exploding. Investments in freely available data from satellite constellations like MODIS, Landsat, and Sentinel have democratized access to timely satellite imagery of the entire globe (albeit at a lower resolution than you're accustomed to seeing on Google Maps). Meanwhile, cloud providers like AWS and Google Cloud have gone so far as to store satellite data for free, further accelerating global usage of these images. The trouble, naturally, is that interpreting the content of satellite imagery is not an easy task. In the field of remote sensing, researchers have been applying algorithmic techniques to the challenge of earth imagery interpretation for over 70 years.

In this paper, we investigate an application of image generation for river satellite imagery. Specifically, we propose a generative adversarial network (GAN) model capable of generating high-resolution and realistic river images that can be used to support models in surface water estimation, river meandering, wetland loss and other hydrological research studies. First, we summarized an augmented, diverse repository of overhead river images to be used in training. Second, we incorporate the Progressive Growing GAN (PGGAN), a network architecture that iteratively trains smaller-resolution GANs to gradually build up to a very high resolution, to generate 256x256 river satellite imagery. With conventional GAN architectures, difficulties soon arise in terms of exponential increase of training time and vanishing/exploding gradient issues, which the PGGAN implementation seems to significantly reduce. Our preliminary results show great promise in capturing the detail of river flow and green areas present in river satellite images that can be used for supporting hydroinformatics studies.

Multi-spectral satellite imaging sensors acquire various spectral band images such as red (R), green (G), blue (B), near-infrared (N), etc. Thanks to the unique spectroscopic property of each spectral band with respective to the objects on the ground, multi-spectral satellite imagery can be used for various geological survey applications. Unfortunately, image artifacts from imaging sensor noises often affect the quality of scenes and have negative impacts on the applications of satellite imagery. Recently, deep learning approaches have been extensively explored for the removal of noises in satellite imagery. Most deep learning denoising methods, however, follow a supervised learning scheme, which requires matched noisy image and clean image pairs that are difficult to collect in real situations. In this paper, we propose a novel unsupervised multispectral denoising method for satellite imagery using wavelet subband cycle-consistent adversarial network (WavCycleGAN). The proposed method is based on unsupervised learning scheme using adversarial loss and cycle-consistency loss to overcome the lack of paired data. Moreover, in contrast to the standard image domain cycleGAN, we introduce a wavelet subband domain learning scheme for effective denoising without sacrificing high frequency components such as edges and detail information. Experimental results for the removal of vertical stripe and wave noises in satellite imaging sensors demonstrate that the proposed method effectively removes noises and preserves important high frequency features of satellite images.

Generative deep learning has sparked a new wave of Super-Resolution (SR) algorithms that enhance single images with impressive aesthetic results, albeit with imaginary details. Multi-frame Super-Resolution (MFSR) offers a more grounded approach to the ill-posed problem, by conditioning on multiple low-resolution views. This is important for satellite monitoring of human impact on the planet -- from deforestation, to human rights violations -- that depend on reliable imagery. To this end, we present HighRes-net, the first deep learning approach to MFSR that learns its sub-tasks in an end-to-end fashion: (i) co-registration, (ii) fusion, (iii) up-sampling, and (iv) registration-at-the-loss. Co-registration of low-resolution views is learned implicitly through a reference-frame channel, with no explicit registration mechanism. We learn a global fusion operator that is applied recursively on an arbitrary number of low-resolution pairs. We introduce a registered loss, by learning to align the SR output to a ground-truth through ShiftNet. We show that by learning deep representations of multiple views, we can super-resolve low-resolution signals and enhance Earth Observation data at scale. Our approach recently topped the European Space Agency's MFSR competition on real-world satellite imagery.

These high-level feature representations are more powerful and robust in typical visual tasks. In the intelligent interpretation of remote sensing images, the automatic target detection (or recognition) and high-resolution remotely sensed image classification are two hot topics, and both of these two tasks are carried out by first computing the low-level features in the raw images. For different kinds of remote sensing images (e.g., SAR images and hyperspectral images), the corresponding specific feature representations are available. Through applying deep learning methods, we are free of these handcrafted low-level features and can automatically learn mid-level and higher-level ones from a large amount of unlabeled raw samples beyond the types and domains of remote sensing images. Deep leaning methods can undoubtedly offer better feature representations for the related remote sensing task, and there is a bright prospect of seeing more and more researchers dedicated to learning better features for the target detection and scene classification tasks by utilizing deep learning methods appropriately. This special issue concentrates on the research in new methods, algorithms, and architectures of deep learning to handle the practical challenges in remote sensing image processing. The papers in this issue can be roughly organized into three main categories: (a) remote sensing imagery classification, (b) change detection of multitemporal remote sensing images, and (c) fusion of diverse types of images.

--One problem found when working with satellite images is the radiometric variations across the image and different images. Intending to improve remote sensing models for the classification of burnt areas, we set two objectives. The first is to understand the relationship between feature spaces and the predictive ability of the models, allowing us to explain the differences between learning and generalization when training and testing in different datasets. We find that training on datasets built from more than one image provides models that generalize better . These results are explained by visualizing the dispersion of values on the feature space. The second objective is to evolve hyper-features that improve the performance of different classifiers on a variety of test sets. We find the hyper-features to be beneficial, and obtain the best models with XGBoost, even if the hyper-features are optimized for a different method. Deforestation has serious implications on biodiversity, on rural communities that depend on forests for survival, and on greenhouse gas emissions that drive the global climate. The machine learning (ML) community can help by providing predictive models that, after learning from a small sample of an image, can automatically classify the whole image. Although previous ML work in forest monitoring has shown good results, the predictive models are often applied on the same location where they were learnt, i.e., the models are trained and tested in samples from the same dataset (e.g., [1]) or time series from the same area (e.g., [2]).

Traditional online maps, widely used on Internet such as Google map and Baidu map, are rendered from vector data. Timely updating online maps from vector data, of which the generating is time-consuming, is a difficult mission. It is a shortcut to generate online maps in time from remote sensing images, which can be acquired timely without vector data. However, this mission used to be challenging or even impossible. Inspired by image-to-image translation (img2img) techniques based on generative adversarial network (GAN), we propose a semi-supervised structure-augmented online map GAN (S$^{2}$OMGAN) model to generate online maps directly from remote sensing images. In this model, we designed a semi-supervised learning strategy to pre-train S$^{2}$OMGAN on rich unpaired samples and finetune it on limited paired samples in reality. We also designed image gradient L1 loss and image gradient structure loss to generate an online map with global topological relationship and detailed edge curves of objects, which are important in cartography. Moreover, we propose edge structural similarity index (ESSI) as a metric to evaluate the quality of topological consistency between generated online maps and ground truths. Experimental results present that S$^{2}$OMGAN outperforms state-of-the-art (SOTA) works according to mean squared error, structural similarity index and ESSI. Also, S$^{2}$OMGAN wins more approval than SOTA in the human perceptual test on visual realism of cartography. Our work shows that S$^{2}$OMGAN is potentially a new paradigm to produce online maps. Our implementation of the S$^{2}$OMGAN is available at \url{https://github.com/imcsq/S2OMGAN}.

Much of the research literature on landscape change has focused on aerial imagery or satellite imagery from remote sensing sources. This is understandable if one were to look at wide areas. However, ground-level photographs also have great value to demonstrate landscape change from a human eye-level perspective. Combing such ground-level photographs with fieldwork and spatial analysis provides the possibility to assess human-environmental factors that have led to sometimes drastic change in the landscape. Yellowstone has become one of the most popular national parks in the United States.

Researchers from Facebook and MIT Labs have proposed a new methodology that uses machine learning and satellite imagery to generate street addresses in areas of the world where individual buildings don't have a unique address. The methodology divides the street addressing into two processes. The first process is segmentation. During segmentation, road pixels are identified using deep learning from 0.5 meter resolution satellite images. The second part of segmentation involves developing the road network from these identified pixels.

Cyber-attack Automated Unconventional Sensor Environment (CAUSE), applies AI/ML-based models to develop novel, automated methods for event-based detection and prediction of cyber-attacks significantly earlier than existing approaches. Forecasting cyber-attack events with actionable details advances the state-of-the-art by enabling threat-specific cyber incident response and defense measures; Creation of Operationally Realistic 3D Environment (CORE3D), uses machine learning and deep learning techniques to develop methods for the construction of a fully automated high fidelity 3D model of the world using remote sensing data; Deep Intermodal Video Analytics (DIVA), leverages machine learning techniques to develop robust automatic activity detection in streaming video across multiple cameras; Finding Engineering-Linked Indicators (FELIX), uses AI for detection of engineering signatures across multiple biological organisms. The goal is to distinguish natural organisms from those that have been engineered; Functional Map of the World Challenge, developed algorithms that would quickly and accurately classify 63 classes of buildings and regions in satellite imagery. All the top participants used various forms of deep learning; Functional Genomic and Computational Assessment of Threats (Fun GCAT), develops AI/ML-based approaches to learn and classify genetic (e.g., DNA) sequence data by genetic taxonomy, sequence function, and threat potential; Mercury Challenge, asked challenge participants to make use of AI/ML approaches to forecast a variety of political events in the Middle East and North Africa region, such as non-violent civil unrest and military activity; Machine Intelligence from Cortical Networks (MICrONS), aims to revolutionize machine learning by reverse-engineering the algorithms of the brain. The program is expressly designed as a dialogue between data science and neuroscience; Machine Translation for English Retrieval of Information in Any Language (MATERIAL), develops machine learning methods to identify foreign language information from speech and text relevant to English queries, and providing evidence of relevance of the retrieved information in English in a meaningful way.