Deep learning usually requires big data, with respect to both volume and variety. However, most remote sensing applications only have limited training data, of which a small subset is labeled. Herein, we review three state-of-the-art approaches in deep learning to combat this challenge. The first topic is transfer learning, in which some aspects of one domain, e.g., features, are transferred to another domain. The next is unsupervised learning, e.g., autoencoders, which operate on unlabeled data. The last is generative adversarial networks, which can generate realistic looking data that can fool the likes of both a deep learning network and human. The aim of this article is to raise awareness of this dilemma, to direct the reader to existing work and to highlight current gaps that need solving.

Recently, a variety of approaches has been enriching the field of Remote Sensing (RS) image processing and analysis. Unfortunately, existing methods remain limited faced to the rich spatio-spectral content of today's large datasets. It would seem intriguing to resort to Deep Learning (DL) based approaches at this stage with regards to their ability to offer accurate semantic interpretation of the data. However, the specificity introduced by the coexistence of spectral and spatial content in the RS datasets widens the scope of the challenges presented to adapt DL methods to these contexts. Therefore, the aim of this paper is firstly to explore the performance of DL architectures for the RS hyperspectral dataset classification and secondly to introduce a new three-dimensional DL approach that enables a joint spectral and spatial information process. A set of three-dimensional schemes is proposed and evaluated. Experimental results based on well knownhyperspectral datasets demonstrate that the proposed method is able to achieve a better classification rate than state of the art methods with lower computational costs.

The UN Sustainable Development Goals allude to the importance of infrastructure quality in three of its seventeen goals. However, monitoring infrastructure quality in developing regions remains prohibitively expensive and impedes efforts to measure progress toward these goals. To this end, we investigate the use of widely available remote sensing data for the prediction of infrastructure quality in Africa. We train a convolutional neural network to predict ground truth labels from the Afrobarometer Round 6 survey using Landsat 8 and Sentinel 1 satellite imagery. Our best models predict infrastructure quality with AUROC scores of 0.881 on Electricity, 0.862 on Sewerage, 0.739 on Piped Water, and 0.786 on Roads using Landsat 8. These performances are significantly better than models that leverage OpenStreetMap or nighttime light intensity on the same tasks. We also demonstrate that our trained model can accurately make predictions in an unseen country after fine-tuning on a small sample of images. Furthermore, the model can be deployed in regions with limited samples to predict infrastructure outcomes with higher performance than nearest neighbor spatial interpolation.

New technologies are generating far more information than ever before to help scientists assess and predict the health and behavior of species and ecosystems, as well as the threats they face. These include cryptic cameras, acoustic sensors, satellite imagery and citizen science apps. Now, researchers and conservation practitioners analyzing large data sets are exploring artificial intelligence, or AI--the ability of a machine or a computer program to think and learn--to help them process, analyze and interpret data to monitor ecosystems and predict results. Computer systems already exist that can host huge amounts of data, use AI with increasingly "smart" algorithms to classify data from the various types of sensors used by scientists, apply modeling results to create reproducible code, and create user interfaces to allow people to monitor natural systems and make predictions with high accuracy. By training computer algorithms with a subset of available data, machines can now learn what they should do for a given challenge--such as classify photos by the species found in them, identify areas of a satellite image containing water or intact forest, or translate speech from one language to another --based on human feedback and data collected from previous experience.

AI is the only thing that can let us see the whole world at once. Not recording it, but seeing it – creating a global real-time database of the world," says Stuart Russell, UC-Berkeley, lead of the AI for Good breakthrough team on AI and satellite imagery. The 2nd AI for Good Global Summit connected AI innovators with public and private-sector decision-makers. Four breakthrough teams – looking at satellite imagery, healthcare, smart cities, and trust in AI – set out to propose AI strategies and supporting projects to advance sustainable development. Teams were guided in this endeavour by an expert audience representing government, industry, academia and civil society.

Remote sensing lacks methods like the word vector representations and pre-trained networks that significantly boost performance across a wide range of natural language and computer vision tasks. To fill this gap, we introduce Tile2Vec, an unsupervised representation learning algorithm that extends the distributional hypothesis from natural language -- words appearing in similar contexts tend to have similar meanings -- to geospatial data. We demonstrate empirically that Tile2Vec learns semantically meaningful representations on three datasets. Our learned representations significantly improve performance in downstream classification tasks and similarly to word vectors, visual analogies can be obtained by simple arithmetic in the latent space.

Advances in sensor technology, cloud computing, and machine learning (ML) continue to converge to accelerate innovation in the field of remote sensing. However, fundamental tools and technologies still need to be developed to drive further breakthroughs and to ensure that the Global Development Community (GDC) reaps the same benefits that the commercial marketplace is experiencing. This process requires us to take a collaborative approach. Data collaborative innovation -- that is, a group of actors from different data domains working together toward common goals -- might hold the key to finding solutions for some of the global challenges that the world faces. That is why Radiant.Earth is investing in new technologies such as Cloud Optimized GeoTiffs, Spatial Temporal Asset Catalogues (STAC), and ML. Our approach to advance ML for global development begins with creating open libraries of labeled images and algorithms.

When you have large data sets of satellite or drone imagery that you have to process to create predictions, classification, or clustering – machine learning (ML) is the way to go. Indeed, ML has started to play a critical role in spatial problem solving given its potential to rapidly scan and unlock insights from petabytes of pixels obtained from hundreds of satellites and drones that are constantly orbiting earth. Orbital Insight, for example, applies machine learning and computer vision technologies to interpret data at petabyte scale to make it actionable for better business and policy decisions. The California based company has developed a powerful method that blends satellite imagery, deep learning, and data science for monitoring fresh-water supplies at local and global scale. Good news is that you don't have to be Orbital or in California,USA to also deploy machine learning. The proliferation of opensource platforms has made machine learning a lot easier to implement both on single personal computers and at scale, and in most popular programming or scripting languages.

Bangalore: In 2016, The Times of India reported that 253 farmers from Lalkheda village in Khargone district of Madhya Pradesh received an average of Rs 2.83 (4 US cents) each as insurance payouts for the loss of their soybean crop under the Pradhan Mantri Fasal Bima Yojana (the flagship prime minister's crop-insurance scheme). The insurer blamed the farmers for taking insufficient cover. The story got 29-year-old Prateep Basu, then an engineer with the Indian Space Research Organisation (ISRO), thinking. How was it possible, he wondered, that a country which had been using remote-sensing technology for decades could fail to use it for accurate crop forecasting when the lives of so many millions crucially depended on such information? The four of them went on to form SatSure, a data-analytics startup with "a mission to evolve crop insurance products and provide an accurate risk assessment of crop yield by integrating climatic variables with geospatial and economic datasets."

Deep learning continues to push state-of-the-art performance for the semantic segmentation of color (i.e., RGB) imagery; however, the lack of annotated data for many remote sensing sensors (i.e. hyperspectral imagery (HSI)) prevents researchers from taking advantage of this recent success. Since generating sensor specific datasets is time intensive and cost prohibitive, remote sensing researchers have embraced deep unsupervised feature extraction. Although these methods have pushed state-of-the-art performance on current HSI benchmarks, many of these tools are not readily accessible to many researchers. In this letter, we introduce a software pipeline, which we call EarthMapper, for the semantic segmentation of non-RGB remote sensing imagery. It includes self-taught spatial-spectral feature extraction, various standard and deep learning classifiers, and undirected graphical models for post-processing. We evaluated EarthMapper on the Indian Pines and Pavia University datasets and have released this code for public use.