Mark Johnson wants to beat the United States Department of Agriculture at its own game: predicting yields of America's crops. The USDA puts boots on the ground, deploying hundreds of workers to survey thousands of farms a month ahead of the October corn harvest, America's biggest crop. Johnson's startup, Descartes Labs, has just 20 employees, and they never leave the office in Los Alamos, New Mexico. Instead, Descartes relies on 4 petabytes of satellite imaging data and a machine learning algorithm to figure out how healthy the corn crop is from space. Corn yield prediction is big business in the US. Billions of dollars are at stake along the ag supply chain each year as corn starts to come out of the ground in August.

Current remote sensing image classification problems have to deal with an unprecedented amount of heterogeneous and complex data sources. Upcoming missions will soon provide large data streams that will make land cover/use classification difficult. Machine learning classifiers can help at this, and many methods are currently available. A popular kernel classifier is the Gaussian process classifier (GPC), since it approaches the classification problem with a solid probabilistic treatment, thus yielding confidence intervals for the predictions as well as very competitive results to state-of-the-art neural networks and support vector machines. However, its computational cost is prohibitive for large scale applications, and constitutes the main obstacle precluding wide adoption. This paper tackles this problem by introducing two novel efficient methodologies for Gaussian Process (GP) classification. We first include the standard random Fourier features approximation into GPC, which largely decreases its computational cost and permits large scale remote sensing image classification. In addition, we propose a model which avoids randomly sampling a number of Fourier frequencies, and alternatively learns the optimal ones within a variational Bayes approach. The performance of the proposed methods is illustrated in complex problems of cloud detection from multispectral imagery and infrared sounding data. Excellent empirical results support the proposal in both computational cost and accuracy.

Deep convolutional neural networks (DCNNs) have been used to achieve state-of-the-art performance on many computer vision tasks (e.g., object recognition, object detection, semantic segmentation) thanks to a large repository of annotated image data. Large labeled datasets for other sensor modalities, e.g., multispectral imagery (MSI), are not available due to the large cost and manpower required. In this paper, we adapt state-of-the-art DCNN frameworks in computer vision for semantic segmentation for MSI imagery. To overcome label scarcity for MSI data, we substitute real MSI for generated synthetic MSI in order to initialize a DCNN framework. We evaluate our network initialization scheme on the new RIT-18 dataset that we present in this paper. This dataset contains very-high resolution MSI collected by an unmanned aircraft system. The models initialized with synthetic imagery were less prone to over-fitting and provide a state-of-the-art baseline for future work. Keywords: Deep learning, convolutional neural network, semantic segmentation, multispectral, unmanned aerial system, synthetic imagery 1. Introduction Semantic segmentation is the pixel-wise classification of an image, i.e., every pixel is assigned its own label.

Are you currently enrolled in either of my Core or Intermediate Spatial Data Analysis Courses? Or perhaps you have prior experience in GIS or tools like R and QGIS? You don't want to spend 100s and 1000s of dollars on buying commercial software for imagery analysis? The next step for you is to gain profIciency in satellite remote sensing data analysis. MY COURSE IS A HANDS ON TRAINING WITH REAL REMOTE SENSING DATA WITH OPEN SOURCE TOOLS!

High-resolution satellite imagery have been increasingly used on remote sensing classification problems. One of the main factors is the availability of this kind of data. Even though, very little effort has been placed on the zebra crossing classification problem. In this letter, crowdsourcing systems are exploited in order to enable the automatic acquisition and annotation of a large-scale satellite imagery database for crosswalks related tasks. Then, this dataset is used to train deep-learning-based models in order to accurately classify satellite images that contains or not zebra crossings. A novel dataset with more than 240,000 images from 3 continents, 9 countries and more than 20 cities was used in the experiments. Experimental results showed that freely available crowdsourcing data can be used to accurately (97.11%) train robust models to perform crosswalk classification on a global scale.

Watch the video below to understand smart processing of satellite imagery using deep learning.

Nigerian health officials won't have to rely on flawed, decade-old census data when they plan deliveries of the measles vaccine next year. Instead, they will have access to what may be the most detailed and up-to-date population map ever produced for a developing country. Created by the Bill & Melinda Gates Foundation in Seattle, Washington, and delivered to Nigerian officials on May 1, the map is based on a detailed analysis of buildings in satellite imagery and more than 2,000 on-the-ground neighbourhood surveys. It is one of several projects that are leveraging remote-sensing data and modern computer-learning algorithms to chart human settlements around the globe with unprecedented precision. Researchers hope that these data will enable better management of public health, infrastructure and natural resources--and improve planning for emergencies.

Facebook uses satellite imagery machine learning and AI to prepare maps for locating unconnected communities across the world. Maps tell us so much more than how to get from A to B, or where C is in relation to D. They can be tools of power and snapshots of history; they can give urban planners the information to plan infrastructure. After a disaster, population density and crisis maps help to direct aid and aid workers. Throughout time, different cultures and industries have produced radically different images of the world. Today there are more than 7 billion humans sprawling across Earth.

Want to know where all the wind and solar power supplies in the US are for some brilliant renewable-energy project? Or plot a round-the-world trip hitting every major soccer stadium along the way? It should be possible with a new tool that lets anyone scan the globe through AI "eyes" to instantly find satellite images of matching objects. Descartes Labs, a New Mexico startup that provides AI-driven analysis of satellite images to governments, academics and industry, on Tuesday released a public demo of its GeoVisual Search, a new type of search engine that combines satellite images of Earth with machine learning on a massive scale. The idea behind GeoVisual is pretty simple.

Agricultural monitoring, especially in developing countries, can help prevent famine and support humanitarian efforts. A central challenge is yield estimation, i.e., predicting crop yields before harvest. We introduce a scalable, accurate, and inexpensive method to predict crop yields using publicly available remote sensing data. Our approach improves existing techniques in three ways. First, we forego hand-crafted features traditionally used in the remote sensing community and propose an approach based on modern representation learning ideas. We also introduce a novel dimensionality reduction technique that allows us to train a Convolutional Neural Network or Long-short Term Memory network and automatically learn useful features even when labeled training data are scarce. Finally, we incorporate a Gaussian Process component to explicitly model the spatio-temporal structure of the data and further improve accuracy. We evaluate our approach on county-level soybean yield prediction in the U.S. and show that it outperforms competing techniques.