Visual search tool for satellite imagery

Posted to Apps  |  Tags: ,  |  Nathan Yau

Terrapattern is a fun prototype that lets you search satellite imagery simply by clicking on a map. For example, you can click on a tennis court, and through machine learning, the application looks for similar areas.

Terrapattern uses a deep convolutional neural network (DCNN), based on the ResNet (“Residual Network”) architecture developed by Kaiming He et al. We trained a 34-layer DCNN using hundreds of thousands of satellite images labeled in OpenStreetMap, teaching the neural network to predict the category of a place from a satellite photo. In the process, our network learned which high-level visual features (and combinations of those features) are important for the classification of satellite imagery.

Give it a try.

Favorites

Real Chart Rules to Follow

There are rules—usually for specific chart types meant to be read in a specific way—that you shouldn’t break. When they are, everyone loses. This is that small handful.

Reviving the Statistical Atlas of the United States with New Data

Due to budget cuts, there is no plan for an updated atlas. So I recreated the original 1870 Atlas using today’s publicly available data.

Shifting Incomes for American Jobs

For various occupations, the difference between the person who makes the most and the one who makes the least can be significant.

Marrying Age

People get married at various ages, but there are definite trends that vary across demographic groups. What do these trends look like?