We are working with Sensefly to map the data from their latest mission on the Matterhorn. Here are three samples of the visualizations we have created from this data, each available as full source code to copy from.
SenseFly R&D engineer Adam Klaptocz launching a Sensefly eBee from the summit of the Matterhorn.
Flying directly from the summit of the Matterhorn using multiple Sensefly eBee fixed-wing drones and technology from Pix4d, Sensefly was able to capture over 10 square miles of 3d data and imagery in two hours. The Matterhorn flight was the most recent project by the Switzerland-based organization that seeks to promote the civilian use of drones. Adam Klaptocz, R&D engineer at Sensefly says:
“The goal of this particular mission was to push the limits of data gathering with drones in the most challenging mountainous conditions. Such a combination of high altitudes, steep rocky terrain and sheer size of dataset has simply not been done before with drones, we wanted to show that it was possible! We’re currently discussing with several agencies on how best they can exploit this dataset, including the Swiss Alpine Club, the Federal Office of Topography and the Zermatt tourism office. Beyond this particular mission, we’re hoping that the lessons that we’ve learned here in the Alps can be applied in similar hard to reach places, whether for mapping landslides to aid first responders or tracking receding glaciers to study climate change.”
SenseFly video of the recent Matterhorn mission.
This is just a first step, we’re excited about working with Sensefly and future possibilities of providing services for high resolution UAV-sourced data. Using all open source technologies to process the data we built this interactive 3d model, as color relief and as an image mosaic:
To visualize the intricacies of the terrain, we have created an interactive 3d point cloud - click on the image to explore it (requires browser with WebGL support). We used LASTools to convert the source LAZ data (compressed point-cloud format) to .asc. For displaying it on the web, we downsampled the point cloud and used WebGl via XB-PointStream to render each x,y,z point.
As a contextual layer, flat color relief maps are often more appropriate than 3d models. For this visualization, we used TileMill and GDAL to render a tiled map of the source data. Note the interactivity layer generated in TileMill, revealing elevation data on mouse over.