Archinect - News

Artificial Intelligence helps mapping urban trees (all of them)

2019-01-07T14:19:00-05:00

To train the model, he identified known locations of tree canopy using lidar data and NAIP imagery over California. Using that as ground truth, the model was trained to classify which pixels contain trees in the corresponding satellite images. The result is a machine-learning model that has learned to identify trees just using four-band high-resolution (~1 meter) satellite or aerial imagery—no lidar required!

Former New York Times cartographer Tim Wallace describes how his current firm, Santa Fe-based Descartes Labs, has built a machine learning model to identify tree canopy from satellite imagery thus making accurate mapping of trees and urban forests far more accessible to cities worldwide.

San Francisco Open Forest Map tree inventory (point data) in comparison with the Descartes Labs tree canopy layer (image data). Image: Tim Wallace/Descartes Labs

"The ability to map tree canopy at a such a high resolution in areas that can’t be easily reached on foot would be helpful for utility companies to pinpoint encroachment issues—or for municipalities to find possible trouble spots beyond their official tree census (if they even have one)," writes Wallace. "But by zooming out to a city level, patterns in the tree canopy show off urban greenspace quirks. For example, unexpected tree deserts can be identified and neighborhoods that would most benefit from a surge of saplings revealed."

New York...

Tracking economic conditions with satellite imagery and shadows

2014-11-20T17:39:00-05:00

Proving that some market somewhere will find a value for anything, a company called Orbital Insight is now tracking "the shadows cast by half-finished Chinese buildings" as a possible indicator for where the country's economy might be headed.

Alvar Aalto gets a close look from Google's Cultural Institute

2014-08-28T15:53:00-04:00

Google's satellite imaging allows us to virtually tour remote or inaccessible locales the world over, and with recently improved resolutions and initiatives from the Google Cultural Institute, our gaze can go farther and more intimately into places we may never physically visit. Google's interest in maintaining this visual resource has implications as far-reaching as its imagery, but it's also a boon for architecture education, where access to imagery and certain spaces can be highly exclusive, and retaining a global (and historical) perspective is exceedingly important.

To this end, Google's Cultural Institute has partnered with the Alvar Aalto Foundation in order to make the famous Finnish architect's spaces (and of course his famous stool no. 60) available online in 360-degree panoramas, detailing every surface. The partnership's collection focuses on eight sites, including Aalto's studio, Säynätsalo Town Hall, and the House Kantola, as well as an exhibition at the Aalto Museum in...

New Google Satellites Could Be Able to See Your Face from the Sky

2014-08-14T20:23:00-04:00

Two months ago, after much lobbying by the biggest satellite company in North America, DigitalGlobe, the US government relaxed restrictions to allow for commercially available satellite imagery up to 25 cm resolution—twice as detailed as the previous limit of 50 cm [...] The extra sharp images from Worldview-3 will greatly increase the maps' level of detail to the point where it can make out 10-inch objects, which means Google will soon be able to see “manholes and mailboxes” [...]

DigitalGlobe launched the first commercial satellite yesterday. Google, Microsoft, and several US government agencies are customers of DigitalGlobe. Such sharp images would be able to make out human faces, which, coupled with facial recognition software, could start to sound like a sci-fi dystopia. But it also would have interesting implications for both daily life and architecture, providing a startling degree of visibility across distances.

NASA's Terra Spacecraft Images Thailand Flooding (amazing image)

2012-03-27T13:03:00-04:00

In this image, vegetation is displayed in red, and flooded areas are black and dark blue. Brighter blue shows sediment-laden water, and gray areas are houses, buildings and roads. The image covers an area of 35.2 by 66.3 miles (56.7 by 106.9 kilometers) and is located at 14.5 degrees north latitude, 100.5 degrees east longitude.

Detached from the content depicted, the full-resolution image itself is too amazing to not circulate...