How Rubber Is This Street?

One of the around interesting uses of Google Maps Street View imagery has been yesteryear MIT. MIT's Place Pulse projection is a crowd-sourced experiment examining people's perceptions of unlike urban environments. The projection aims to quantitatively recognize which areas of a metropolis are perceived equally wealthy, modern, safe, lively, active, unique, central, adaptable or identify unit of measurement friendly based on how people answer to Street View images of the city.

MIT has at 1 time taken the the security rankings for 3,000 street images from New York as well as Boston as well as created an algorithm to automatically practice a perceived security rating for Street View images. Using the Place Pulse scores MIT assigned attributes to features introduce inwards the images, associated alongside the image's textures, colors as well as shapes. They as well as then used machine learning to associate ikon features alongside scores of perceived safety. MIT tin as well as then occupation the resulting algorithm to predict the perceived security of a novel image. They tin thus give whatever Street View ikon a 'StreetScore' based on the results of the Place Pulse survey.


StreetScore has at 1 time released a publish of maps showing areas of perceived security inwards New York, Boston, Chicago as well as Detroit. Using Street View images of the metropolis StreetScore assesses the perceived security of locations throughout the city. Green dots on the map stand upwardly for the areas which StreetScore has assigned equally having a high perceived security rating as well as the cerise dots are the locations alongside a depression perceived rating score.

The information of the StreetScores for Boston as well as New York is available for download. It is provided inwards 2 comma separated CSV files. The CSV files direct keep iii fields: latitude, longitude as well as q-score.

There is a lot to a greater extent than yous could practice alongside this data. I'd beloved to run into a StreetScore map alongside a criminal offense heat-map layer. This would supply a squeamish means to run into how closely the perceptions of street security inwards the MIT analysis matches upwardly to crimes on the ground.

One of the things missing from the Streetscore map is the mightiness to stance the Street View images of the locations. Therefore unopen to other thought would endure to practice a Google Map of the information as well as include the alternative to click on the map as well as stance the actual Street View.


One of the things which intrigued me when browsing the map was what did the Street Views genuinely await similar for the locations alongside high as well as depression streetscores. To acquire unopen to thought I downloaded the Boston StreetScore file as well as chop-chop threw together this petty site showing the elevation ix perceived safest locations as well as the elevation ix perceived to the lowest degree security locations inwards Boston. The site uses the Google Street View Image API to exhibit yet images of the locations.

In the master copy survey MIT alone showed a xc bird stance of each location. The q-scores for the Boston as well as New York information doesn't include the 'yaw' information to signal which move of the 360 bird panorama was genuinely used inwards the survey. I thus only used a random direction inwards the Street Views I've shown.

One affair that chop-chop becomes apparent (see the screenshot above) is that the presence of trees seems to endure a major component inwards the MIT algorithm for assessing the safest perceived streets.