One of the nearly 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 dissimilar urban environments. The projection aims to quantitatively recognize which areas of a metropolis are perceived every bit wealthy, modern, safe, lively, active, unique, central, adaptable or household unit of measurement friendly based on how people answer to Street View images of the city.
MIT has straightaway 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 acquaint inwards the images, associated alongside the image's textures, colors as well as shapes. They so used machine learning to associate icon features alongside scores of perceived safety. MIT tin so purpose the resulting algorithm to predict the perceived security of a novel image. They tin hence give whatever Street View icon a 'StreetScore' based on the results of the Place Pulse survey.
StreetScore has straightaway released a disclose 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 correspond the areas which StreetScore has assigned every bit having a high perceived security rating as well as the cherry-red 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 get got 3 fields: latitude, longitude as well as q-score.
There is a lot to a greater extent than you lot could practice alongside this data. I'd dear to run across a StreetScore map alongside a criminal offense heat-map layer. This would render a prissy agency to run across how closely the perceptions of street security inwards the MIT analysis matches upward to crimes on the ground.
One of the things missing from the Streetscore map is the might to persuasion the Street View images of the locations. Therefore simply about other thought would endure to practice a Google Map of the information as well as include the selection to click on the map as well as persuasion the actual Street View.
One of the things which intrigued me when browsing the map was what did the Street Views genuinely hold back similar for the locations alongside high as well as depression streetscores. To acquire simply about thought I downloaded the Boston StreetScore file as well as rapidly threw together this picayune site showing the summit ix perceived safest locations as well as the summit 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 ninety score persuasion of each location. The q-scores for the Boston as well as New York information doesn't include the 'yaw' information to betoken which business office of the 360 score panorama was genuinely used inwards the survey. I hence simply used a random administration inwards the Street Views I've shown.
One affair that rapidly becomes apparent (see the screenshot above) is that the presence of trees seems to endure a major element inwards the MIT algorithm for assessing the safest perceived streets.
