This step obtains street view images for a provided *.csv file of buildings addresses such as the one from the San Francisco soft story property list. Note that the *.csv file must contain a PROPERTY and STATUS column. Entries in the PROPERTY column must be of the form house_number␣street_name.
To run the script you first need to obtain a Google Street View Static API key on the gcp website. To start downloading street view images run:
python Download_Images.py --G_API_key <your Google API key>
Depending on your connection, this will take between two and five hours to complete. To get started on a small test set of 62 buildings run:
python Download_Images.py --G_API_key <your Google API key> --test
By default, addresses in EQanalytics/1_Pre_Processing/Locations/SF.csv are used. Downloaded images will be saved in EQanalytics/Data/Street_View_Images and the index file under EQanalytics/Data/Street_View_Images/results.csv. Use the flags --address_file, --result_folder and --result_file to change this behavior. In all cases you must provide an absolute path.
Other possible flags are --city, which is set to SAN FRANCISCO by default, and --pitch, which determines the angle of the street view camera. The default pitch value is set to 10 and 0 would mean a completely horizontal picture.