Boundary: We filtered the original building boundary shp file by only selecting the common type and calculating the centroid of each object. For all other codes that used the shp file, we will use the updated one.
Image info: We run a loop over the folder of a whole year's image, track down the information of each image and the houses we are going to crop out in the image, and then store all the information into one Excel file.
Houses Crop: We crop the houses within each image into a new folder
Completeness Check: We check if the images fully cover the whole area of toronto
Image resize: We resize the image to 640*640 and generate empty file for labeling of images without objects.
RGB_Adapter: Convert pictures with four color channels into RGB bands.
House_Image_Classifier: Use a GUI to display images from previous years and move them to a specified folder. (old)
House Image Classifier GUI: A Python GUI application for classifying house images.
Roof Change GUI: A Python GUI for identifying house roof change.
Year 2011-2018: EPSG:2019
Year 2019: EPSG:2958 (File "TorontoBoundarywgs84" passed the completeness check)
Year 2020-2022: EPSG:2952