This is a repository to track development of a new model for DeepForest, a Python package for predicting tree crowns. Please see the wiki to get a better understanding how we implemeneted Boston Specific DeepForest model by training it over Local data to get better predictions.
The ideal outcome of this project is:
-
The creation of a new, accurate machine learning model focused on urban tree canopies
-
The creation of a simple web interface for the Parks Department to upload new aerial imagery for analysis
-
The ability for the Parks Department of Boston to generate a list of statistics on tree counts to ensure that it continues to plants trees in an equitable manner across the city.