This notebook serves as a demonstration on the use of Machine Learning and Image Extraction algorithms to obtain useful information from an old map.
Python libraries used in this project:
The IPYNB file is better served using a running Jupyter Lab or Jupyter Notebook. Click here to run this notebook in a Google Colab session.
The "Jogjakarta en Omstreken" is a map from Colonial era of Indonesia in 1925. The map depict titular area of Yogyakarta, a center of cultural and historical importance in Indonesia. The map shows Town plan of Yogyakarta in Java, Indonesia, showing roads, boundaries, water features, vegetation, cultivation, swamps, buildings and built-up areasm and also relief which depicted by contours. The map is a collection of https://trove.nla.gov.au/m and could be accessed freely from here
- Display the old map and resulting features using Folium
- Divide the map into tiles, so that processing is faster. Probably using SuperMercado
- Testing different Machine Learning Algorithm, such as Random Forest Classifier
- Experiments with Cloud Optimised GeoTIFF (COG)
- Deep Learning using training Model Transfer
This works and its subsidiary are Licensed under GNU General Public License v3.0