Identifying a machine learning approach to weed recognition in Australian grain production systems

The project will create an open source weed image library to make weed recognition data freely available to the weed control industry. This will provide a path to market for weed recognition technologies that can make use of the database. The project will determine best database structure for facilitating weed recognition algorithm development. A machine learning approach will be identified and implemented to demonstrate the utility of the database. Open access to the database will then allow developers to develop weed recognition software for growers.