Shazam for plants: new software identifies species from single leaf photo

RELATED
Landscape Design - Sustainability Awards finalists 2015
Biophilic design: it’s intuitively obvious but we need to document it

A computer science researcher at Edith Cowan University (ECU) has developed a unique software that can automatically identify plants from a single digital photo of the leaf.

With more than 250,000 flowering plants on Earth, the only way to identify any plant species is to refer the sample to a botanist. However, the project by ECU School of Computer and Security Science PhD candidate Oluleye Hezekiah Babatunde has led to the development of a computer program that uses high-tech image recognition software to identify the species of a plant based on a digital photo of a single leaf.

Similar to the popular music identification app Shazam, the program uses a combination of highly complex mathematical algorithms and artificial intelligence to identify leaves based on their colour, shape and texture. According to test results, the software has shown 93 per cent accuracy in plant identification.

The software is expected to have extensive applications in the agricultural sector. The processes used by the program to distinguish between different species could be used to monitor damage to crops by pests, weather and herbicide. Babatunde added that there’s potential for farmers to use this tool as an app to get real time updates on their crops with a simple smartphone image.

Babatunde is now seeking partnerships to take the project out of the lab and into the field.