The application, when completed would detect diseases of four crops – maize, cassava, cashew and tomatoes, Dr Patrick Kwabena Mensah, a Senior Lecturer at the Department and the Principal Investigator of the AI for Agriculture and Food Systems (AI4AFS) Project at UENR said in an interview with the Ghana News Agency in Sunyani.

Dr. Mensah said the project is under the funding of the Kenya-based International Development Research Centre (IDRC) which is also sourcing funds from Canada and is expected to be completed in 18 months, with six months already done.

He said currently the application was at the development phase, saying by July this year the team would have been done with the development and then engage the farmers around September to train them about how the application works.

Touching on the modalities of the application, Dr Mensah explained the images of the four selected crops would be taken, where diseased and healthy parts of the plant would be taken and then trained as an AI model, which would then be developed into a mobile application and then installed on an android phone and iPhone Operating System (IOS).

He further said a web application would be developed and a farmer with access to a computer must pluck a leaf of a diseased crop, so that with the help of the computer webcam, the AI application could detect the type of diseases on the computer.

Dr Mensah said with the mobile phone, a farmer on the farm realising a crop either had been infected or had an unusual appearance “have to take a picture of that part and the application will identify the diseases affecting the crop and further prescribe environmentally friendly recommendations about how to control the specific disease.”

He said the team realised most farmers were either not having phones and computers or their phones were not Android/IOS inclined, hence as part of the project an ‘E-kiosk’ would be set up in five communities in the pilot phase.

The ‘E-kiosk’, he said, would comprise either a computer, mobile phone and an attendant to assist farmers without these gadgets to detect diseases from their affected crops.

Dr Mensah said the application when operational would help boost crop production since it would ensure early detection of pests and diseases in the farms and give prevention controls to mitigate crop losses.