Farmer Bienvenue Nzonje Ngoul has been growing tomatoes in southwestern Cameroon for four years. Unfortunately, his yields have dwindled recently due to unknown diseases. "My crops are regularly attacked by illnesses and these can cripple you in one day," Ngoul said. To avoid a repeat of this, the farmer set out to look for a solution.
That is when he met 27-year-old Landry Doko who co-founded Agrix Tech, an artificial intelligence (AI) based mobile application that detects plant diseases at primary stage by analyzing photos of the sick crops, and offers both chemical and physical treatment as well as preventive measures.
Last December, Doko went to Ngoul's farm to test his work. The farm is in a remote area in Dibombari where the network signal is weak like in most of Africa's rural zones, but Doko's app has everything built inside and can be used without internet.
It turned out that Ngoul's tomatoes suffered from early blight, a common disease that can occur nearly every season. The diagnosis is delivered through text and voice for the easy understanding of farmers who could be less literate. Users can choose standard or pidgin English and French, as well as African languages like Hausa.
With backgrounds in machine learning, software development, and crop pathology, Doko's team has now made available a working prototype covering tomato, grape, potato, and corns. According to the Food and Agriculture Organization (FAO), there are around 33 million small farms of less than two hectares in sub-Saharan Africa, representing about 80 percent of all farms in the region. The ambition of Doko's team is to help these smallholder farmers "sufficiently live from the fruits of their production and at a lower cost."