Brief Description:

Potato farming plays an important role in food and nutrition security. However, potato production in Kenya has not achieved its potential due to several production constraints with pests and diseases dominating and resulting in 80 to 100% losses in potato yield. Lack of disease free-seeds has led to the propagation of potatoes diseases such as bacterial wilt. Most potato farmers are small-scale holders in rural areas, therefore, low productivity results in low income which in turn increases poverty levels, poor lifestyle, food insecurity and nutrition associated problems. This study proposes an artificial intelligence (AI) based system for early disease detection in potatoes that can be utilized by both small-scale and large-scale farmers. For the farmers from the hardship areas, this study proposes a code-based model that will apply an Unstructured Supplementary Service Data (USSD) framework to access a trained classifier model that will predict a potato disease based on an inputted USSD code text (signs and symptoms). While farmers in urban setup, the study proposes a CV-based model that will predict a potato disease based on a captured potato image. AI-based monitoring solutions are characterized as fast, accurate, efficient, robust, effective, and non-subjective.

Amount: Ksh: 500,000

Funding Organization: MMUST

Grant Number: FY22/23-URF-004

Dr. Cedric Okinda: - PI

Mr. Hesbon Amwayi:  Co- PI

Dr. Jairus Odawa:  Investigator

Dr. Evelyne Mmbone:  Investigator

Project Activities and Events