Research Project 2025

AI-Powered Bean Disease Detection and Recommendation System

Supporting Tanzanian farmers through machine learning and intelligent disease diagnosis.

This project uses Convolutional Neural Networks (CNN) and artificial intelligence to help farmers identify bean crop diseases early and receive timely treatment recommendations, reducing crop losses and improving productivity.

Contact Us

99%+

Accuracy

6

Disease Categories

12K+

Training Images

Overview

Project Overview

Detailed information about the research behind this system

Project Title

Recommender System for Bean Crop Diseases Detector Using Machine Learning

Researcher

Makombe Kalebu Oscar

Institution

Sokoine University of Agriculture (SUA)

Academic Program

Bachelor of Information Technology

Supervisor

Dr. Michael P. J. Mahenge

Year

2025

The Challenge

Problem Statement

Many bean farmers in Tanzania struggle to accurately identify crop diseases due to limited access to agricultural experts. Incorrect diagnosis often leads to poor treatment decisions, reduced yields, and economic losses.

Limited Expert Access

Rural farmers lack access to trained agricultural extension officers

Misdiagnosis

Incorrect disease identification leads to wrong treatments

Economic Losses

Crop failures cause significant financial impact on farming communities

Why This Matters

Beans are a staple food crop in Tanzania, providing essential nutrition and income for millions of smallholder farmers. Protecting these crops is crucial for food security.

Goals

Project Objectives

Main Objective

Develop a machine-learning-based application that helps farmers identify diseases affecting bean crops.

Objective 1

Build a comprehensive bean disease image dataset.

Objective 2

Train an AI model capable of accurate disease classification.

Objective 3

Develop a user-friendly web system for disease diagnosis and recommendations.

Objective 4

Improve farmers' decision-making through timely information.

Capabilities

Key Features

Powerful tools designed to support farmers and agricultural officers

AI Disease Detection

CNN-powered image analysis for accurate disease identification

Treatment Recommendations

Actionable treatment advice based on detected diseases

Mobile-Friendly

Responsive interface accessible on any device

Officer Support

Dedicated tools for agricultural extension officers

Prediction History

Track and review past disease detections over time

Analytics & Reports

Comprehensive reporting and data analytics dashboard

Bilingual Support

Full Swahili and English language support

Early Identification

Detect diseases early before they cause major damage

Results

Research Impact

0
Images Used for Training
0
Disease Categories
99-100%
Classification Accuracy
Designed for Tanzanian Farmers

This project contributes to sustainable agriculture and food security in Tanzania by enabling early, accurate disease detection. By empowering farmers with AI-driven insights, we aim to reduce post-harvest losses, improve crop management practices, and support the livelihoods of smallholder farming communities.

Get in Touch

Contact Us

Have questions about the research? Want to collaborate? Reach out!

Researcher Information

Name

Makombe Kalebu Oscar

WhatsApp

+255 615 090 662

Institution

Sokoine University of Agriculture

Send Email Chat on WhatsApp