Fight hunger through machine learning-based crop classification in Uganda.
Agriculture is a core sector of Uganda’s economy and the largest employer. According to the Uganda National Household Survey (UNHS) 2016/17, the bigger proportion of the working population is engaged in agriculture, forestry, and fishing (65%). Among the females in the working population, 70% are engaged in agriculture compared to 58% of the males. Plantains, cassava, sweet potato, and maize are major subsistence crops. The major export crop is coffee, but tea, tobacco, and cotton are also important. Although many farmers sold food crops to meet short-term expenses, the government attempt to encourage diversification in commercial agriculture that would lead to a variety of nontraditional exports. The agriculture sector had a total contribution to GDP at current prices of 24.9 percent in FY 2016/17 compared to 23.7 percent in FY 2015/16. The food crop subsector registered the highest contribution within the agricultural sector of 13.6 percent in FY 2016/17, an increase from 12.1 percent in FY 2015/16. The government has therefore concluded that investing in agriculture to achieve higher growth rates is the most effective way of reducing poverty.
Uganda’s agricultural sector presents multiple highly profitable investment opportunities both for profit-oriented investments and partnerships. While some steps are being taken to provide insurance against crop failures, access to finance for small-scale farmers is limited. The high cost and limited availability of improved farm inputs, including hybrid seeds and post-harvest technology, over-stretched extension services, poor transport networks, a lack of market information, inadequate production and post-harvest facilities, and weak value chain linkages all hinder and frustrate subsistence farmers. Despite the enormous progress in poverty reduction, about 40 percent of all rural people still live below the poverty line; the poorest regions being in the north and north-east, where civil conflict has severely disrupted the lives and agricultural production of small farmers.
The goal is to build a machine learning model to fight hunger in Uganda. We hope to provide the solution for crop production, increasing degradation of resources, climate change, etc.
The project results will be made open source. The machine learning model-based crop classification will help the government and private institutions to make data-driven decisions on where to allocate resources for crop growth most effectively.
1. To develop a machine learning-based crop classification in Uganda based on the same models built for Nepal.
2. To establish a dataset for the crop.
3. To do image preprocessing, training, and testing the classifier.
1. Data collection, preprocessing, and dataset building.
2. Develop, train and test deep learning models for image classification.
Link to Original Project: https://omdena.com/projects/cropclassification/
We will be running an AI project soon…. Stay Tuned!
Uganda Chapter Leads
Experienced Researcher with a Bachelor of Science in Computer Engineering and demonstrated history of working in the Data Science and Machine Learning industries. Gets happiness in finding solutions to my community problems using AI technologies. Recently Worked on some of the big issues of my community like Smart Traffic System and Crop Disease Detection. Working as Machine Learning Engineer for Kin-Keepers and Omdena.
Msc Data Science and Analytics [UK] candidate and Machine Learning Engineer at Omdena Uganda. He is passionate about building and evangelising to enthusiasts about creating cutting edge solutions in the space ML & AI Applications that will revolutionise life as we know it.