| dc.description.abstract | Traditional disease identification approaches either rely on the farmers’ knowledge base on the existing diseases or rely on the support of agricultural extension organizations, but these organizations are limited in the country with low logistical and human infrastructure capacity, and are expensive to scale up.
With the basis of the research being on the interest of the bananas; these are an important starchy food and cash crop in Uganda and the Great Lakes region of East Africa. Widespread reports of declining yields in Uganda since the 1930s and the low yields today do raise serious sustainability and food security concerns, especially as food demand increases with a population growth rate of 3.2% per annum. Farmers cite soil fertility decline, pests (banana weevils and nematodes) and moisture stress as the major factors responsible for yield decline. In response, several organic and mineral fertilization experiments have been carried out at research stations and in farmers’ fields in Uganda since the 1950s. Although pests are controlled in some trials, researchers have often failed to embrace a systems approach, quite often leaving out factors, such as moisture stress and soil physical conditions that affect the responses to fertilization. In order to set proper banana research priorities to benefit farmers in Uganda, the objective of this study was to develop a web-based Application system with integrated AI for banana disease detection.
The mathematical approach to find the relationship between two or more variables is known as Regression in AI and is the method we employed in development of the banana diseases detection system. Regression is widely used in Machine Learning to predict the behaviour of one variable depending upon the value of another variable. Unlike the classification models, the regression models output numeric values. It also has continuous values for both dependent and independent variables, and for the most part, Regression is classified as supervised learning. There is potential to obtain real time help at the comfort of one’s plantation or farm land and also providing collective approach to dealing with the never-ending threat to Banana crops which are one of the most consumed food crops in our motherland. Suggestions among farmers’ experiences involved in dealing with specific disease infections are also realized. The web-based application system with integrated AI for banana disease detection is a potential remedy to some of the challenges faced by farmers growing bananas. | en_US |
