Pendugaan tingkat keamanan jagung dengan menggunakan pengolahan citra digital dan jaringan syaraf tiruan

Abstract

Food security is the main problem in food production in Indonesia. One of the reason that the food from corn is not safe to be consumed caused by contaminated by aflatoxins which produced by fungi. Determination of food contaminants is usually carried out by laboratory analysis. This method is very expensive and need a lot of time. The objective of the research was to predict the security of corn due to aflatoxin contaminant by using image processing and artificial neural network (ANN). The image of seed corn was taken by using digital camera and processed by image processing program. ANN model was developed with 10 input parameters, 20 hidden layers and 4 targets. The targets of safety levels was four such as very safe, safe, less safe and not safe. The accuracy of the model was 74%.