Detection of optimum maturity of maize using image processing and artificial neural networks

Abstract

Maize is one of the most widely cultivated crops in Nigeria having good contents of protein, carbohydrate and fat. The leaves of maize are also very good source of food for grazing livestock like cows, goats, sheep, etc. However, in Nigeria the leaves are left in farm lands to dry up completely (brownish) or even rot with little or no nutritional value to livestock. At the maturity of maize the leaves are still maintaining green coloration and the stalk are still fresh. Changes in the maize leaf coloration during maturity were studied. A CCD camera for image acquisition of the different green colorations of the maize leaves at maturity was used. Different color features were extracted from the image processing system (MATLAB) and used as inputs to the artificial neural network that classify different levels of maturity.

Keywords: Maize, Maturity, CCD Camera, Image Processing, Artificial Neural Network