Application of Image Processing and Artificial Neural Networks to Identify Ripeness and Maturity of the Lime (citrus medica)

The object of this study is to identify maturity and ripeness of lime (citrus medica) using Image Processing and Artificial Neural Network. Image processing method was developed and applied to samples of lime from three levels of maturity and ripeness based on their harvest time. The area, shape factor, energy and color indexes were extracted from image processing development system. The feature extracted from image processing were used as artificial neural network inputs and trained by using back propagation methods with 4 model neural network and used momentum constant’s value 0.8, learning rate constant value 0.8 with 1 until 2000 iteration. The result showed that the use of 4 back propagations neural network models with 3, 6, 9 and 12 hidden layers provided the 100% accurateness in classifying the lime based on their maturity and ripeness.