Quality of Leaves Classification Using CNN

This study demonstrates the incredible benefits of information technology (IT) to agriculture by demonstrating how well Convolutional Neural Networks (CNNs) with dropout can be used to classify the quality of rubber tree seedlings through their leaves. The researcher has gone through every fundamental step such as pre-processing, structuring of the model, training and evaluation of the model, and producing classification report. Data augmentation has been applied to increase the number of data sample during the training of CNN and it gives 96% accurate result for new images. It was noticed that during the evaluation of the model it posted 98% accuracy result in classifying the data with the same procedure set during the training the model. However, the classification report of each data sample that is not included in the datasets of training and evaluation.