| dc.description.abstract | Corn is one of the staple food ingredients in Indonesia. The price of corn commodities has been increasing over the years due to high demand. In addition to being used as food, corn is also used as biofuel and feedstock in various countries. Along with the increasing demand for corn commodities, it has been found that corn harvests often fail due to pests or diseases. One way to identify diseases in corn plants is through changes that occur in the corn leaves. The objects used in this study are leaf diseases in corn plants that have similarities in their types, including leaf blight, leaf spot, leaf streak, and leaf rust diseases. The total number of datasets to be trained is 1000, with 600 data for training, 200 data for validation, and 200 data for testing. The training data processing is performed with 64 epochs and a batch size of 32. The training process then generates a model to be used in the testing process. Additionally, there is a validation dataset comprising 20% of the total research data, amounting to 200 data points. The combination of Convolutional Neural Network and K-Means Clustering with the pre-trained Sequential model is able to detect corn leaf diseases with an accuracy of 92%. | en_US |
