<![CDATA[Potatoes are an important horticultural commodity that is susceptible to various diseases such as black scurf, blackleg, common scab, dry rot, and pink rot. Early disease detection is crucial for reducing losses and increasing productivity. This study aims to develop an image-based potato disease classification model using the Convolutional Neural Network method. The dataset consists of 1,760 images divided into eight classes, with a data split of 70% training data, 20% validation data, and 10% test data. The data processing involves preprocessing and augmentation to improve the model’s generalisation. The convolutional neural network architecture used consists of four convolutional layers combined with batch normalisation, max pooling, dropout, and fully connected layers. The model was trained using the Adam optimiser with a learning rate of 0.0003 and a batch size of 32. Testing results show that the model achieved an accuracy of 86%, with an average precision of 0.87, a recall of 0.86, and an F1-score of 0.86. The evaluation results indicate that the model is capable of recognising most classes well, although errors still occur in those with high visual similarity. This study demonstrates that the CNN method is effective for classifying potato diseases based on digital images and has the potential for further development on larger datasets. ]]>
Copyrights © 2026
