NeuraMold: Harnessing Convolutional Neural Networks for Anthracnose Detection Precision

There is a widespread fungal disease known as anthracnose that affects a broad range of crops. This disease is one of the most severe threats to agriculture all over the globe. To accomplish the goals of this inquiry, this paper provides a novel approach to the detection of anthracnose that takes the use of convolutional neural networks (CNNs). Convolutional Neural Networks (CNNs) were trained & tested for finding anthracnose using a large dataset of 885 images. The diseases included are white spot, anthracnose, healthy, bird eye spot, brown blight, red leaf spot, gray light, and algal leaf. After completing extensive training and evaluation, this CNN model was able to obtain a validation accuracy of 84.2%. There was a low validation loss of 0.471, which suggested that the model could learn and generalize complex patterns related to anthracnose in plant photos. The model showed both of these capabilities. The validation loss was rather minor, which was a nice complement to this great performance.