Abstract:
Enset is a monocarpic perennial crop that belongs to the Schistaminae order
and the Musaceae family. Enset is a significant food security crop in
Southern Ethiopia, where almost 20 million people depend on it for
survival. Plant leaf diseases and damaging pests are foremost challenges in
Enset production. This study looks into the use of deep learning to detect
bacterial wilt disease and Enset mealy bug pest, where data is obtained in
small amounts and collected under minimally controlled conditions. We
employed data augmentation to get over the limits of the dataset size. The
proposed approach is divided into four stages. The initial part entails
gathering healthy and diseased Enset images with the support of agricultural
experts, from various farms and research institutes. Then image processing
tasks, resizing and segmentation are applied on the collected dataset in
order to get an enhanced (simpler) image and to extract region of interest
from the dataset images. Finally, using the collected dataset, the created
model is trained and evaluated, and it is compared to the state of the art
pre-trained convolutional neural network models: AlexNet,
ResNet-50, Inception v3, DenseNet-201, VGG16 and EfficientNetB3.
The proposed approach is implemented using Google collaboratory or
"colab" for short. To detect and classify Enset diseases, the model has an
accuracy of 99.68% for training and 98.87% for testing
