We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Boosting the Accuracy of Deep Learning Models for Detecting and Classifying Multiple Diseases in Plant Leaves using a Voting Classifier.
- Authors
Desai, Jesal; Ganatra, Amit
- Abstract
Indian economy largely depends upon agriculture. Emergence of Deep Learning has given solutions to many issues related to farming. One of the agricultural issues is plant disease detection and classification. Much research has been done and is still going on in this field as it directly relates to productivity of crops and plants. Loss in productivity of crops impacts farmer's income and thus country's economy. It has now become possible to avoid huge losses as many models have been implemented lately and various performance metrics have been used to evaluate them and compare their outcomes. In this study, we look into and analyze Deep Learning methods, that have been developed over the period of last few years. This article concentrates on a two-step approach. The first step is to find and classify nine types of tomato leaf diseases using six Deep Learning architectures: VGG16, VGG19, InceptionNet, ResNet50 V2, DenseNet 121 and MobileNet and analyze their performances based on precision, recall, F1- Score, testing and validation accuracy. The second phase is an ensemble technique known as Voting Classifier (VC), which trains an ensemble of many models and predicts an output (class) based on the highest probability of the chosen class being the outcome. Output of Deep Learning Models are compared with the proposed approach. The experimental analysis signifies that the proposed approach is effective and acheives better accur.
- Subjects
DEEP learning; AGRICULTURE; CROPS; FOLIAGE plants; VOTING
- Publication
Grenze International Journal of Engineering & Technology (GIJET), 2023, Vol 9, Issue 1, p90
- ISSN
2395-5287
- Publication type
Article