Plant Leaf Disease Detection and Identification Using Different Deep Learning Techniques
Date
2022
Student ID
Research Supervisor
Editor
Journal Title
Volume
Issue
Journal Title
Journal ISSN
Volume Title
Abstract
In this paper, we will be discussing disease detection using four types of Deep Learning approaches. Our research was based on 11 types of plant leaves and their diseases. The main motto of our project was to detect diseases of leaves. Here, we merged two datasets that include tea leaf illnesses and cherry, blueberry, maize, apple, peach, potato, tomato, soybean, strawberry, and raspberry leaves. As it is a supervised learning-based project, we had to build a model to which we introduced our dataset and successfully achieved the intended results. From four of the methods we learned in our research throughout the course period, we practically built a CNN (Convolutional Neural Network) based model. And then, we tried to train a few pre-trained models with our custom dataset. These models include ResNet-50, Inception V3, and ResNet-152V2. We achieved the best result with our CNN model, starting from 92% to a maximum of 96%. The lack of permission to access several revolutionary papers has made it harder for us to maximize our potential in collecting knowledge and schemes to achieve an impeccable overall outcome and declare our model to be accurate. Our dataset has not been used in any of the papers we have found. Therefore, a clear comparison between our created model and the paper’s models was quite difficult for us to present to our readers. A thorough reading of this paper will clarify our approaches and results.
Description
Keywords
Citation
Department Name
Electrical and Computer Engineering
Publisher
North South University