Tomato Leaf Disease Classification with Optimized Hyperparameter: A DenseNet-PSO Approach

DOI

10.2991/978-94-6463-413-6_23How to use a DOI?

Keywords

Tomato leaf disease; Hyperparameter; CNN; DenseNet; PSO

Abstract

Tomato stands out as one of the foremost horticultural crops worldwide, including in Indonesia that profoundly influences the agricultural economy. Tomato leaf disease poses a critical and direct threat to quality, yield, and overall production of tomato crops, demanding swift attention and action. Manual detection of tomato leaf diseases not only consumes significant time, effort, and financial resources, but also demands specialized expertise and skills, hampering the urgency and effectiveness of disease control measures. Utilizing CNN, a robust AI approach known for its high precision in image classification, serves as a valuable tool for the automated identification of diseases in tomato leaves in large datasets. DenseNet is a type of CNN architecture that establishes connections between each layer and all subsequent layers, ensuring comprehensive connectivity throughout the network. This approach effectively minimizes the distance of gradient flow during backpropagation, addressing and alleviating issues related to vanishing gradients. Notably, hyperparameter optimization shows significant potential for improving the performance of CNN. In this work, we propose an innovative algorithm driven by Particle Swarm Optimization (PSO) to swiftly and efficiently converge on optimal hyperparameter configurations, including number neuron in fully connected layer, dropout rate, learning rate, activation function and optimizer for DenseNet architecture. Through this optimization, we aim to boost the classification accuracy of the DenseNet architecture in distinguishing among nine distinct types of tomato leaf diseases. The proposed DenseNet-PSO achieves up to 7.67% improvement in accuracy.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Cynthia Ayu Dwi Lestari
AU  - Syaiful Anam
AU  - Umu Sa’adah
PY  - 2024
DA  - 2024/05/13
TI  - Tomato Leaf Disease Classification with Optimized Hyperparameter: A DenseNet-PSO Approach
BT  - Proceedings of the First International Conference on Applied Mathematics, Statistics, and Computing (ICAMSAC 2023)
PB  - Atlantis Press
SP  - 228
EP  - 239
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-413-6_23
DO  - 10.2991/978-94-6463-413-6_23
ID  - Lestari2024
ER  -