Enhancing Tea Leaf Disease Classification: Leveraging Data Augmentation, Diverse Feature Extraction Techniques, and Ensemble Stacking with Machine Learning Models

DOI

10.2991/978-94-6463-740-3_14How to use a DOI?

Keywords

Bag of Visual Words; Color Histogram; Decision tree; Ensemble Stacking; F1 score; Pixel feature extraction; Support Vector Machine; Tea leaf disease; XGB classifier

Abstract

In recent years, the tea industry has gained newfound importance, driven by advancements in technological innovation and automation. These developments have played a pivotal role in boosting productivity and ensuring the quality of tea production. Within the field of tea agriculture, a crucial challenge lies in the accurate detection and classification of diseases that can afflict tea leaves. In this research, a comprehensive methodology is presented for addressing this challenge. Leveraging the Tea Plant Disease Dataset, a widely accessible and reputable resource containing diverse tea leaf disease images, this approach unfolds in five key stages: (1) Data Augmentation, (2) Feature Extraction Methods, (3) Machine Learning, (4) Ensemble Stacking and (5) Evaluation of F1 score.

In this research, it is observed that while using the data without data augmentation, XGBoost (XGB) classifier demonstrated superior performance for Color Histogram feature extraction, while for Pixel Feature extraction, Random Forest excelled. For Bag of Visual Words (BoVW) feature extraction, Support Vector Machine (SVM) emerged as the top-performing model. After data augmentation, Random Forest demonstrated superior performance for Color Histogram feature extraction, while for Pixel Feature extraction, XGB Classifier excelled. For Bag of Visual Words (BoVW) feature extraction, Support Vector Machine (SVM) emerged as the top-performing model. These insights provide valuable guidance for optimizing the F1 score in tea leaf disease classification. Then ensemble stacking was used for all the machine learning models to identify the model with the best accuracy while using ensemble techniques. The ultimate aim for this research is to make these trained models accessible to farmers for early disease detection and classification, acting as a preventive measure to safeguard tea leaves. This research marks a significant stride towards sustainable agriculture, empowering farmers with technology and machine learning to protect yields and ensure food security.

Copyright

© 2025 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  - K. Somesh
AU  - C. Shanmukh Srinivas Sai
AU  - Nithin Mude
AU  - B. Surendiran
AU  - J. Dhakshayani
PY  - 2025
DA  - 2025/06/25
TI  - Enhancing Tea Leaf Disease Classification: Leveraging Data Augmentation, Diverse Feature Extraction Techniques, and Ensemble Stacking with Machine Learning Models
BT  - Proceedings of the 6th International Conference on Deep Learning, Artificial Intelligence and Robotics (ICDLAIR 2024)
PB  - Atlantis Press
SP  - 149
EP  - 168
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-740-3_14
DO  - 10.2991/978-94-6463-740-3_14
ID  - Somesh2025
ER  -