EUVIMEDEUVIMED entdecken
Crossref

Enhance KNN Model Performance using Stepwise Regression Feature Selection Method for Prediction of Tomato Leaf Diseases

Kumar Sanjeev, Suneeta Paswan, Pramod Prabhakar, Deban Kumar Choudhary, Mrinalini Kumari

INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES · 2026-06 · 22 (01) · 153

DOI: 10.59467/ijass.2026.22.153 ↗

Kurz gefasst

Fachartikel in der Zeitschrift INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES, 2026-06. Verfasst von Kumar Sanjeev und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

It is estimated that a significant loss in tomato yield is caused by various diseases. In this study, we utilized tomato leaf images collected from the Plant Village dataset, which contains 4,898 images categorized into three disease classes: Bacterial Leaf Spot, Early Blight, and Septoria Leaf Spot. The data analysis and processing were conducted using MATLAB 2018b. A total of 76 features related to color, texture, and shape were extracted from the images with help of image processing technique. Feature selection plays a crucial role in pattern classification, as it significantly affects both accuracy and processing time-especially when the initial feature set is large. The developed model employs the K-Nearest Neighbors (KNN) algorithm for predicting and classifying the tomato leaf diseases. Stepwise regression was used as the feature selection technique. The model achieved an accuracy of 85.58% without any feature selection. However, by selecting 45 out of the 76 features using stepwise regression, the accuracy improved to 88.15%. This enhancement demonstrates that effective feature selection can significantly improve classifier performance, leading to more accurate disease prediction.. KEYWORDS :Tomato, Plant disease, Feature extraction, Feature selection, Classification.

Zitieren

APA

Sanjeev, K., Paswan, S., Prabhakar, P., Choudhary, D. K., & Kumari, M. (2026). Enhance KNN Model Performance using Stepwise Regression Feature Selection Method for Prediction of Tomato Leaf Diseases. INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES, 22(01), 153. https://doi.org/10.59467/ijass.2026.22.153

Vancouver

Sanjeev K, Paswan S, Prabhakar P, Choudhary DK, Kumari M. Enhance KNN Model Performance using Stepwise Regression Feature Selection Method for Prediction of Tomato Leaf Diseases. INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES. 2026;22(01):153. doi:10.59467/ijass.2026.22.153

BibTeX
@article{Sanjeev2026,
  author = {Sanjeev, Kumar and Paswan, Suneeta and Prabhakar, Pramod and Choudhary, Deban Kumar and Kumari, Mrinalini},
  title = {Enhance KNN Model Performance using Stepwise Regression Feature Selection Method for Prediction of Tomato Leaf Diseases},
  journal = {INTERNATIONAL JOURNAL OF AGRICULTURAL AND STATISTICAL SCIENCES},
  year = {2026},
  volume = {22},
  number = {01},
  pages = {153},
  doi = {10.59467/ijass.2026.22.153}
}
Kostenlos starten →