Tomato Leaf Disease Detection Using Deep Learning

Abstract

Tomato leaf diseases cause major losses in crop production and seriously affect agriculture. Identifying these diseases at an early stage is very important for proper treatment and prevention. In this work, we present a new method for tomato leaf disease detection using deep learning with a Counting Sort–based feature extraction technique. Traditional image-based disease detection methods often have problems such as unnecessary features and high computational cost. To solve this, we use Counting Sort to efficiently analyze the distribution of pixel intensity values in different color channels of tomato leaf images. This helps in extracting important color and texture features related to different diseases. These extracted features are then given to a Convolutional Neural Network (CNN) to classify tomato leaves into early blight, late blight, or healthy categories. The model is tested on a publicly available tomato leaf image dataset. Experimental results show that the proposed method achieves better accuracy and faster processing compared to traditional deep learning methods that use raw images. Overall, this approach provides an effective and automatic way to detect tomato leaf diseases, which can help farmers improve crop management and increase food production.

Introduction

Tomato crops, vital to global agriculture, are highly susceptible to leaf diseases like early and late blight, which reduce yield and quality. Traditional disease detection through manual inspection is slow, labor-intensive, and error-prone. To address this, the proposed system integrates AI, deep learning, and IoT technologies for automated disease detection and crop monitoring. Leaf images are processed using Counting Sort–based feature extraction and classified with a Convolutional Neural Network (CNN) into Healthy, Early Blight, or Late Blight categories. Environmental factors such as temperature, humidity, and soil moisture are monitored with DHT11 and soil moisture sensors, while a camera sensor captures leaf images. The NodeMCU ESP8266 transmits data to the cloud, enabling real-time monitoring, alert generation, and automatic irrigation control.

This system allows early disease detection, reduces pesticide misuse, improves crop management, and supports precision agriculture by combining intelligent image analysis, environmental monitoring, and IoT-enabled automation.

References

[1] Saravanan, M. S., & Badabagni, L. D. (2024). Tomato plant disease prediction system with a new framework SSMAN using advanced deep learning techniques. International Journal of Power Electronics and 15(1), Drive Systems, 940. https://doi.org/10.11591/ijece.v15il.pp940-948. [2] Mewada, H., Sundar, L. S., Desai, M., & Mohammed, N. (2024). Harnessing Transdisciplinary Knowledge: Integrated Deep Learning Techniques for Accurate Tomato Leaf Disease Classification. https://doi.org/10.22545/2024/00264. [3] Subramanya, S. K., & B, N. (2024). Enhanced detection of tomato leaf diseases using ensemble deep learning: INCVX-NET model. IAES International Journal of Artificial Intelligence, 13(4), 4757.https://doi.org/10.11591/ijai.v13.i4.pp4757-4765. [4] Patwal, P., Chauhan, R., Bhatt, C., & Devliyal, S. (2024). Automated tomato disease detection and classification using image processing and machine learning for precision agriculture. 481-486. https://doi.org/10.1201/9781003559085-84. [5] Yang, J. (2024). Deep Learning for Tomato Leaf Disease Recognition Environment. in Balanced Data https://doi.org/10.1109/ccsb63463.2024.10735683. 74-78. [6] Vinutha, H. R., G.K., P., & Karegowda, A. (2024). An Effective Analysis of Tomato Plant Leaf Disease Identification Using Deep Learning. 1-4. https://doi.org/10.1109/icdsns62112.2024.10691220

Copyright

Copyright © 2026 M. Hema, P. Jayasree, R. Jogendra Reddy , B. Yaswanth , K. Pavan Kumar. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.