AI Based Mango Leaf Disease Detection and Management System

Abstract

Leaf diseases pose a serious difficulty challenge in the management of mango growers and may cause significant yield loss if diagnosis is delayed. In this paper, we describe a novel deep learning-based approach for the classification of diseases in mango leaves using images. The method proposed takes advantage of a pretrained MobileNetV2 model to predict whether a leaf image is healthy or belongs to one of its diseased classes, including Anthracnose, Bacterial Canker, Powdery Mildew Cutting Weevil (Die Back), Gall Midge and Sooty Mould. In addition to detecting diseases, it provides treatment suggestions and general tips for leaf care, such as organic solutions. High-risk periods for disease spread are continuously analyzed, and alerts are generated based on prevailing weather conditions. System supports multiple languages (Marathi,English and Hindi) which makes the user interface Farmer Friendly; increase usability in different regions, this will contribute for early diagnosis improved crop management.

Introduction

This project presents an AI-powered Mango Leaf Disease Detection and Management System that helps farmers identify diseases early, receive treatment recommendations, and monitor disease risks using real-time weather data. Mango cultivation is a major agricultural activity in India, but diseases such as Anthracnose, Bacterial Canker, Powdery Mildew, Cutting Weevil, Die Back, Gall Midge, and Sooty Mould can significantly reduce crop quality and yield. Traditional disease diagnosis relies on expert visual inspection, which is often slow, costly, and inaccessible in rural areas.

The proposed system uses MobileNetV2, a lightweight deep learning model based on transfer learning, to classify mango leaf images as healthy or diseased by analyzing visual features such as color, texture, and patterns. Farmers can upload leaf images through a Streamlit-based user interface, where the system quickly identifies the disease and provides suitable treatment recommendations, including both organic and chemical control measures.

To enhance disease prevention, the system integrates real-time weather monitoring, analyzing temperature, humidity, and rainfall data to generate disease-risk alerts. Since environmental conditions strongly influence disease spread, these alerts help farmers take preventive actions before outbreaks occur. The application also includes multilingual support (Hindi, Marathi, and English), making it accessible to farmers from different regions.

The methodology involves collecting and preprocessing mango leaf images, training the MobileNetV2 model for disease classification, and combining the prediction results with weather-based risk analysis. The system is designed to be simple, efficient, and user-friendly while reducing dependence on agricultural experts.

The expected outcomes include accurate disease detection, timely treatment guidance, reduced crop losses, improved productivity, and support for sustainable farming practices. By integrating deep learning, climate analysis, treatment recommendations, and multilingual accessibility, the proposed system provides a practical and intelligent solution for mango disease management and smart agriculture.

Conclusion

The system delivers an intelligent method for the identification of the mango leaf diseases using a deep learning model. In this work we look at the identification of healthy and sick leaves via use of MobileNetV2.The system not only identifies the leaf diseases but it also gives treatment recommendations, organic solutions and general leaf care tips. The integration of real time weather analysis along with the multilingual support in languages such as Marathi, Hindi and English makes the user interface farmer- friendly.

References

[1] T. M. Saravanan, M. Jagadeesan, P. A. Selvaraj, M. Aravind, G. D. Raj,P. Lokesh, “ Prediction of diseases in mango leaves using aconvolutional neural network approach,” in Proc. 2023 Int. Conf. on Computer Communication and Informatics (ICCCI), Coimbatore, India, 2023. [2] A S. Kini, P. K. V. Reddy, and S. N. Pai, “A review of deep learning and image processing methods for plant leaf disease detection,” Int. J. Electr. Comput. Eng. (IJECE), vol. 13, no. 3, pp. 3029–3040, 2023. [3] T. R. Mahesh, V. Vinoth Kumar, R. Sivakami, I. Manimozhi, N.Krishnamoorthy, and B. Swapna, “Early-stage plant leaf disease detection using a MobileNetV2-based predictive model,” Int. J. Intell. Syst. Appl. Eng.(IJISAE), vol. 11, no. 2, pp. 46–54, 2023. [4] D. David, “A review of machine learning–based weather-driven prediction models for crop diseases and pests,” Asian J. Agric. Extens.,Econ. Sociol., vol. 41, no. 11, pp. 334–345, 2023. [5] B. U. Mahmud, A. Al Mamun, M. J. Hossen, G. Y. Hong, and B.Jahan,“A lightweight deep learning model for fast and efficient mango leaf disease classification,” Emerg. Sci. J., vol. 8, no. 1, pp. 28–42, Feb. 2024. [6] T.M. Saravanan, M.Jagadeesan, P. A. Selvaraj,M. Aravind, G. D.Raj, and P. Lokesh, “ Prediction of diseases in mango leaves using a convolutional neural network approach,” in Proc. 2023 Int. Conf. on Computer Communication and Informatics (ICCCI), Coimbatore, India, 2023. [7] A S. Kini, P. K. V. Reddy, and S. N. Pai, “A review of deep learning and image processing methods for plant leaf disease detection,” Int. J. Electr.Comput. Eng. (IJECE), vol. 13, no. 3, pp. 3029–3040, 2023. [8] T. R. Mahesh, V. Vinoth Kumar, R. Sivakami, I. Manimozhi, N.Krishnamoorthy, and B. Swapna, “Early-stage plant leaf disease detection using a MobileNetV2-based predictive model,” Int. J. Intell. Syst. Appl.Eng.(IJISAE), vol. 11, no. 2, pp. 46–54, 2023. [9] D. David, “A review of machine learning–based weather-driven prediction models for crop diseases and pests,” Asian J. Agric. Extens., Econ. Sociol.,vol. 41, no. 11, pp. 334–345, 2023. [10] V. Krishna Pratap and N. Suresh Kumar, “Deep learning approaches for classifying and evaluating mango leaf diseases,” ResearchGate Preprint,2024. [11] B. Uddin Mahmud, A. Rahman, and S. Islam, “A lightweight Dense Net-based deep learning model for mango leaf disease identification,” Emerg.Sci. J., vol. 8, no. 1, pp. 112–124, 2024. [12] P. Mohanty, D. Hughes, and M. Salathé, “Image-based plant disease identification using deep learning,” Front. Plant Sci., vol. 7, pp. 1–10, 2016.

Copyright

Copyright © 2026 Prerana Patil, Shravani Bhatkar, Mubashsheera Momin, Gauri Tanpure, Dr. Minakshee M. Patil. 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.