Home / Current Issue / Paper 1715542
An AI-Powered Mobile Application for Crop Disease Detection Using MobileNetV2 and Flutter
Subject area: Science,Engineering and Technology · Area of research: Computer Vision and Deep Learning
DOI: https://doi.org/10.64388/IREV9I9-1715542
Abstract
Agriculture forms the basis of the Indian economy. However, crop disease has been a major problem in reducing crop yields. Small and marginal farmers cannot get expert advice in time. In this paper, we propose a mobile application known as SmartAgri Doctor, implemented using Flutter and TensorFlow Lite, to detect crop diseases in real time. The proposed system works on the concept of transfer learning with MobileNet V2 and the Plant Village dataset to detect diseases in crops like tomato, rice, brinjal, and sugarcane. TensorFlow Lite's model is quite light, and the proposed system runs in full offline mode. Once the disease has been identified, recommendations are provided for both organic and chemical treatments depending upon the severity of the disease. The results show that the validation accuracy of the proposed system is around 93.5%. It runs in real time and can be executed in normal Android devices. The proposed system provides a scalable, accessible, and sustainable solution for smart agriculture.
Keywords
Crop Disease Detection, MobileNet V2, TensorFlow Lite, Flutter, Smart Agriculture, On-Device AI, Plant Village Dataset
How to cite this paper
@article{1715542,
author = {Dinakaran k, Ponmozhi K},
title = {An AI-Powered Mobile Application for Crop Disease Detection Using MobileNetV2 and Flutter},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {9},
pages = {2705-2709},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1715542.pdf},
abstract = {Agriculture forms the basis of the Indian economy. However, crop disease has been a major problem in reducing crop yields. Small and marginal farmers cannot get expert advice in time. In this paper, we propose a mobile application known as SmartAgri Doctor, implemented using Flutter and TensorFlow Lite, to detect crop diseases in real time. The proposed system works on the concept of transfer learning with MobileNet V2 and the Plant Village dataset to detect diseases in crops like tomato, rice, brinjal, and sugarcane. TensorFlow Lite's model is quite light, and the proposed system runs in full offline mode. Once the disease has been identified, recommendations are provided for both organic and chemical treatments depending upon the severity of the disease. The results show that the validation accuracy of the proposed system is around 93.5%. It runs in real time and can be executed in normal Android devices. The proposed system provides a scalable, accessible, and sustainable solution for smart agriculture.},
keywords = {Crop Disease Detection, MobileNet V2, TensorFlow Lite, Flutter, Smart Agriculture, On-Device AI, Plant Village Dataset},
month = {March},
doi = {https://doi.org/10.64388/IREV9I9-1715542}
}