Kisan Seva: Transfer Learning-Based Plant Disease Detection using Mobilenetv2 with Comparative Evaluation of Traditional Machine Learning Techniques on the Plantvillage Dataset
- Version
- Download 26
- File Size 307.76 KB
- File Count 1
- Create Date 20 February 2026
- Last Updated 20 February 2026
Kisan Seva: Transfer Learning-Based Plant Disease Detection using Mobilenetv2 with Comparative Evaluation of Traditional Machine Learning Techniques on the Plantvillage Dataset
Atharva Shivade, Prashant Jadhav, Makarand Shimpi, Vaibhav Sanap
Department of Computer Engineering (SOCSE) Sandip University
Mahravni, Nashik,India
Emails: shivadeatharva1450@gmail.com , prashantjadhav110301@gmail.com, makarand0604@gmail.com ,
vbsanap2004@gmail.com
Abstract
Early and accurate detection of plant diseases is essential for improving agricultural productivity, reducing crop losses, and ensuring global food security. Traditional disease identification methods rely heavily on manual visual inspection, which is time-consuming, subjective, and impractical for large-scale farming environments. Recent advancements in artificial intelligence, particularly deep learning and computer vision, provide promising automated solutions for real-time plant disease diagnosis. This study proposes Kisan Seva, an AI-driven plant disease detection framework based on transfer learning using the MobileNetV2 convolutional neural network, designed to achieve high classification performance while maintaining computational efficiency suitable for deployment on resource-constrained agricultural devices. The system is trained and evaluated using the widely adopted PlantVillage dataset. To rigorously validate the effectiveness of the proposedapproach, a comparative evaluation is conducted against traditional machine learning classifiers, including Support Vector Machine (SVM) and k-Nearest Neighbor (KNN). The framework further incorporates infected area estimation to provide quantitative disease severity analysis. The proposed system aims to contribute toward AI-enabled precision agriculture and farmer-centric decision support systems.
Keywords—Plant Disease Detection, Deep Learning, Transfer Learning, MobileNetV2, Machine Learning, Precision Agriculture.
Download