An Intelligent Image Analysis Approach for Early Detection of Potato Diseases Using Machine Learning
Potato production is significantly affected by foliar diseases that can reduce crop quality and productivity when identification and intervention are delayed. Conventional disease diagnosis commonly depends on visual inspection, which may be influenced by environmental conditions, observer expertise, and the similarity of symptoms across disease categories. This paper presents an intelligent image analysis approach for early detection of potato diseases using machine learning. The proposed approach integrates digital image preprocessing, discriminative feature extraction, machine-learning-based classification, and disease-severity analysis into a unified analytical framework. The methodological foundation is derived from prior research on potato disease recognition, image segmentation, support vector machines, convolutional neural networks, clustering, local binary patterns, and digital image processing. Particular emphasis is placed on the applicability of convolutional representations for learning disease-related visual patterns and on the role of segmentation and feature engineering in improving classification reliability. The PlantVillage dataset is considered as a suitable image source for controlled experimental development, while studies conducted under uncontrolled environmental conditions highlight the importance of robustness beyond laboratory settings. The resulting framework provides a structured pathway from raw leaf images to disease classification and supports early agricultural decision-making. Analytical findings indicate that combining preprocessing, feature representation, and machine-learning classification is more appropriate than relying on a single computational technique. The study also identifies illumination variation, background complexity, symptom similarity, dataset bias, and generalization to field environments as important limitations requiring further investigation.
Authors
- Rohan Kulkarni
- Neha Verma
Publication Details
- Journal
- International Journal of Advanced Artificial Intelligence Research
- Published
- 2026-11-20
- DOI
- https://doi.org/10.55640/ijaair-v03i11-09
- Primary Topic
- Smart Agriculture and AI
- Type
- article
- Field-Weighted Citation Impact
- 0.00