Farm Ease App Crop Information And Disease Prediction Using Machine Learning For Farmer Info

Abstract

Crop disease prediction remains a major challenge in precision agriculture, and numerous methods have been developed and evaluated to tackle this problem. Because plant diseases are affected by factors such as climate, weather conditions, soil characteristics, fertilization practices, and seed varieties, accurate prediction requires the integration of multiple datasets. This demonstrates that plant disease prediction is a complex process involving several interconnected stages rather than a simple, direct task. Farmers are central to the agricultural ecosystem, and agriculture plays a vital role in national development by contributing significantly to a country’s Gross Domestic Product (GDP). The overall performance of agriculture largely depends on farmers who cultivate and manage crops. To assist them, a real-time plant disease prediction prototype was developed using the Python programming language, integrating hybrid machine learning techniques with data analysis. Agricultural productivity is closely tied to economic growth, and due to the widespread occurrence of plant diseases, early detection is essential for sustaining the agricultural sector. If plant diseases are not addressed in a timely manner, they can severely affect crop quality and yield. This research proposes an automated image segmentation–based approach for detecting and classifying plant leaf diseases and provides a review of various disease classification techniques. Image segmentation, which is a crucial step in leaf disease detection, is carried out using a genetic algorithm.

Plant phenotyping relevance

植物葉の病徴を画像から検出・分類する自動セグメンテーション手法とリアルタイム予測プロトタイプが研究の中心であり、植物の病害状態を直接推定するため。

Code and data availability

The paper describes a plant disease prediction app using Kaggle-sourced leaf images and the DiaMOS plant dataset, but provides no authors' public dataset, code, model, or supplement with availability language or URLs. The mentioned datasets are third-party inputs, not paper-specific public assets.

No evidence-backed public reproduction asset is currently recorded.