Plant Leaf Disease Detection using SVM

Abstract

Agriculture and crop production play an important role in our everyday lives.The primary source of food and clothing is agriculture.Food and clothing are currently being lost due to contaminated crops that decrease production rates.There are a variety of diseases that affect the plant's leaves, fruits, and stem.Bacteria, fungi, viruses, and other microorganisms are the most common causes of plant disease.Diseases are often difficult to monitor, and observations made with the naked eye are unreliable in detecting them.If we can't detect the disease quickly, we won't be able to take the appropriate action.The quality and quantity of goods would improve if pesticides are used less in agriculture.The technique is now mostly used in image processing for the identification of plant diseases.The technology is used in this method for detecting and classifying leaf diseases using SVM classification.Image acquisition, image preprocessing, feature extraction, and classification are all steps in this technology.Banana, pepper, and rice were the three crops we used.A total of 400 leaf sample images were used.From there, 80% will be used for preparation and 20% for research.With an accuracy of 92.99 percent, this device can successfully diagnose and identify the disease.

Plant phenotyping relevance

植物葉の画像から病害状態を抽出・分類するSVM手法が研究の中心であり、画像取得から特徴抽出、分類、精度評価まで記述されているため、植物フェノタイピング手法として含める。

Code and data availability

The paper describes an SVM-based leaf disease detection system using ~400 leaf images, but provides no public dataset deposit, no author code/model release, and no availability statements. Images were gathered from unspecified websites and field visits, with no repository or URL given.

No evidence-backed public reproduction asset is currently recorded.