Scholarlink

The recognition of fruit diseases is rapidly becoming a significant area in computer vision. Besides reducing fruit production, plant diseases also have a profound impact on the country's economy. Therefore, predicting plant illnesses and implementing remedial actions are of utmost importance. Plant illness directly affects the crop’s quality and overall productivity. To achieve swift and accurate disease prediction, an automated method is used to identify and diagnose affected leaves within a short span of time. Through the appropriate use of diagnosis and classification techniques, potential losses can be prevented. The features used for identifying plant diseases include color, strength, area of attack, and dimension. The current research primarily focuses on guava leaf disease prediction using K-means clustering techniques. In this proposed study, we aim to identify orange fruit and leaf diseases. To achieve this, K-means clustering is applied to segment the orange fruit images into distinct clusters. We leverage the gray-level co-occurrence matrix (GLCM) approach with the clustered data to extract the global features as feature vectors, and the orientation gradient approach is utilized to uncover real features. Implementing machine learning algorithms has significantly enhanced the accuracy of the system, achieving a remarkable 96% accuracy rate.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning Based Citrus Fruit Disease Prediction Using Support Vector Machine Algorithm

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • D. Mohanapriya,
  • S. K. Kabilesh,
  • R. Swaranambigai,
  • B. Saritha,
  • S. V. K. Deepikapriya

摘要

The recognition of fruit diseases is rapidly becoming a significant area in computer vision. Besides reducing fruit production, plant diseases also have a profound impact on the country's economy. Therefore, predicting plant illnesses and implementing remedial actions are of utmost importance. Plant illness directly affects the crop’s quality and overall productivity. To achieve swift and accurate disease prediction, an automated method is used to identify and diagnose affected leaves within a short span of time. Through the appropriate use of diagnosis and classification techniques, potential losses can be prevented. The features used for identifying plant diseases include color, strength, area of attack, and dimension. The current research primarily focuses on guava leaf disease prediction using K-means clustering techniques. In this proposed study, we aim to identify orange fruit and leaf diseases. To achieve this, K-means clustering is applied to segment the orange fruit images into distinct clusters. We leverage the gray-level co-occurrence matrix (GLCM) approach with the clustered data to extract the global features as feature vectors, and the orientation gradient approach is utilized to uncover real features. Implementing machine learning algorithms has significantly enhanced the accuracy of the system, achieving a remarkable 96% accuracy rate.