Machine Intelligence for the Detection of Plant Diseases Using Image Processing

EasyChair Preprint 1634

Abstract

The economic growth of the country depends upon the quality and quantity of the plant production. The diagnosis of the diseases is needed at an earlier stage before it becomes worse. Proper preventive care will increase the production of fruits and vegetables. Thus, automated recognition of diseases in plant plays a vital role in agricultural yields. Several image processing approaches have been developed for disease diagnosis in a short span. In this work, various machine intelligence techniques include SVM, Back Propagation network, Naïve Bayesian are analyzed and observed and concur Gray Level Co-occurrence Matrix (GLCM) based neural network performs better and gives high accuracy. The experimental results validate that the proposed model achieved a test accuracy is found that to around 93% for alternaria alternate, anthracnose, black spot, bacterial blight, cercospora leaf spot diseases in fruit and leaf.

Keyphrases: Classification, Gray Level Co-occurrence Matrix (GLCM), Plant disease detection, image processing

@booklet{EasyChair:1634,
  author    = {Liba Manopriya J and Arockia Jansi Rani and M.Asha Paul},
  title     = {Machine Intelligence for the Detection of Plant Diseases Using Image Processing},
  howpublished = {EasyChair Preprint 1634},
  year      = {EasyChair, 2019}}