Please use this identifier to cite or link to this item:
Title: Vectored Machine Learning Rearing Process: Early Detection of Leaf Diseases
Authors: Prasad, B Rajendra
Ramashri, T
Naidu, K Rama
Keywords: Clustering;Disease Severity;Plant leaf analyses;Quantification;Recognition
Issue Date: Jul-2020
Publisher: NISCAIR-CSIR, India
Abstract: Over the past years, the plant leaf analyses through image processing have drawn a remarkable approach in assessing leaf disease severity through accurate and precise conclusions. We proposed, ‘Scale Invariant Feature Transform’ (SIFT) based Distinctive Scale Invariant Mapping Procedure (DSIMP) for training images. Random Separation Propagation (RSP) Procedure and Redundant multiclass Support Vector Machine (RM-SVM) are implemented to detect the rice and groundnut leaf diseases at its early stages. Discriminative Gray Level Co-occurrence Matrix (DGLCM) and K means clustering is used for recognition and quantification to give the best color analysis. Experiments with 1000 samples of rice and groundnut leaf images show promising performance.
Page(s): 619-625
ISSN: 0975-1084 (Online); 0022-4456 (Print)
Appears in Collections:JSIR Vol.79(07) [July 2020]

Files in This Item:
File Description SizeFormat 
JSIR 79(7) 619-625.pdf380.53 kBAdobe PDFView/Open

Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.