Open AccessPlant Classification Based on Extraction Feature Gray Level Co-Occurrence Matrix Using k-nearest Neighbour
Author(s) -
Fuzy Yustika Manik,
Singgih Saputra,
Dewi Sartika Br Ginting
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1566/1/012107
Subject(s) - gray level , pattern recognition (psychology) , co occurrence matrix , artificial intelligence , feature extraction , classifier (uml) , k nearest neighbors algorithm , nearest neighbour , computer science , mathematics , geography , image (mathematics) , image processing , image texture
Indonesia is one of the countries with high plant diversity. Almost every region in Indonesia has distinctive plants and may not be present in other countries. Based on these facts required a strategic step to record and identify plants in Indonesia. One method that can be used to leaf image feature extraction is the Gray Level Co-occurrence Matrix (GLCM). This research will implement k-Nearest Neighbor (k-NN) method to classify type of plants based on leaf texture. The classification result based on GLCM using k-NN classifier showed that the accuracy using k = 3 was 83%. The use of parameter k influence classification results, the greater the value of k then the accuracy would be smaller. Classification errors for some types of leaf images occurred because the value extraction traits generated by GLCM was very similar and had a small range of values.