Classification of Mint Leaf Types Based on the Image Using Euclidean Distance and K-Means Clustering with Shape and Texture Feature Extraction

  • Trinugi Wira Harjanti Sekolah Tinggi Teknologi Informasi NIIT
  • Hari Setiyani Sekolah Tinggi Teknologi Informasi NIIT
  • Joko Trianto Sekolah Tinggi Teknologi Informasi NIIT
  • Yuri Rahmanto Universitas Teknokrat Indonesia

Abstract

Mint is a plant that has many benefits and uses. However, some people are not familiar with the types of mint leaves because they cannot tell the difference. Actually, if you look closely, mint leaves have their own characteristic shape and texture. However, most people judge mint leaves to have a shape similar to other leaves so it is difficult to tell them apart. This paper aims to classify the types of mint leaves using the Euclidean distance algorithm and K-Means clustering with shape and texture feature extraction. The K-Means Clustering Algorithm functions as a segmentation so that the image to be classified can be separated from other objects. In the feature extraction process, metric and eccentricity parameters are used. Meanwhile, for texture feature extraction, use the parameters in the Gray Level Co-occurence Matrix (GLCM). Furthermore, the classification process uses the Euclidean Distance algorithm which has a function to represent the level of similarity between two images by taking into account the distance value from the identified image. Based on the results of the evaluation using a confusion matrix by calculating precision, recall and accuracy, the precision value is 82%, recal is 84% ​​and accuracy is 83%.

Published
2022-03-25
How to Cite
HARJANTI, Trinugi Wira et al. Classification of Mint Leaf Types Based on the Image Using Euclidean Distance and K-Means Clustering with Shape and Texture Feature Extraction. Tech-E, [S.l.], v. 5, n. 2, p. 115-124, mar. 2022. ISSN 2581-1916. Available at: <https://jurnal.buddhidharma.ac.id/index.php/te/article/view/940>. Date accessed: 27 sep. 2022. doi: https://doi.org/10.31253/te.v5i1.940.
Section
Articles

Most read articles by the same author(s)