Feature Extraction of Gram-Negative Bacteria Texture Using Grey Level Co-Occurrence Matrix and Scale-Invariant Feature Transform

DOI

10.2991/assehr.k.200303.019How to use a DOI?

Keywords

Gram-negative bacteria, segmentation, Grey Level co-occurrence matrix, scale-invariant feature transform

Abstract

Gram-negative bacteria are one of the pathogenic bacteria in the respiratory tract. The presence of these bacteria should be identified correctly so that a doctor can do the proper handling of antibiotic therapy. Observations are still made under a microscope by a microbiology clinic or hospital team. This research uses image processing to replace visual inspection. One of the stages in image processing is feature extraction. This research uses a texture approach that is the Grey Level co-matrix matrix and Scale-Invariant Feature Transform. The first method uses the grey level approach, while the second method uses the texture object direction approach. Data used in this research get from a sample of 50 patients. It exposed to Gram-negative pathogenic bacteria. The selected bacteria are Klebsiella pneumonia and Pseudomonas aeruginosa. Of the two methods used, the results were compared to obtain information on which method was more suitable for the observation of Gram-negative bacterial objects. The results of the view by examining the effect of object features with ground truth found that the level of accuracy using the grayscale with the similarity of objects approaching and using the object texture direction approach obtained an accuracy of 89%.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Budi Dwi Satoto
AU  - Imam Utoyo
AU  - Riries Rulaningtyas
PY  - 2020
DA  - 2020/03/06
TI  - Feature Extraction of Gram-Negative Bacteria Texture Using Grey Level Co-Occurrence Matrix and Scale-Invariant Feature Transform
BT  - Proceedings of the 1st International Multidisciplinary Conference on Education, Technology, and Engineering (IMCETE 2019)
PB  - Atlantis Press
SP  - 70
EP  - 74
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.200303.019
DO  - 10.2991/assehr.k.200303.019
ID  - Satoto2020
ER  -