DETEKSI DAN IDENTIFIKASI CITRA DIGITAL JENIS BERAS MENGGUNAKAN METODE ANFIS DAN PREWITT

  • Nisrina Hasna Nataraharja Universitas Singaperbangsa Karawang
  • Riza Ibnu Adam Universitas Singaperbangsa Karawang
  • Garno Garno Universitas Singaperbangsa Karawang

Abstract

Indonesia is among the third most rice producing countries in the world. Thanks to that there are many types of rice in Indonesia. This study aims to identify types of rice using the Adaptive Neuro Fuzzy Inference (ANFIS) method that can be trained to identify types of rice. Using the values ​​contained in the form feature extraction which includes metric and eccentricity and texture feature extraction which includes LBP, contrast, correlation, and energy. This study also aims to determine the effect of adding Prewitt edge detection to accuracy in identification of rice types. This edge detection is added to the extraction of shape characteristics. Extraction values ​​were taken from 100 training images and 100 test images with 25 images per rice type each. The results of the research conducted showed that ANFIS proved to be quite good in identifying types of rice with an average accuracy above 70%, while the use of Prewitt edge detection influences 1-5% of accuracy.

Downloads

Download data is not yet available.
Published
Nov 24, 2020
How to Cite
NATARAHARJA, Nisrina Hasna; ADAM, Riza Ibnu; GARNO, Garno. DETEKSI DAN IDENTIFIKASI CITRA DIGITAL JENIS BERAS MENGGUNAKAN METODE ANFIS DAN PREWITT. INFOMATEK: Jurnal Informatika, Manajemen dan Teknologi, [S.l.], v. 22, n. 2, p. 101 - 108, nov. 2020. ISSN 2614-7807. Available at: <https://journal.unpas.ac.id/index.php/infomatek/article/view/3250>. Date accessed: 23 jan. 2021. doi: http://dx.doi.org/10.23969/infomatek.v22i2.3250.