Pengaruh Ciri Tekstur Pada Metode Klasifikasi LVQ Untuk Hasil Akurasi Identifikasi Citra Batik Tradisional Solo

Noor Abdul Haris, Kusrini Kusrini, Hanif Al Fatta

Abstract


The use of appropriate features to determine identification is very important for the results of accuracy, especially the image of traditional batik in Solo. Characteristics are considered as good if they have distinguishing abilities, so they can be used for recognition with a high degree of accuracy. The purpose of this study was to determine what texture characteristics affect the level of accuracy in the identification of traditional batik in Solo. The method used is KFold cross-validation, with a K value of 2, 3, 5, 6, 10 used to test the training data and validation data. The best parameters are chosen from the highest accuracy value, and then the best final weight, value α, dec α, and min α are stored. The resulting values are used in testing the data. The texture characteristics tested were 0o energy, 45o energy, 90o energy, 135o energy, 0o entropy, 45o entropy, 90o entropy, 135o entropy, 0o contrast, 45o contrast, 90o contrast, 135o energy, 0o homogeneity, 45o homogeneity, 90o homogeneity. , homogeneity 135 o, correlation 0 o, correlation 45 o, correlation 90 o, correlation 135o. The test results obtained from K = 2, 3, 5, 6, 10 is the highest accuracy value obtained at 10-fold with an accuracy of 80.2%. The results of testing using k-fold can be concluded that the use of texture characteristics can affect the accuracy of 18.89%.

Keywords


K-Fold; Batik; Texture; LVQ.

Full Text:

PDF


DOI: https://doi.org/10.35970/infotekmesin.v11i2.265

Article Metrics

Abstract view : 0 times
PDF - 0 times

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Indexed by:

 
   

 

Editorial Office:

Pusat Penelitian dan Pengabdian Masyarakat (P3M)
Politeknik Negeri Cilacap
Jln. Dokter Soetomo No.1, Karangcengis Sidakaya Cilacap
Jawa Tengah 53212
Telepon: (0282) 533329
Email: infotekmesin.ejournal@pnc.ac.id

 

   Flag Counter

 

View My Stats

 

Creative Commons License
INFOTEKMESIN is licensed under a Creative Commons Attribution 4.0 International License.