Ekstraksi Fitur Citra Tanda Nomor Kendaraan Bermotor Dengan Metode Moment Invariant

  • Andi Yulia Muniar STMIK AKBA
Keywords: moment invariant, otsu, euclidean distance, ekstraksi fitur, resize, segmentasi citra, akurasi

Abstract

Every country has a different feature of vehicle license plate, it is a challenge to extract the feature of vehicle license plate which is needed an accurate method to extract the features of vehicle license plate. The aims of the study were to implement and find the level of accuracy feature extraction of license plate using moment invariant. The data was obtained through Library Research. The method used in this research was Moment Invariant used for the process of the feature extract and Euclidean. The result of the study indicated that the test result obtained the total of the accuracy level was 78,43%. The best accuracy level was 90% obtained from level of accuracy obtained from the testing based on the the distance, the smallest accuracy level was 43,75% obtained from the testing based on images skewness and the nonstandard characters. The dark images obtained the accuracy level 50%.

Published
2020-04-30