IMPLEMENTASI IMAGE PROCESSING DENGAN METODE LEARNING VECTOR QUANTIZATION UNTUK APLIKASI PENGENALAN BUAH-BUAHAN

Winarto, Winarto (2020) IMPLEMENTASI IMAGE PROCESSING DENGAN METODE LEARNING VECTOR QUANTIZATION UNTUK APLIKASI PENGENALAN BUAH-BUAHAN. Tugas Akhir thesis, University of Technology Yogyakarta.

[img]
Preview
Text
NASKAH PUBLIKASI-WINARTO-5150411300.pdf

Download (1MB) | Preview

Abstract

An ability to recognize objects is indispensable for human beings at this time. A discovery of a new recognition method can simplify human daily activities. Pattern recognition is often applied to a variety of objects, such as pattern recognition in fruit types. The limitations of human memory in remembering many things become a weak point of human beings. Based on these issues, the research is conducted to identify the types of fruits based on their features using digital image processing techniques and Learning Vector Quantization (LVQ) methods. The LVQ is a classification method in which a system works with training and testing stages that will produce classes in the form of apples, oranges, and bananas. Image Processing is used to obtain the parameter values from fruit images, using the RGB (Red, Green, and Blue) average color extraction features and the shapes (area and perimeter) with 5 parameters. This application uses 3 weight initialization data imageries, 30 training data imageries, and 15 test data imageries. With 0.025, 0075, 0.05 and 0.1 learning rate, the reduction of learning rate is 0.1, minimum 0.01, 0.001, 0.0001 and 0.00001 with a maximum number of iterations/Epoch 10. The level of accuracy generated by the system as much as 99.85375% for data obtained from https://www.kaggle.com/moltean/fruits/data and accuracy of 70% obtained from original fruit imagery photographed by the researcher himself. Keywords: Fruits, Image Processing, Learning Vector Quantization, Pattern Recognition.

Item Type: Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Sains Dan Teknologi > S1 Informatika
Depositing User: Kaprodi S1 Informatika UTY
Date Deposited: 26 Mar 2020 07:38
Last Modified: 26 Mar 2020 07:38
URI: http://eprints.uty.ac.id/id/eprint/4902

Actions (login required)

View Item View Item