KLASIFIKASI JENIS BUAH KURMA MENGGUNAKAN METODE GRAY LEVEL CO-OCCURANCE MATRIX DAN K-NEAREST NEIGHBOR

BUDIASA, AHLAQ (2023) KLASIFIKASI JENIS BUAH KURMA MENGGUNAKAN METODE GRAY LEVEL CO-OCCURANCE MATRIX DAN K-NEAREST NEIGHBOR. Other thesis, UPN "Veteran" Yogyajarta.

	Text ABSTRAK SKRIPSI_AHLAQ BUDIASA_123180021.pdf Download (35kB)
	Text COVER SKRIPSI_AHLAQ BUDIASA_123180021.pdf Download (150kB)
	Text DAFTAR ISI SKRIPSI_AHLAQ BUDIASA_123180021.pdf Download (225kB)
	Text DAFTAR PUSTAKA SKRIPSI_AHLAQ BUDIASA_123180021.pdf Download (203kB)
	Text HALAMAN PENGESAHAN PEMBIMBING SKRIPSI_AHLAQ BUDIASA_123180021.pdf Download (284kB)
	Text HALAMAN PENGESAHAN PENGUJISKRIPSI_AHLAQ BUDIASA_123180021.pdf Download (342kB)
	Text SKRIPSI FULLTEXT_AHLAQ BUDIASA_123180021.pdf Restricted to Repository staff only Download (2MB)

Abstract

ABSTRAK
Tujuan : Perancangan sistem yang dapat mengklasifikasikan jenis buah kurma
menggunakan metode Gray Level Co - Occurance Matrix (GLCM) dan K-Nearest Neighbor
(K-NN) berdasarkan nilai warna dan tekstur. Pengujian tingkat akurasi pada penggunaan
metode GLCM dan KNN dalam klasifikasi jenis buah kurma
Perancangan/metode/pendekatan : Penerapan metode Gray level Co-Occurence Matrix
(GLCM) sebagai ekstraksi fitur tekstur dan K-Nearest Neighbor (K-NN) sebagai metode
klasifikasi.
Hasil : Penerapan ekstraksi fitur tekstur pada metode Gray Level Co-Occurance Matrix
(GLCM) menggunakan nilai besaran Angular Second Moment (ASM), Contrast, Inverse
Different Moment (IDM), dan Correlation serta ekstraksi ciri RGB dengan nilai rata-rata
Red, Green, dan Blue yang dapat berfungsi dengan baik sebagai parameter untuk klasifikasi
menggunakan metode K-Nearest Neighbor. Pada proses pengujian model menggunakan
perhitungan metode confussion matrix didapatkan nilai rata-rata akurasi pada K=3 sebesar
96%, nilai rata-rata presisi pada K=3 sebesar 100%, dan nilai rata-rata recall pada K=3
sebesar 98%. Hasil ini didapatkan dengan menggunakan data sejumlah 320 data. Data
tersebut terbagi menjadi 3 label, yaitu 1000 data citra Kurma Ajwa , 1000 data citra Kurma
Medjool dan 1000 data citra Kurma Rutab. Total data tersebut terbagi menjadi 2 jenis data,
yaitu data training dan data testing. Perbandingan jumlah data training dan data testing, yaitu
80%:20%. Jumlah data training yang digunakan 2400 data dan jumlah data testing sebanyak
600.
Keaslian/ State Of The Art : Penelitian ini dirancang untuk mengklasifikasikan jenis dan
kualitas biji kopi menggunakan metode Gray Level Co-Occurance Matrix (GLCM) dan K-
Nearest Neighbor (K-NN) dengan menggunakan buah kurma yang di ambil dari Kaggle
Kata Kunci : Klasifikasi, Kurma, Grey level Co-occurence Matrix, K-Nearest Neighbor
ABSTRACT
Purpose : The design of a system that can classify the type dates fruit using the Gray
Level Co-Occurance Matrix (GLCM) and K-Nearest Neighbor (K-NN) methods based on
color and texture values. Testing the level of accuracy of the use of the GLCM and KNN
methods in the classification of the type dates fruit.
Design/Method/Approach : Application of Gray level CoOccurrence Matrix (GLCM)
method as texture feature extraction and K-Nearest Neighbor (K-NN) as classification
method.
Result : The application of texture feature extraction to the Gray Level Co-Occurance
Matrix (GLCM) method uses Angular Second Moment (ASM), Contrast, Inverse Different
Moment (IDM), and Correlation and RGB feature extraction values with an average value
of Red, Green, and Blue can function well as a parameter for classification using the K-
Nearest Neighbor method. In the process of testing the model using the confusion matrix
calculation method, the average accuracy value at K=3 is 96%, the average precision value
at K=3 is 100%, and the average recall value at K=3 is 98%. These results were obtained
using data of 320 data. The data is divided into 3 labels, namely 1000 data of the Ajwa Date
image, 1000 image data of the Medjool Date and 1000 image data of the Rutab Date. The
total data is divided into 2 types of data, namely training data and testing data. Comparison
of the amount of training data and test data, namely 80%:20%. The amount of training data
used is 2400 data and the amount of test data is 600.
Originality/Value/State Of The Art : This study was designed to classify the type and
quality of coffee beans using the Gray Level Co-Occurance Matrix (GLCM) and K-Nearest
Neighbor (K-NN) methods using dates taken from kaggle.
Keywords : Classifications, Dates, Grey level Co-occurence Matrix, K-Nearest Neighbor

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Classifications, Dates, Grey level Co-occurence Matrix, K-Nearest Neighbor
Subjects:	Z Bibliography. Library Science. Information Resources > ZA Information resources
Divisions:	Faculty of Engineering, Science and Mathematics > School of Engineering Sciences
Depositing User:	Eko Yuli
Date Deposited:	01 Dec 2023 04:21
Last Modified:	01 Dec 2023 04:21
URI:	http://eprints.upnyk.ac.id/id/eprint/38270

Actions (login required)

View Item