Prediksi Tingkat Kematangan Buah Apel berbasis GLCM, CIE LAB, dan Machine Learning

Vieri, Gabriel (2022) Prediksi Tingkat Kematangan Buah Apel berbasis GLCM, CIE LAB, dan Machine Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 02311740000072-Undergraduate_Thesis.pdf - Accepted Version Download (2MB)

Abstract

Buah apel sebagai salah satu komoditas pertanian yang disukai karena tekstur, gizi, rasa, dan daya tarik visual. Apel dapat diklasifikasi berdasarkan tingkat kematangannya dari tekstur dan warna kulit buah apel. Analisa tekstur dilakukan dengan ekstraksi fitur Gray Level Co occurence Matrix (GLCM), skala warna CIE Lab untuk menganalisa warna, dan mengkombinasikan fitur keduanya dalam algoritma machine learning. Penentuan tingkat kematangan buah apel secara fisiologis dilakukan dengan menghitung starch conversion rate yang terbentuk akibar reaksi dengan larutan iodine. Gambar dilakukan preprocessing dengan crop dan resize. Fitur yang diekstrak oleh GLCM pada jarak 1, 2, 3 dan sudut 0O , 45O , 90O , 135O adalah kontras, ASM, energi, homogeneitas, dan disimilaritas. Pelatihan model untuk klasifikasi tingkat kematangan dilakukan dengan skenario penggunaan seleksi fitur dan reduksi dimensi sebelum masuk sepuluh algoritma machine learning. Seleksi fitur dilakukan dengan menghitung Z-value pada Standard Error of difference means (SEd). Untuk reduksi dimensi fitur dilakukan analisa PCA. Hasilnya adalah dengan algoritma machine learning seperti Nearest Neighbors, Linear SVM, Gaussian Process, Decision Tree, Random Forest, Neural Net, AdaBoost, Naive Bayes, dan QDA didapatkan akurasi 100%. Sehingga klasifikasi apel berdasarkan tingkat kematangan dapat dilakukan dengan menggunakan GLCM, Z-value SEd, dan analisa PCA.
====================================================================================================================================
Apples are one of the preferred agricultural commodities because of their texture, nutrition, taste, and visual appeal. Apples can be classified based on the level of maturity of the texture and color of the skin of the apple. Texture analysis is performed by feature extraction of Gray Level Co-occurence Matrix (GLCM), CIE Lab color scale to analyze color, and combining both features in machine learning algorithms. Determination of the level of physiological maturity of apples is done by calculating the starch conversion rate formed by the reaction with iodine solution. The image is preprocessed by cropping and resizing. The features extracted by GLCM at distances 1, 2, 3 and angles 0O, 45O, 90O, 135O are contrast, ASM, energy, homogeneity, and dissimilarity. Model training for maturity level classification is carried out by using feature selection and dimension reduction scenarios before entering ten machine learning algorithms. Feature selection is done by calculating the Z-value on the Standard Error of difference means (SEd). For feature dimension reduction, PCA analysis was performed. The result is that machine learning algorithms such as Nearest Neighbors, Linear SVM, Gaussian Process, Decision Tree, Random Forest, Neural Net, AdaBoost, Naive Bayes, and QDA get 100% accuracy. Therefore, the classification of apples based on the level of ripeness can be done using GLCM, Z-value SEd, and PCA analysis.

Item Type:	Thesis (Other)
Additional Information:	RSF 006.31 Vie p-1 2022
Uncontrolled Keywords:	CIE Lab, GLCM, kematangan apel, machine learning, apple maturity
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	- Davi Wah
Date Deposited:	27 Sep 2024 08:54
Last Modified:	27 Sep 2024 08:54
URI:	http://repository.its.ac.id/id/eprint/115682

Actions (login required)

View Item