IMPLEMENTASI ALGORITMA RANDOM FOREST DENGAN LOCAL BINARY PATTERN DAN HUE SATURATION VALUE UNTUK DETEKSI KESEGARAN SAYUR DAN BUAH

Genta Alima Persada, - (2025) IMPLEMENTASI ALGORITMA RANDOM FOREST DENGAN LOCAL BINARY PATTERN DAN HUE SATURATION VALUE UNTUK DETEKSI KESEGARAN SAYUR DAN BUAH. S1 thesis, Universitas Pendidikan Indonesia.

Abstract

Sebagai negara beriklim tropis, Indonesia menghadapi tantangan besar dalam menjaga kualitas sayur dan buah, terutama selama proses distribusi. Suhu dan kelembapan yang tinggi mempercepat penurunan kualitas, yang berujung pada kerugian finansial signifikan bagi para petani. Masalah ini diperparah oleh proses penentuan kesegaran secara manual yang seringkali subjektif dan tidak efisien, menyebabkan kerugian pasca-panen yang lebih besar. Penelitian ini bertujuan merancang dan mengevaluasi sistem deteksi kesegaran otomatis menggunakan machine learning. Metode yang diimplementasikan adalah algoritma Random Forest (RF) yang diperkuat dengan kombinasi ekstraksi fitur Local Binary Pattern (LBP) untuk analisis tekstur dan Hue Saturation Value (HSV) untuk analisis warna. Proses penelitian meliputi pengumpulan dataset, pra-pemrosesan citra, ekstraksi fitur, dan evaluasi model menggunakan metrik accuracy, precision, recall, dan F1-score. Hasil penelitian menunjukkan bahwa penerapan ekstraksi fitur Local Binary Pattern (LBP) dan Hue Saturation Value (HSV) memberikan peningkatan kinerja yang signifikan. Akurasi model berhasil meningkat dari 96,26% (tanpa ekstraksi fitur) menjadi 99,93%. Disimpulkan bahwa kombinasi algoritma Random Forest dengan ekstraksi fitur Local Binary Pattern (LBP) dan Hue Saturation Value (HSV) sangat efektif dan mampu menghasilkan klasifikasi kesegaran dengan tingkat akurasi sangat tinggi, menawarkan solusi objektif dan akurat untuk masalah penyortiran manual. --------- As a tropical country, Indonesia faces significant challenges in maintaining the quality of fruits and vegetables, especially during the distribution process. High temperatures and humidity accelerate quality degradation, leading to significant financial losses for farmers. This problem is exacerbated by the manual freshness assessment process, which is often subjective and inefficient, causing greater post-harvest losses. This research aims to design and evaluate an automated freshness detection system using machine learning. The implemented method is the Random Forest (RF) algorithm, enhanced by a combination of Local Binary Pattern (LBP) for texture analysis and Hue Saturation Value (HSV) for color analysis. The research process includes dataset collection, image pre-processing, feature extraction, and model evaluation using accuracy, precision, recall, and F1-score metrics. The results show that the application of Local Binary Pattern (LBP) and Hue Saturation Value (HSV) feature extraction provides a significant performance improvement. The model's accuracy successfully increased from 96.26% (without feature extraction) to 99.93%. It is concluded that the combination of the Random Forest algorithm with Local Binary Pattern (LBP) and Hue Saturation Value (HSV) feature extraction is highly effective and capable of producing freshness classifications with a very high accuracy rate, offering an objective and accurate solution to the problem of manual sorting.

	Text S_RPL_2100245_Title.pdf Download (1MB)
	Text S_RPL_2100245_Chapter 1.pdf Download (268kB)
	Text S_RPL_2100245_Chapter 2.pdf Restricted to Staf Perpustakaan Download (532kB) | Request a copy
	Text S_RPL_2100245_Chapter 3.pdf Download (548kB)
	Text S_RPL_2100245_Chapter 4.pdf Restricted to Staf Perpustakaan Download (1MB) | Request a copy
	Text S_RPL_2100245_Chapter 5.pdf Download (188kB)
	Text S_RPL_2100245_Appendix.pdf Restricted to Staf Perpustakaan Download (1MB) | Request a copy

Item Type:	Thesis (S1)
Additional Information:	https://scholar.google.com/citations?hl=en&user=zU2unHsAAAAJ&scilu=&scisig=ACUpqDcAAAAAaJRiSEPAFhU93Q_G7s9RIDE2QT4&gmla=AH8HC4yfUPvhNexQUE8L4VtgA6GdJWU_dJzCDU-8w749AnujmATI9vR5HMP6W__lPDGfLoOjbZQ-BnTh8QvsCvLPzwPrf9t4RE-SxpY&sciund=5711596763115394914 ID Sinta Dosen Pembimbing: MOCHAMAD IQBAL ARDIMANSYAH: 6658552 Yulia Retnowati: 6852573
Uncontrolled Keywords:	Random Forest, Local Binary Pattern, Hue Saturation Value, Kesegaran Sayur, Kesegaran Buah, Klasifikasi Citra, Random Forest, Local Binary Pattern, Hue Saturation Value, Fruit Freshness, Vegetable Freshness, Image Classification.
Subjects:	L Education > L Education (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software S Agriculture > S Agriculture (General) T Technology > T Technology (General)
Divisions:	UPI Kampus cibiru > S1 Rekayasa Perangkaat Lunak
Depositing User:	Genta Alima Persada
Date Deposited:	19 Aug 2025 07:10
Last Modified:	19 Aug 2025 07:10
URI:	http://repository.upi.edu/id/eprint/135293

Actions (login required)

View Item