Oktavianita, Sherli (2026) Rancang Bangun Spektrometer Near-Infrared (NIR) Portabel untuk Mengevaluasi Kesegaran Buah Apel dengan Terintegrasi Sensor Karbon Dioksida (CO2). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 2042221101_Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
2042221101_Undergraduate_Thesis.pdf Restricted to Repository staff only Download (5MB) | Request a copy |
Abstract
Apel (Malus domestica Borkh.) merupakan salah satu komoditas hortikultura bernilai ekonomi tinggi yang banyak dikonsumsi secara global, karena memiliki berbagai kandungan nutrisi seperti karbohidrat, pektin, vitamin, dan mineral. Sebagai buah klimakterik, apel tetap mengalami peningkatan laju respirasi setelah panen, yang berdampak pada penurunan tingkat kesegaran. Saat ini, evaluasi kesegaran buah apel umumnya masih dilakukan secara konvensional, seperti pengamatan visual dan titrasi kimia, namun metode tersebut bersifat destruktif. Penelitian ini bertujuan untuk merancang dan mengembangkan sistem evaluasi kesegaran buah apel secara non-destruktif menggunakan spektrometer near-infrared (NIR) portabel dengan prinsip kerja spektroskopi interactance pada rentang panjang gelombang 730-940 nm. Data spektral yang digunakan dalam penelitian ini diadopsi dari hasil akuisisi penelitian sebelumnya, yang diproses menggunakan pendekatan chemometrics melalui metode partial least squares (PLS). Selanjutnya, data spektral diintegrasikan dengan data laju respirasi karbon dioksida (CO₂) hasil respirasi buah apel yang diperoleh melalui pembacaan sensor SCD30. Data gabungan ini digunakan sebagai input pada model prediktif berbasis artificial neural network (ANN) dengan arsitektur multilayer perceptron (MLP). Hasil pengujian menunjukkan bahwa model ANN dengan arsitektur tiga hidden layer (64-32-16) mampu mengklasifikasikan tingkat kesegaran buah apel dengan akurasi sebesar 98,33%, precission sebesar 98,41%, recall sebesar 98,33%, dan F1-score sebesar 98,33%.
====================================================================================================================================
commodities and is widely consumed worldwide due to its rich nutritional content, including carbohydrates, pectin, vitamins, and minerals. As a climacteric fruit, apples continue to undergo respiration after harvest, resulting in a decline in freshness during storage. Currently, apple freshness evaluation is generally performed using conventional methods, such as visual inspection and chemical titration; however, these methods are destructive. This study aims to design and develop a non-destructive apple freshness evaluation system using a portable near-infrared (NIR) spectrometer based on the interactance spectroscopy principle within the wavelength range of 730–940 nm. The spectral data used in this study were adopted from a previous study and processed using a chemometric approach based on the Partial Least Squares (PLS) method. Subsequently, the spectral data were integrated with carbon dioxide (CO₂) respiration rate data obtained from an SCD30 sensor. The combined dataset was used as the input for an Artificial Neural Network (ANN)-based predictive model employing a Multilayer Perceptron (MLP) architecture. The experimental results demonstrated that the ANN model with three hidden layers (64–32–16 neurons) achieved an accuracy of 98.33%, a precision of 98.41%, a recall of 98.33%, and an F1-score of 98.33% in classifying the freshness level of apples.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Kesegaran apel, spektroskopi near-infrared (NIR), laju respirasi CO2, artificial neural network (ANN), Malus domestica, apple freshness, near-infrared (NIR) spectroscopy, CO₂ respiration rate, artificial neural network (ANN). |
| Subjects: | Q Science > QC Physics > QC100.5 Measuring instruments (General) S Agriculture > S Agriculture (General) T Technology > T Technology (General) > T57.5 Data Processing |
| Divisions: | Faculty of Vocational > Instrumentation Engineering |
| Depositing User: | Sherli Oktavianita |
| Date Deposited: | 17 Sep 2026 06:15 |
| Last Modified: | 17 Sep 2026 06:15 |
| URI: | http://repository.its.ac.id/id/eprint/144512 |
Actions (login required)
 |
View Item |