Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/105758
Title: Deteksi Cepat Kesegaran Telur Ayam Negeri Menggunakan Portable Near-Infrared Spectrophotometer (NIRS)
Other Titles: Rapid Prediction of Egg Freshness using Portable Near-Infrared Spectrophotometer (NIRS)
Authors: Purwanto, Y. Aris
Widodo, Slamet
Ardiamsyah, Vondra
Issue Date: 2021
Publisher: IPB University
Abstract: Telur merupakan bahan pangan yang sangat mudah mengalami penurunan kualitas kesegaran. Hal tersebut dapat terjadi akibat beberapa faktor seperti lama penyimpanan, suhu penyimpanan yang tinggi dan kelembaban relatif ruang penyimpanan yang rendah. Faktor-faktor ini yang menyebabkan hilangnya kadar CO2 pada telur sehingga menyebabkan kerusakan jaringan-jaring dalam telur dan naiknya pH pada albumen telur. Kesegaran telur sangat sulit dilihat dengan pengamatan biasa, sehingga dibutuhkan alat yang dapat mendeteksi kesegaran telur secara non-destruktif. Portable Near-Infrared Spectrophotometer (NIRS) SCiO merupakan alat yang dapat membantu dalam memprediksi kandungan bahan pangan. SCiO bekerja pada rentang gelombang 740nm hingga 1070nm, dimana pada rentang gelombang tersebut sesuai dengan overtones kelompok gugus C-H, N-H dan O-H. Prediksi kesegaran telur dilakukan dengan menggunakan metodePartial Least Square Discriminant Analysis (PLS-DA) dan dilanjutkan denganpenerapan tabel confusion matrix guna mencari luaran berupa akurasi, sensitivitas,spesifisitas dan false positive rate (FPR). Hasil penelitian menunjukkan bahwamodel terbaik untuk memprediksi kesegaran telur adalah dengan pre-treatmentderivative 2nd. Berdasarkan hasil kalibrasi diperoleh nilai rataan akurasi 93.12%,sensitivitas 91.32%, spesifisitas 94.47%, dan FPR 5.53%. Sedangkan untukvalidasi diperoleh nilai rataan akurasi 91.22%, sensitivitas 88.02%, spesifisitas93.63%, dan FPR 6.37%
Eggs are a food ingredient that is very prone to degradation of freshness. That can occur due to several factors such as storage time, high temperature, and low humidity. These factors cause the loss of CO2 levels in the egg, causing damage to the egg's tissue and making the pH rise in the albumen. The freshness of eggs is very difficult to see with ordinary observation, so we need a tool to detect egg freshness with a non-destructive technique. Portable Near-Infrared Spectrophotometer (NIRS) SCiO is a tool that can help us to predict food content. The SCiO portable NIRS works in the wavelength 740 to 1070nm, where the wavelength corresponds to the overtones of the C-H, N-H, and O-H group. Prediction of egg freshness was carried out using Partial Least Square Discriminant Analysis (PLS-DA) method. It continued with the confusion matrix table to find the output of accuracy, sensitivity, specificity, and false-positive rate (FPR). The result showed that the best model to predict the egg freshness is using derivative 2nd pre-treatment. The average value of accuracy is 93.12% on the calibration data, sensitivity is 91.32%, specificity is 94.47%, and the FPR is 5.53%. On the validation data, the average value of accuracy is 91.22%, sensitivity is 88.02%, specificity is 94.47%, and the FPR is 6.37%
URI: http://repository.ipb.ac.id/handle/123456789/105758
Appears in Collections:UT - Agricultural and Biosystem Engineering

Files in This Item:
File Description SizeFormat 
Cover.pdf
  Restricted Access
Cover429.37 kBAdobe PDFView/Open
F14160049_Vondra Ardiamsyah.pdf
  Restricted Access
Fullteks976.91 kBAdobe PDFView/Open
Lampiran.pdf
  Restricted Access
Lampiran1.26 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.