Determination of egg storage time at room temperature using a low-cost NIR spectrometer and machine learning techniques

Abstract

Currently, consumers are more concerned about freshness and quality of food. Poultry egg storage time is a freshness and quality indicator in industrial and consumer applications, even though egg marking is not always required outside the European Union. Other authors have already published works using expensive laboratory equipment in order to determine the storage time and freshness of eggs. This paper presents a novel alternative method based on low-cost devices for the rapid and non-destructive prediction of egg storage time at room temperature (23 ± 1 °C). H&N brown flock with 49-week-old hens were used as a source for the sampled eggs. Samples were scanned for a period of 22 days beginning from the time the egg was laid. The spectral acquisition was performed using a low-cost near-infrared reflectance (NIR) spectrometer which has a wavelength range between 740 nm and 1070 nm. The resulting dataset of 660 samples was randomly split according to a 10-fold cross-validation in order to be used in a contrast and optimization process of two machine learning algorithms. During the optimization, several models were tested to develop a robust calibration model. The best model used a Savitzky Golay pre-processing technique with a third derivative order and an artificial neural network with ten neurons in one hidden layer. Regressing the storage time of the eggs, tests achieved a coefficient of determination (R-squared) of 0.8319 ± 0.0377 and a root mean squared error in cross-validation test set (RMSECV) of 1.97 days. Although further work is needed, this technique shows industrial potential and consumer utility to determine an egg's freshness using a low-cost spectrometer connected to a smartphone.

Innovative and rapid analysis for rice authenticity using hand-held NIR spectrometry and chemometrics

“…MSC was found to give neat cluster trends and hence superior to the others applied in this study. This could be explained in that, MSC has the power to remove undesirable scatter effect of spectra from a data matrix before modelling [29] and this has proved highly useful in this study. Also, the scatter effect is known to influence modelling as it contains unwanted information for prediction.…”

“…The average of the triplicate scans (whole spectral data set) were preprocessed before further analysis. The activity of preprocessing the spectra data is an integral part of modelling to eliminate background information and noise from the useful properties of the scanned samples [29,30]. In this study, four spectra preprocessing techniques were applied comparatively, namely: de-trending (DT), mean centering (MC), multiplicative scatter correction (MSC), and standard normal variant (SNV) were employed as the specral models developed using unprocessed did not yield good results.These few preprocessing techniques were selected after initial background studies.…”

“…MSC is a well known technique used since 1983 for the removal of undesirable scatter effect from spectra data matrix before modelling [29,32]. It is also used to compensate the effect of non-uniform scattering induced by diverse particle size, uneven distribution and other effects in the spectral data and it works by linearizing each spectrum to an "ideal"spectrum, which corresponds to the average spectrum of the calibration set and the average spectrum is fitted through the method of least squares [33].…”

“…However, these spectra did contain very useful yet unexploded information, therefore the need to pretreat the spectra data with suitable pretreatment techniques was apparent. The selection of the most suitable pretreatment techniques can be quite cumbersome, and this was mainly performed using a trial and error approach involving several selections [29,46]. Finally, three preprocessing techniques were selected and were compared with unprocessed (raw) spectral profile.…”

Application of Deep Learning in Food: A Review

“…The second problem is how to realize miniaturization of sensing equipment. Except the large spectrometers used in the laboratory, some portable sensing devices combined with traditional machine learning algorithms and Internet of things technology have been applied in fruit quality evaluation, egg freshness prediction (Coronel‐Reyes, Ramirez‐Morales, Fernandez‐Blanco, Rivero, & Pazos, ), and achieved acceptable results (Wang et al., ). Miniaturized hardware computing platforms are also beginning to be used, such as NVIDIA Jetson TX2 (Partel, Kakarla, & Ampatzidis, ), which is expected to be utilized to realize local accelerated computation without the help of the internet and cloud servers.…”