Application of VIS/NIR Spectroscopy and SDAE-NN Algorithm for Predicting the Cold Storage Time of Salmon

Abstract

The cold storage time of salmon has a significant impact on its freshness, which is an important factor for consumers to evaluate the quality of salmon. The efficient, accurate, and convenient protocol is urgent to appraise the freshness for quality checking. In this paper, the ability of visible/near-infrared (VIS/NIR) spectroscopy was evaluated to predict the cold storage time of salmon meat and skin, which were stored at low-temperature box for 0~12 days. Meanwhile, a double-layer stacked denoising autoencoder neural network (SDAE-NN) algorithm was introduced to establish the prediction model without spectral pre-preprocessing. The results showed that, compared with the common methods such as partial least squares regression (PLSR) and back propagation neural network (BP-NN), the SDAE-NN method had a better performance due to its high efficiency in decreasing noise and optimizing the initial weights. The determination coefficient of test sets (R 2 test ) and root mean square error of test sets (RMSEP) have been calculated based on SDAE-NN, for the salmon meat (skin), the R 2 test can reach 0.98 (0.92), and the RMSEP can reach 0.93 (1.75), respectively. It is highlighted that the algorithm is efficient and accurate and that the salmon meat would be more suitable for predicting freshness than the salmon skin. VIS/NIR spectroscopy combined with the SDAE-NN algorithm can be widely used to predict the freshness of various agricultural products.

Emerging Approach for Fish Freshness Evaluation: Principle, Application and Challenges

“…In addition, researchers have used near-infrared spectroscopy in the 800-2500 nm region to predict the potential of bacterial populations in salmon stored at 4 • C. The PLSR model was built on 72 data points and predicted the total number of aerobic slabs for the calibration and validation datasets with R 2 values of 0.95 and 0.64, respectively [102]. Similarly, Wu et al [103] investigated the ability to use VIS/NIR to predict salmon cold storage time. At the same time, a double-layer stacking denoising self-encoding neural network (SDAE-NN) algorithm is introduced to establish a prediction model without spectral preprocessing.…”

Sensing Technology for Fish Freshness and Safety: A Review

“…In the last couple of years, deep learning techniques have been used more and more in this field: for example, Wu et al [ 96 ] coupled FTIR spectroscopy (in the range of 400–1000 nm) with a stacked denoising autoencoder neural network (SDAE-NN) to predict the cold storage time of salmon, obtaining an R 2 of 0.98 in prediction and a root mean square error in prediction (RMSEP) of 0.93 days whereas Agyekum et al [ 97 ] studied the potentiality of a genetic algorithm to quantify volatile TMA concentrations in silver carps, using spectra acquired with an FT-NIR spectrometer and an optical fiber, resulting in an R 2 p of 0.980 (RMSEP = 5.1 mgN/100 g). A similar paper was published by the same research group in 2020, where FT-NIR was coupled with an ant colony PLS (ACO-PLS) algorithm to evaluate the K-value in silver carps, obtaining an R 2 p of 0.98 (RMSEP = 3.98).…”

Bayesian Fusion Model Enhanced Codfish Classification Using Near Infrared and Raman Spectrum

“…The determination coefficient of test sets (R 2 test) and root mean square error of test sets (RMSEP) have been calculated based on SDAE-NN; for the salmon meat (skin), the R 2 test can reach 0.98 (0.92), and the RMSEP can reach 0.93 (1.75), respectively. [ 15 ]. In addition, Raman spectroscopy was applied using a 532 nm laser for the classification of 12 frozen types of frozen fish fillets.…”