Effects of chilled storage on fish freshness using computer vision and artificial neural network modeling

Abstract

The present study investigates the impact of refrigeration storage on the freshness and shelf life of European sea bass (Dicentrarchus labrax). This investigation utilises computer vision systems and artificial neural networks (ANNs) to analyse the dynamics of the process. A non-destructive assessment approach was established by analysing the eye colour characteristics (RGB, Lab*, and HSI values) of fish stored at +4 °C for 15 days, with sampling occurring every three days. There were considerable changes in the colour range throughout the time, particularly a reduction of brightness (L*), which can be one indicator of the progressive deterioration of the fish›s freshness. The neural network multilayer perceptron was optimised with 20 neurons in the hidden layer and demonstrated a high correlation coefficient (R² = 0.98) between predicted and experimental shelf life values. The data indicates that the values of rack life, which were initially determined to be cautious, exhibited a high degree of correlation with the estimated values. The R2 value was determined to be 0.98. The technique offers a rapid and reliable non-destructive method for determining the freshness of fish, with potential applications in relevant areas such as quality control and natural security examination for aquaculture products. The present study investigates the impact of refrigeration storage on the freshness and shelf life of European sea bass (Dicentrarchus labrax). This investigation utilises computer vision systems and artificial neural networks (ANNs) to analyse the dynamics of the process. A non-destructive assessment approach was established by analysing the eye colour characteristics (RGB, Lab*, and HSI values) of fish stored at +4 °C for 15 days, with sampling occurring every three days. There were considerable changes in the colour range throughout the time, particularly a reduction of brightness (L*), which can be one indicator of the progressive deterioration of the fish›s freshness. The neural network multilayer perceptron was optimised with 20 neurons in the hidden layer and demonstrated a high correlation coefficient (R² = 0.98) between predicted and experimental shelf life values. The data indicates that the values of rack life, which were initially determined to be cautious, exhibited a high degree of correlation with the estimated values. The R2 value was determined to be 0.98. The technique offers a rapid and reliable non-destructive method for determining the freshness of fish, with potential applications in relevant areas such as quality control and natural security examination for aquaculture products.