Accurate fish-freshness prediction label based on red cabbage anthocyanins

Biosafe colorimetric labels that can accurately evaluate food freshness have been widely investigated in recent years. Here, red cabbage anthocyanin labels and back propagation (BP) neural network are combined to form a system for monitoring fish freshness. Anthocyanins extracted from red cabbage were used as color response pigments and carboxymethyl chitosan/oxidized sodium alginate (CMCS/OSA) as the solid matrix. They were dispersed in silica sol to obtain colorimetric labels using the screen-printing approach. The label is recognized by the mobile phone to obtain freshness information, rather than the traditional method with the color card. The labels underwent color gradation during the storage period which was driven by response of anthocyanins to changes in pH. Computers are more sensitive to changes in color than the human eye. The labels are divided into three categories according to the freshness of the fish. BP neural network trained with labeled red cabbage anthocyanin label images predicted fish freshness with an overall accuracy of 92.6%. Integrating a BP neural network into a smartphone application forms a simple system for fast label scanning and real-time identification of fish freshness. The system can be used for food quality control throughout the supply chain.

Automated summary

Biosafe colorimetric labels using red cabbage anthocyanin and a neural network were developed to monitor fish freshness. The labels were recognized by smartphones and classified the fish freshness based on the color gradation driven by anthocyanins' response to pH changes. The overall accuracy of the BP neural network prediction reached 92.6%.