No abstract

BACKGROUND: This paper proposes a novel method to improve accuracy and efficiency in detecting the quality of blueberry fruit, taking advantage of deep learning in classification tasks. We first collected 'Tifblue' blueberries at seven different stages of maturity (10-70 days after full bloom) and measured the pigments of the blueberry skin and the total sugar and the total acid of the pulp. We then established a skin pigment contents prediction network (SPCPN), based on the correlation between the pigments and blueberry pictures, and also a fruit intrinsic qualities prediction network (FIQPN), based on the correlation between the pigments and fruit qualities. Finally, the SPCPN and FIQPN were consolidated into the blueberry quality parameters prediction network (BQPPN). RESULTS:The results showed that the anthocyanins in the blueberry skin were significantly correlated with the total sugar, total acid, and sugar / acid ratio of the fruit. After verification, the results also indicated that, for the prediction of anthocyanins, chlorophyll, and the anthocyanin / chlorophyll ratio, the SPCPN network model was found to achieve higher R 2 (RMSE) values of 0.969 (0.139), 0.955 (0.005), 0.967 (15.4), respectively. The FIQPN network model was also able to evaluate the value of total sugar (R 2 = 0.940, RMSE = 4.905), total acid (R 2 = 0.930, RMSE = 2.034), and the sugar / acid ratio (R2 = 0.973, RMSE = 0.580). CONCLUSION:The above results indicated the potential for utilizing deep learning technology to predict the quality indicators of blueberry before harvesting.

No abstract