Non-destructive detection of blueberry skin pigments and intrinsic fruit qualities based on deep learning.

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 R2 (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 (R2 = 0.940, RMSE = 4.905), total acid (R2 = 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. © 2020 Society of Chemical Industry.

Nitrogen and phosphorus runoff losses were influenced by chemical fertilization but not by pesticide application in a double rice-cropping system in the subtropical hilly region of China.

As one of the important nitrogen (N) and phosphorus (P) pollution sources of waters, the paddy water N and P runoff losses are still poorly understood in the double rice cropping system under the interaction of chemical fertilizer and pesticide. In the subtropical hilly region of China, we conducted a 1.5-year continuous and high-frequency monitoring of paddy water N and P concentrations, runoff N and P losses, and grain yield in a double rice-cropping system with different chemical fertilizer and pesticide application rates. The results showed that the high-risk periods for N loss were in the first 5 days after the base fertilizer (BF) application and the first 10 days after the topdressing fertilizer application in both early and late rice seasons, while the high-risk periods for P loss were in the first 5 days after BF application in the early rice season and the first 15 days after BF application in the late rice season. The N and P runoff losses in the early rice season were greater than those in the late rice season, due to that the N and P fertilizers use efficiencies were lower, and thus paddy water N and P concentrations were higher in the early rice season. The paddy N and P concentrations and N and P runoff losses increased significantly with increased fertilizer application rates, while the pesticide application rate did not significantly affect N and P losses. Therefore, special effects (e.g., avoiding high irrigation, fertilizer deep application) should be taken during the high-risk periods of N and P losses to reduce the N and P runoff losses in the double rice cropping system, especially in the early rice season. There are also potentials to reduce fertilizer and pesticide input without reducing rice grain yield for the double rice cropping system in the subtropical hilly region of China.

[Atmospheric Nitrogen Dioxide, Nitric Acid, Nitrate Nitrogen Concentrations, and Wet and Dry Deposition Rates in a Double Rice Region in Subtropical China].

Nitrogen dioxide (NO2) and nitric acid (HNO3) are nitrogen-containing acidic gases in the atmosphere, and they are important precursors of nitrate in aerosol and rainwater. The emission intensity of atmospheric nitrogen oxides is high in the subtropical region of China, but the concentrations and deposition rates of atmospheric nitrogen dioxide, nitric acid, particulate nitrate-nitrogen (NO3--Np), and rainwater nitrate-nitrogen (NO3--Nr) in a double rice region in subtropical China are still unclear,. In this study, the atmosphere concentrations of NO2-N, HNO3-N, NO3--Np in PM10, and NO3--Nr and related meteorological parameters were simultaneously monitored in a typical double rice region within a subtropical hilly region of China, with the aim of determining the characteristics and influencing factors of NO2-N, HNO3-N, NO3--Np, and NO3--Nr concentrations and quantifying the wet and dry deposition rates. The results showed that the annual mean concentrations of NO2-N, HNO3-N, NO3--Np, and NO3--Nr were 4.2 µg·m-3, 0.7 µg·m-3, 4.0 µg·m-3, and 1.0 mg·L-1, respectively, and the deposition rates were 1.5, 3.2, 2.3, and 6.1 kg·hm-2, respectively. The NO2-N concentrations were negatively correlated with air temperatures, and the HNO3-N concentrations were negatively correlated with wind speeds. TheNO3--Np concentrations were negatively correlated with air temperatures, positively correlated with NO2-N concentrations, but not significantly correlated with HNO3-N concentrations, thus indicating that NO2-N concentrations were an important limiting factor forNO3--Np pollution in this study area. The NO3--Nr concentrations were negatively correlated with rainfall, as well as the concentrations of HNO3-N and NO3--Np. The annual total dry and wet depositions of the atmospheric NO2-N, HNO3-N, NO3--Np, and NO3--Nr were 13.0 kg·hm-2, which indicates that these compounds are important sources of nitrogen in paddy fields and may have significant impacts on paddy fields and surrounding ecosystems.