基于视频与数字图像比色的甘薯多酚氧化酶活力检测

多酚氧化酶(polyphenol oxidase,PPO)是引起甘薯酶促褐变的主要原因,常用的PPO活力检测方法操作繁琐,建立一种简便快捷的检测方法意义重大。实验采用手机拍摄甘薯切片的褐变视频,提取视频图像RGB变化数据,建立了PPO活力预测模型。研究发现,0~30 s图像R值的变化量(ΔR)与PPO活力呈极强的正相关(相关系数0.946),G值与B值的变化量(ΔG和ΔB)与PPO活力呈中等强度正相关(相关系数分别为0.799和0.620)。采用多元线性回归模型对ΔR、ΔG、ΔB与PPO活力间的关系进行拟合,所得模型的拟合确定系数(R 2)达到0.903,模型的预测相关系数(r p)、预测均方误差和标准偏差比分别为0.956,2.079和3.459。结果表明,采用视频与数字图像比色快速检测甘薯多酚氧化酶活力是可行的。

Rapid determination of polyphenol oxidase activity in sweet potato based on video and digital image colorimetry

Polyphenol oxidase(PPO)is the main cause of enzymatic browning of sweet potato.The commonly used method for detecting PPO activity is complicated,so it is of great significance to establish a simple and fast method for the detection of PPO activity.In the present study,the browning process of sweet potato slices were videotaped by a cell phone,and the changes of RGB values in these videos were extracted.The PPO activity prediction model was also be established.The results showed that the change of R value(ΔR)from 0 to 30 s was highly positive correlated with PPO activity(correlation coefficient was 0.946),and the change of G and B value(ΔG andΔB)was moderately positive correlated with PPO activity(correlation coefficient was 0.799 and 0.620,respectively).Multivariate linear regression model was used to fit the relationship betweenΔR,ΔG,ΔB and PPO activity.The coefficient of determination(R 2)of the model reached 0.903,the predicted correlation coefficients(r p),root mean square error of prediction(RMSEP)and standard deviation ratio(SDR)of the model were 0.956,2.079 and 3.459,respectively.The results indicated that the methodology developed here is feasible to predict PPO activity of sweet potato by video and digital image colorimetry.