基于机器视觉和人工神经网络的切花玫瑰瓶插寿命预测

摘要

In order to predict rose vase life of cut flowerwe, using machine vision to extract color and shape feature parame-ters,a total of 15,and combined with artificial neural network ANN prediction model was established. The ANN model con-sisted of 15,15, and 1 neurons in the input,hidden,and output layers. The ANN model using the validated data showed a good predictability between the predicted and observed values(R2=0.8507, RMSEP=1.2051), similar to using the trained data (R2=0.9394, RMSEC=0.7849). The results showed that the ANN model based on machine vision can predict the vase life of cut roses accurately.%对切花玫瑰瓶插寿命进行预测,利用机器视觉提取颜色和形状特征参数共15个,并结合人工神经网络ANN建立预测模型.该预测模型的输入层、隐含层和输出层的神经元数分别为15、15和1,训练集相关系数R2=0.9394,训练集均方根误差RMSEC=0.7849;其预测集相关系数R2=0.8507,预测均方根误差RMSEP=1.2051.结果表明,机器视觉结合人工神经网络能够较准确地预测切花玫瑰的瓶插寿命.