用数字图像技术评估番茄苗期钙素水平  

Evaluating Calcium Level of Seedling Tomato by Digital Image Technique

作　　者：郭小清[1] 范涛杰 GUO Xiao-qing;FAN Tao-jie(College of Information Science and Technology,Nanjing Agricultural University,Nanjing,Jiangsu 210095)

机构地区：[1]南京农业大学信息科学与技术学院,江苏南京210095

出　　处：《安徽农业科学》2018年第36期192-195,215,共5页Journal of Anhui Agricultural Sciences

基　　金：中央高校基本科研业务费项目(KYZ201547)

摘　　要：[目的]探讨数字图像技术对番茄钙素营养水平无损评估预测及诊断的可行性,明确该技术的最佳图像特征。[方法]2015年以番茄为材料进行钙素梯度(钙离子0、80、160 mg/L)试验。利用数码相机获取叶片数字图像数据,明确采集叶片钙素含量水平,确立了钙素营养参数模拟预测方程。同时建立了基于颜色特征R/G标准差和R/G阈值的叶片钙含量诊断模型。[结果]顶叶与侧2枝顶叶钙素含量实测值与预测值决定系数R2分别为0.853和0.902,均方根误差RMSE分别为0.128和0.016,基于阈值的诊断模型准确率超过80%。[结论]叶片颜色特征可以表征番茄苗期钙素营养水平,可为基于数字图像技术预测与诊断番茄苗期钙水平提供参考。[Objective]To explore the feasibility of digital image technology in predicting and diagnosing tomato calcium nutrition level nondestructively,and to identify the best image characteristics of this technology.[Method]Calcium gradient( 0,80,160 mg/L) test was conducted in tomato in 2015. The digital image data of leaves were acquired by digital camera. The level of calcium content in leaves was clearly collected,and the simulation and prediction equation of calcium nutrition parameters was established.At the same time,a diagnostic model of leaf calcium content based on R/G standard deviation and R/G threshold of color feature was established.[Result]The determination coefficient R2 of measured and predicted values of calcitonin content in top leaf and lateral 2 branch top leaf was 0.853 and 0.902,respectively,and RMSE of RMS was 0.128 and 0.016,respectively.The accuracy of threshold-based diagnostic model was over 80%.[Conclusion]It can provide reference for predicting and diagnosing calcium level of tomato seedling stage based on digital image technology.

关 键 词：钙素胁迫 营养诊断 叶片钙含量 

分 类 号：S126[农业科学—农业基础科学]