磷含量的图片比色法测定与人工神经网络模拟  

Image Colorimetric Determination of Phosphorus Content and Artificial Neural Network Simulation

作　　者：高光芹[1] 李瑞歌[1] 谢普会[1] 黄家荣[2] 李伟[1] GAO Guangqin;LI Ruige;XIE Puhui;HUANG Jiarong;LI Wei(College of Science,Henan Agricultural University,Zhengzhou 450002,China;College of Forestry,Henan Agricultural University,Zhengzhou 450002,China)

机构地区：[1]河南农业大学理学院,郑州450002 [2]河南农业大学林学院,郑州450002

出　　处：《实验室研究与探索》2022年第3期59-62,共4页Research and Exploration In Laboratory

基　　金：河南省教育科学“十三五”规划2020年度一般课题(2020YB0047);河南农业大学教学改革研究与实践项目(202026)。

摘　　要：图片比色法、颜色识别器和人工神经网络等信息技术综合应用于有机磷含量的测量。结果表明:图片比色法与721分光光度计测定的测量的结果基本一致;红绿蓝(RGB)色度值与浓度呈非线性关系。待测磷溶液浓度,分光光度计测定结果为2.0276μg/mL,人工神经网络模型计算结果为2.0158μg/mL,其相对精度为99.42%。用图片比色法定点快速测定,再用模型进行浓度计算的方法是实用、可行。该方案具有便携、操作简单、速度快、结果精准、可实现实时、在线检测等优点。Image colorimetry, color recognizer and artificial neural network were used to measure the content of organophosphorus. The results show that the results obtained by picture colorimetry are consistent with those obtained by 721 spectrophotometer. The chromaticity value of red, green and blue(RGB) is non-linear with the concentration. To measure the concentration of phosphorus solution, spectrophotometer determination result is 2.027 6 μg/mL, the calculated result of artificial neural network model is 2.015 8 μg/mL, its relative precision is 99.42%. It is practical and feasible to use picture colorimetric point for fast determination and model for concentration calculation. The scheme has the advantages of portability, simple operation, high speed, accurate results, real-time and online detection.

关 键 词：磷含量 数字图像分析 人工神经网络 分光光度法 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]