Design Of Remote Monitoring System For Micro Reagent Water Quality In Situ Monitor Based On Machine Vision

Chromium(?)is one of the core evaluation indicators for the pollution degree of heavy metals in the surface water environment.Concentration monitoring of chromium(?)in water is of great significance.At present,the measurement methods of chromium(?)include spectrophotometry,spectrometry,and electrochemical analysis.In recent years,with the development of emerging science and technology and widespread application in the industrial field,water quality monitoring technology has also developed in an integrated and intelligent direction.Water quality monitoring method based on machine vision is an important research direction of water quality monitoring technology.According to the requirements of chromium(?)water quality detection,a remote monitoring system for machine vision micro-reagent in-situ water quality monitor was designed based on machine vision technology,embedded technology,sequential injection technology and remote communication technology.This paper mainly includes the following parts.First of all,the principle of machine vision inspection is explained.Combined with the detection principle of chromium(?)spectrophotometry,a water quality detection method based on machine vision is proposed,and its detection principle is elaborated.The image processing technology in machine vision is analyzed,and the principles of color model and commonly used filtering algorithms are studied.The image acquisition experiment is designed based on the detection principle of machine vision.According to the RGB color model in the color space and the image processing principle in machine vision,the acquired images are pre-processed using Gaussian filtering algorithm and morphological operation algorithm.A multiple linear regression algorithm is used to establish a mathematical model and verify the detection principle.Secondly,the hardware platform design of the system.It mainly includes the design of the detection pool structure,the design of the machine vision detection system,the improved design based on the sequential injection platform and the hardware design of the embedded system.Through the design of the image acquisition and detection pool,the imaging quality is improved and the visual detection effect is improved.By improving the sequential injection platform,the inspection process is simplified,the inspection efficiency is improved,and the key components of the platform are analyzed and selected.The software part completes the design of the STM32 MCU-based sequential injection platform control software which ?C/OS-? platform is applied for multi-task management,and realizes the GPRS remote transmission and detection process control.Developed industrial computer software and remote monitoring platform software using Python,Open CV and Py Qt5.Finally,referring to the detection method in GB7464-198 and the experimental platform of this system,the experimental procedure was designed.Using an improved sequential injection platform and machine vision system,a chromium(?)detection experiment was performed.After the image data is obtained,it is processed based on the actual situation of the image,and a shadow compensation algorithm is designed to improve the quality of the image.Using Python's machine learning library to perform multiple linear regression analysis on the image data,a regression model based on the solution color characteristic value and the chromium(?)concentration was established.Experiments were designed to test the regression model,and compared with the results of spectrophotometry,the feasibility of the principle of machine vision detection was verified.