Smoke Of Ringelmann Black Remote Video Monitoring System Applied Research

With the rapid development of image acquisition and gradual maturation of the corresponding application technology,it has been an inevitable trend to use networking computer image capture means to give automatic monitoring to black smoke pollutants comparing with conventional methods.The situation of black smoke Ringlemann degree monitoring is introduced.The work principles of three kinds of equipments currently used are specified,and the advantages and disadvantages of each of them are pointed out.After the analysis of the development trend of Ringelmann degree monitoring,the practicability and application prospect of the research are pointed out.The main contents of the study are:the capture of smoke image,the extraction of smoke region,spatio-temporal joint background deduction,identification and output of Ringlemann degree based on pattern recognition method.This paper designs to realize manual and automatic monitoring functions to smoke Ringlemann degree in common sky background.The specific function is respectively: image gathering and store at the front-end,image data compound processing,automatic identification of Ringlemann degree and pan&tilt control at the front-end,multi-channel preview,store and pan&tilt control at the monitoring center.Through comprehensive use of image collection,digital image processing and pattern recognition technology,this research fundamentally avoid subjective measurement deviation of Ringelmann degree,thus it breaks away from the laggard artificial subjective evaluating status.In addition,comparing with the high-end products,it emphasizes reducing the cost of instruments.Based on the secondary development of a high performance DSP image acquisition card,the system realizes four channels real-time data acquisition.Image compression adopts advanced h.264 format.Through a comprehensive procedure of digital image processing,it selects effective image areas,deducts the sky background influence to the selected region brightness.Furthermore,by the using of pattern recognition,it recognises Ringelmann degree and outputs the result effectively. Through the programming based on Software Development Kit of the acquisition card,it realized a network monitoring system of C/S structure.Still,it realizes the control to camera and pan&tilt based on the programming to serial communication.According to the relevant domestic standard and technical material requirements,this project has completed the principal function design of the system.The artical expounds the design idea and the key technologies,presents the experimental results of main function modules and certain simulation results of some modules.This design puts forward the idea of using neural network in Ringelmann degree monitoring.On considering of both the system efficiency and performance,it fully embodies the better real-time feature and high throughput rate.Moreover,it increases the sample data quantity properly,which is more advantageous to improve the accuracy of the monitoring records than the existing technical means.