Study On Secondary Identification Method Of Paper Disease In FPGA+ Computer Mode

With the rapid development of paper-making industry and the continuous improvement of automation in paper production,the current paper machine web has become wider and its speed has been continuously increasing.In addition,uncertainties such as paper fibers,process flow,and mechanical equipment make the paper-making process face more risks(common appearance of paper diseases),and people have higher and higher requirements for the quality of paper products.Traditional detection technology relying on manual detection of paper defects can no long meet the actual needs.Therefore,it is very important to study an automatic detection device based on machine vision to become the preferred solution,which is especially important for improving detection efficiency and improving the accuracy of paper disease identification.In this paper,the technical scheme of secondary identification of paper diseases in FPGA +computer mode is proposed by analyzes the causes and characteristics of common paper diseases in paper-making industry and study the advantages and disadvantages of the existing on-line detection scheme for paper diseases.On the basis of this,we improve the hardware platform,determine the classification algorithm,and develop the on-line detection software of paper disease.The main research work and contents are as follows:(1)Determination of online detection technology for paper diseases.At present,the “industrial camera +computer” structure model is generally used for on-line detection of paper diseases.Analyzing the working process and the method of identification,since the computer as the core of the detection system deals with image data in a serial way,which makes it increasingly difficult to deal with huge amount of data generated by high speed and wide paper machine in real time.As this kind of pattern detection system increasingly restricted bottleneck problems of the quickness and accuracy,a detection system structure mode of “industrial camera + FPGA + computer” is put forward to realize the technical scheme of “secondary identification” of paper disease.FPGA is responsible for the identification of the paper disease,and sent the suspected paper area to the computer to complete the accurate secondary identification of the suspected paper disease area.(2)Paper disease analysis and algorithm research.This paper analyzes the pathogenic mechanism and its characteristics of common appearance paper disease,and uses digital image processing technology to process the collected images,mainly including image preprocessing and suspected paper disease region segmentation of FPGA implementation;the upper computer terminal realizes the feature extraction of image and classification of paper disease.In the algorithm research,it's mainly through the simulation statistics and analysis to verify the processing effect of each algorithm to the paper disease,and select the independent and complete characteristic quantity to improve the identification precision.(3)Realization of on-line detection software system for paper disease.According to the plan demand,the software system functional requirements are determined and the system is divided into two parts:(1)the FPGA side design,the FPGA side is divided into six functional modules design and implementation,mainly to complete the image acquisition module,image preprocessing module,edge detection module,suspected of pathogenic region segmentation module,SDRAM cache module,Ethernet transmission module,etc.;(2)in the software development platform of VS2010,it adopts C++ language to program the functions of each module in the MFC framework,and adopts multithreading technology to meet the real-time requirements of system operation.Software according to the module function design of image acquisition module,data processing module,database interface module and display module,etc.,thus it will be well realize the real-time detection of the paper surface defect,real-time identification and statistics of paper disease,paper syndrome and log information storage and query functions.Finally,through debugging the whole paper disease detection system,gradually improve the hardware structure,optimize the light source and improve functions of the software system during the debugging process.After tests showed that the hardware structure and light source optimization used by this topic have certain guidance and reference for visual inspection of similar product,and the paper disease detection software system can fast and stable operation to recognize and classify paper disease,which well meet the demand of detection.At the same time,the secondary identification method of paper disease is proposed to construct a distributed paper disease detection hardware system,which have good portability and expansibility and also can maximize the decomposition of paper disease detection of quickness and accuracy problems.