Font Size: a A A

Design And Implementation Of Marker Recognition System Of Seamless Stell Tubes

Posted on:2009-09-21Degree:MasterType:Thesis
Country:ChinaCandidate:J G WangFull Text:PDF
GTID:2178360278466082Subject:Software engineering
Abstract/Summary:PDF Full Text Request
In the seamless steel tubes production line of metallurgical industry,each tube for producing would have an one and only serial number. The number is spraied on the ektexine of the tube,being called marking generally. Marking machine system is an indispensably process in the seamless steel tubes production line.The information of the steel tube will be recorded in database of the manufacturer after marker is been sprayed.It will be tracked and monitored in the subsequent production process. Operatora count the regular and irregular tubes according to the markers, and record the information of each tube. Customer can find out the specific information of each tube by the marker while using them. If quality problems are appear in project usage,management can ascend to each original cylindrical bloom by the marker,even to the heat of continuous casting. Strict quality standards are request of the tubes,which used in oil exploration and production and used in making boilers and for oil and gas. So a perfect retrospective system must be built. Tube marker is crucial for retrospective process. The tube marker include pivotal information such as raw materials chemical composition,casting heat,emblem mark of manufacture,date of production and so on. After the markers are spraied by the marking machine,the tubes will through the screwing hoop station and arrived airing and packing station before leaving factory. In the subsequent flows after marking,with the detection results,the markers will be recorded in the database of the manufacturer for future reference. Any mistake should not be made about the markers of the seamless stell tubes production flow in the metallurgical industry which meeting API standard. The maker must be clearly and identifiable. In the subsequent flows after spraying,now recognizing the makers is clear or not depend on the operators' watching by naked eyes.If the maker is not clear,the operator will adjust the marking machine and respray the tube. Development of intelligent marker recognition system is placed onto a board or committee in the technological transformation of industrial enterprise of some factories for decreasing the intensity of labor and increasing automation.Through analysis the process of seamless stell tubes and a definitive or characteristic feature of tube marker,this article put forward further the design scheme of intelligent marker recognition system based on video image pattern recognition and BP-artificial neural nets(BP-ANN). Video image pattern recognition use CCD video camera to gain video image including tube marker. It pick-up the marker characters through feature analysis of the pictures and intelligently identify the characters by artificial neural net and save the identifiability result to the products database. This scheme can implement mark detection and recognition automation.It will greatly raise productivity and reduce resources of labour power.This article describes following parts:1. Analysis the production engineering and production environment of seamless steel tubes.Focal point is the ambient light and the station moving Analysis.2. Analysis the characteristic feature of tube marker,and determine processing methods to deal with various situation of the object.3. Comfirm the architecture and composition of the marker recognition system.4. Analysis the characteristic of the target images and design the image preprocessing algorithm.5. Analysis the feature of the characters in the images and comfirm the characters localization algorithm model.Implement gross orientation of the characters.6. Comfirm the segmentation algorithm model of characters,Implement the partitioning every single character from picture. 7. Implement characters recognition by Artificial Neural Networks, and increase the correctness of dynamic characters recognition by topology analysis and probability statistics.8. Complete function module design of the marker recognition system.9. Complete module program design and software test of the marker recognition system.The Intelligent Steel Tube Marker Recognition System of Seamless Steel Tubes is designed for metallurgical industry users in requirement today. With the development of the recognition algorithm and the improvement of the system,this marker recognition system would be used more widely in other industries where online recognition of characters is needed.
Keywords/Search Tags:marker recognition, image preprocessing, edge detection, image segmentation, BP-artificial neural nets
PDF Full Text Request
Related items