| The technology of computer image processing involves the basic theories and methods of all kinds of image information processing by means of the computer. Presently.This technology has been widely applied to many fields,such as medical diagnoses,industrial automatic measuring and control,intelligent robot,transportation, communication and geology.With drastic development of home appliances,automobile and electronic industry etc,not only the requirement of quantity but also the requirement of quality of steel strip product are more and more in market.Shape quality is a key index of cold rolled strip and an important factor for its competition in market,therefore,the technology of shape recognition plays a very important role in the process of shape rolling.In terms of the practical application of the technology of shape recognition at currently iron and steel factories,the shape is obtained indirectly by measuring the inner-stress of plate during the process of plate rolling,because of the main shortage of its touch-measuring mode and the factor of outer environment and the sensor,this method can't ensure the reliability of shape measuring.How to recognise shape quality of cold rolled strip effectively becomes an urgent problem to solve, so it is very necessary to do some research on it.This paper according to technology reform requirement of the line at Cold Rolling department,ChongQing Steel Group Corporation,analyses and studies the computer image processing technology applying in the shape recognition based on the existing theories of shape and Flatness pattern recognition,in the paper,on the basis of existing technology of shape recognition and syntheses of the modern recognition technology,how up to the minute study harvest of the computer image processing is used to moden shape recognition is the primacy topic,the software system of shape recognition based on computer image processing technology has been developed successfully using production locale.The new system adopt Visual C++6.0,make use of CCD pick up camera,Image Collection card,Industry Control machine exploited a suit of shape imaging system,take advantage of current technology of computer image processing,instead of manual work recognition method,This kind of new recognition method is novel with simple equipment,improve recognition rate,which can realize untouched recognition and meet precision requirement of most product line of coldl rolling company,carry out untouch recognition in cold rolling successfully;In process of image filter smoothing the noise,aim at the model of Perona&Malik's shortage,improve the model of Perona&Malik,by experiment validate,it can filt the noise and keep image edge successfully,the effect of filter smoothing the noise is good;in image edge detection,according to the idiographic on shape defection image,based on the shortage of traditional canny algorithm,improve traditional canny algorithm,compare diversified edge detection technology,the experiment shows that the improved canny algorithm not only keeps traditional canny's excellent performance,but also improves the performance in the detail edge detection and good detection,and handle time sooner,which were more suit for defects image processing and the features extraction of sharp recognition system;Features extracted by the system are discussed,and the back-propagation neural network has been chosen to as the method of image classification,they was efficient of processing defects image of shape.The result shows that main types of shape defects,which were investigated in this paper,can be recognized efficiently,and the average recognition rate is 90%. |
PDF Full Text Request