3741

3535

كارشناسي ارشد

معماري كامپيوتر

اصفهان :دانشگاه صنعتي اصفهان ،دانشكده برق و كامپيوتر

1384

نه ، 98 ،[I]ص. : مصور،جدول ،نمودار

ص.ع. به فارسي

رسول امير فتاحي، شادرخ سماوي

محمدرضا احمد زاده

04/09/86

عليمحمد دوست حسيني، سعيد صدري

مهندسي برق و كامپيوتر

ID3535

به فارسي و انگليسي :قابل رؤيت در نسخه ديجيتال

AbstractSurface inspection of steel sheets is Vital for Steel Industries On the other hand visualinspection manually not only is a dangerous task but also has low performance Toimprove quality steel sheets must be classified into defected and defect free categories Then each defect category can be divided in more subclasses of defects There are anumber of different automatic methods for this purpose and we used neural networks inour work We considered four surface defects including Holes Scratches Coil breaks andRusting Images of defect free sheets were also included and using neural networks andsimple characteristics to categories the images Parallelism and high adaptability towardenvironmental changes are among advantages of using neural networks as compared toother identification methods If a new class of defects is added to the existing classes wecan train network again to classify the whole set and therefore there is no need for a newalgorithm to classify the added category Simulation results show that our approach hashigh capability in detection and recognition of surface defects In a manner that the systemis always successful in classifying faulted and fault free categories In addition to highperformance the suggested approach due to high parallelism is suitable for real timeimplementation

رسول امير فتاحي، شادرخ سماوي

محمدرضا احمد زاده

عليمحمد دوست حسيني، سعيد صدري