Comparison of Edge Detection Techniques Applied in the Identification of Centerline Segregation on Steel Slabs | Science Publications

American Journal of Applied Sciences

André Riyuiti Hirakawa¹  Luciene Coelho Lopez Queiroz¹ 

关键词: Edge Detection;    Digital Image Processing;    Centerline Segregation;    Baumann's Method;   

DOI  :  10.3844/ajassp.2015.567.571

学科分类：自然科学（综合）

来源: Science Publications

【 摘 要 】

The method Baumann, or Sulphur print as it is also known, isone of the tools used to evaluate the operating conditions of continuouscasting machine and quality control of the production of steel slabs. Thecorrect evaluation of centerline segregation severity, analyzed in Sulphurprint, is essential to control this process. However, given the classification complexity between the differentseverity levels, the classification process becomes dependent on expertsexperience and knowledge. In the light of human interference in this analysis,differences on classification results are possible, since the activity ismanual and occasionally an expert can be stricter than others during theclassification process. The evaluation of this scenario has motivated thesearch for the development of computational resources able to identify andclassify the defect of centerline segregation. The edge detection techniquesare applied to identify the centerline segregation, to reduce the amount ofdata to be processed and discard information considered irrelevant image whilepreserving the structural features of the regions of interest that will be usedlater to classify the severity of the defect. This paper presents a comparisonbetween Sobel, Laplacian of Gaussian and Canny edge detection methods appliedto the digital images of the samples generated during Baumann method foridentifying the defect of centerline segregation.

【 授权许可】

Unknown