Title :
Classification of carbide distributions using scale selection and directional distributions
Author :
Wiltschi, Klaus ; Lindeberg, Tony ; Pinz, Axel
Author_Institution :
Inst. for Comput. Graphics & Vision, Tech. Univ. Graz, Austria
Abstract :
We present an automatic method for the classification of steel quality based on scale-space operations. The carbide distribution of microscopic specimen images is assessed by classifying according to so-called `degree´ and `type´ of the specimen. `Degree´ is represented by features extracted with automatic scale selection, and `type´ information is computed from second-moment descriptors. In combination with a morphological verification scheme, this pattern classifier shares large similarities with current manual techniques. Compared to previous work, the new classification scheme has several advantages: the significant scale of the carbide agglomeration is calculated explicitly, and the method is less sensitive to the variance of spatial connectivity than a morphological approach
Keywords :
automatic optical inspection; carbon compounds; feature extraction; image classification; impurity distribution; object detection; production engineering computing; steel; automatic method; automatic scale selection; carbide agglomeration; carbide distributions; classification; degree; directional distributions; feature extraction; microscopic specimen images; morphological verification scheme; pattern classifier; scale selection; scale-space operations; second-moment descriptors; spatial connectivity; steel quality; type information; Computer graphics; Computer vision; Data mining; Detectors; Feature extraction; Kernel; Laboratories; Microscopy; Shape; Steel;
Conference_Titel :
Image Processing, 1997. Proceedings., International Conference on
Conference_Location :
Santa Barbara, CA
Print_ISBN :
0-8186-8183-7
DOI :
10.1109/ICIP.1997.632012