International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 74 - Number 21 |
Year of Publication: 2013 |
Authors: Shailendra M. Mukane, Feiroz F. Shaikh |
10.5120/13044-0122 |
Shailendra M. Mukane, Feiroz F. Shaikh . Identification of Metallurgical Surface Finish Images-In Manufacturing Process using Fuzzy Classifier. International Journal of Computer Applications. 74, 21 ( July 2013), 36-40. DOI=10.5120/13044-0122
Surface metrology with image processing is a challenging task having wide application in the industry. Analysis of surface finish can be done using image processing. The proposed system aims at identifying three classes of surface finish images, viz. Casting, Milling and shaping. The system proposes an effective combination of features for analysis of the engineering surfaces. The prototype system developed for classification of surface finish images using Discrete Wavelet Transform (DWT) and Fuzzy Logic classifier. The feature extracted using DWT are Standard Deviation and Mean. The system out performs the earlier methods and gives 92. 78% of average success rate for only 72 number of features. The system is also analysed by different wavelet filters for maximum success rate and minimum success rate comparison.