作者: Veerendra Singh , R. Mishra
DOI: 10.1016/J.SURFCOAT.2006.05.031
关键词: Digital camera 、 Image processing 、 Materials science 、 Principal component analysis 、 Artificial intelligence 、 Machine vision system 、 Pattern recognition 、 Entropy (information theory) 、 Artificial neural network
摘要: Abstract These studies are carried out to classify the three different spangle patterns found on galvanized steel sheets by image processing and artificial neural network. Images of 200 × 200 pixel sizes from samples were captured using optical filter digital camera. images preprocessed Haralicks (energy, entropy, contrast homogeneity) Laws (LE/EL, LS/SL, LR/RL, ES/SE SR/RS) textural parameters calculated. Principle component analysis was generated database this used train test The network could be able pattern up a reliable extent overall accuracy 80.09% for investigated samples. proposed methodology can quantification develop an online system classification. Matlab® 7 studies.