@misc{Khedkar_Sandip_S._Enhanced,
 author={Khedkar, Sandip S. and Damahe, Lalit and Chaudhari, Sharad S. and Fulzele, Punit R. and Algburi, Sameer and Mohammed, Salah J. and Majdi, Hasan S.A.},
 howpublished={online},
 publisher={Zielona Góra: Uniwersytet Zielonogórski},
 language={eng},
 abstract={An automated technique for detecting and classifying defects in valve-casting radiography pictures is presented in this research. The suggested system offers solutions for challenging radiography image interpretation issues. The over-segmentation issue is resolved by the multiple morphological image processing used in this work, which also provides good results for identifying the size and shape of the faults. The suggested system is a collection of methods to implement an automated system for inspecting valve-casting radiography images.},
 abstract={Film digitization, image pre-processing aimed primarily at noise reduction and elimination, contrast enhancement and discriminate feature enhancement facing interpretation, multi-level morphological image processing, and defect region segmentation are steps in automated defect detection and classification. After the segmented region`s features are recovered, principal component analysis (PCA) is carried out. On PCA data, classification methods like artificial neural networks (ANN) are used. A casting manufacturing company evaluated the new economical working style for a variety of valve castings.},
 title={Enhanced Defect Detection in Valve-Casting Radiography Using Morphological Image Processing and Machine Learning},
 type={artykuł},
 keywords={casting defects, digitalization, morphological image processing, ANN},
}