A Part Classification System Using a Shape-Analysis Based Heuristic Approach

Autor: Debjit Roy, U. M. B. S. Sarma, Manoj Kumar Tiwari, S. K. Mukhopadhyay
Rok vydání: 2003
Předmět:
Zdroj: The International Journal of Advanced Manufacturing Technology. 21:355-364
ISSN: 1433-3015
0268-3768
DOI: 10.1007/s001700300041
Popis: The classification of the parts for various manufacturing applications is still a labour-intensive and time-consuming process. In this paper, attempts are made to automate some elements of this process using a shape analysis model. In this work, a 24-vector directional template is used to represent the feature elements of the parts (candidate and prototype). Various transformation processes such as deformation, straightening, bypassing, insertion and deletion are used to match the candidate parts with their prototypes. The least-cost paradigm is then adopted to ensure the efficient matching of the candidate with the prototype. Two underlying concepts, namely threshold based deterministic approaches and probabilistic approaches, are considered for the series of transformation processes to map the candidate part with different prototypes. The proposed approaches are applied to classify 100 candidate parts and to classify their prototypes to demonstrate the effectiveness of the system.
Databáze: OpenAIRE