Open AccessPre-Forming Inspection System to Detect Deep Drawing Defect Due to Punch-Die Misalignment using Image Processing Technique
Author(s) -
Alimi Abdul Ghafar,
Ahmad Badri Abdullah
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/530/1/012015
Subject(s) - die (integrated circuit) , sheet metal , centroid , offset (computer science) , image processing , deep drawing , computer science , toolbox , automotive industry , engineering drawing , engineering , artificial intelligence , mechanical engineering , image (mathematics) , programming language , aerospace engineering
Deep drawing processes are extensively used in the industry for the fabrication of a wide range of sheet metal products from household appliances to automotive parts. In a deep drawing process, the alignment of the punch and die is extremely important to avoid defects such as variation of thickness and intolerable thinning condition of the drawn part. In this paper, a computer vision system is proposed that is able to identify the degree of misalignment in the punch-die configuration. The apparatus consists of an image capture and analysis system that can be used as a tool for pre-forming inspection setup. The image analysis implements the centroid recognition approach using the Image Processing Toolbox function in MATLAB. The offset value between centre of the punch and the centre of die indicates the severity of misalignment between punch and die. The system was successfully tested for a square profile, with more accurate results obtained for a colored punch as compared to the punch with the original metallic surface. Verification of the system was then conducted to align the centroids of a punch and die, and the results show that a good alignment of the centroids was achieved within two alignment attempts.