ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Designing an Affordable System for Early Defect Detection Using Image Processing

KS, Puneeth and Sahay, A and GH, Shreya and Biswas, P and Arora, M and Chakrabarti, A (2019) Designing an Affordable System for Early Defect Detection Using Image Processing. In: 17th International Conference on Manufacturing Research, ICMR 2019, 10-12 Sept 2019, Belfast, pp. 121-128.

Full text not available from this repository. (Request a copy)
Official URL: https://doi.org/10.3233/ATDE190022


The affordability of high technology inspection systems for Micro, Small and Medium Enterprises (MSME) is a big challenge since, if not implemented, it will affect productivity and quality of the product. A case study has been carried out to understand inspection practices at an MSME in India (at an orthotic footwear company), within its medical device industry where safety and risk management are critical. Raw materials are not inspected during the initial stages of production, due to which the whole product is rejected if any defect is subsequently found. Due to this rejection, manufacturing time and man hours are squandered at the expense of quality. To enable early inspection that is also economical, existing methodology is integrated to develop a system that can boot from a Next Unit Computing (NUC) with real time processing. The inspection system uses Fault Detection and Isolation (FDI) technique using 'canny edge detection algorithm' for image processing; defects are identified as blobs by convolving the image matrix. The images are captured by low cost off-the-shelf cameras arranged in-line perpendicular to the bespoke set up. A set of light sources is used for illumination and the algorithm is evaluated for its accuracy under different light conditions. © 2019 The authors and IOS Press. All rights reserved.

Item Type: Conference Paper
Publication: Advances in Transdisciplinary Engineering
Publisher: IOS Press BV
Additional Information: The copyright for this article belongs to IOS Press BV
Keywords: Accident prevention; Defects; Edge detection; Fault detection; Industrial research; Inspection; Inspection equipment; Light sources; Risk management; Shoe manufacture, Canny edge detection; Fault detection and isolation techniques; Inspection practices; Inspection system; Manufacturing time; Medical device industry; Realtime processing; Small and medium enterprise, Image processing
Department/Centre: Division of Mechanical Sciences > Centre for Product Design & Manufacturing
Date Deposited: 01 Dec 2021 14:42
Last Modified: 01 Dec 2021 14:42
URI: http://eprints.iisc.ac.in/id/eprint/70038

Actions (login required)

View Item View Item