Academic and researcher Ir Ts Dr Tham Mau Luen from Lee Kong Chian Faculty of Engineering and Science (LKC FES) successfully secured an RM10,000 research grant from the Tin Industry (Research and Development) Board Research Grant 2021. 

The grant will facilitate his research titled "Deep Learning-based Machine Vision for Tin Soldering Defect Detection in Low Powered Embedded Platform". His research explained that with the emergence of Industry 4.0 and the rapid development of artificial intelligence (AI), automating visual inspection systems using machine learning has become a popular trend. Tin soldering, which is an integral process of printed circuit board (PCB) manufacturing, can adopt AI techniques to detect defects. However, deploying these AI models on resource-limited embedded systems remains an open issue. Therefore, this project aims to develop a deep learning-based machine vision system for tin soldering defect detection in the low powered embedded platform. His research is joined by another LKC FES academic Chean Swee Ling.

“The importance of tin solder technology is visible through several forums and workshops organised by the Malaysian Tin Products Manufacturers' Association (MTPMA). Defect detection has become a crucial task during the soldering process. The early detection of tin soldering defects and the removal of the elements that may produce them are essential to improve product quality and to reduce the economic impact caused by discarding defective products. This will indirectly ensure the efficient use of tin in Malaysia. Besides that, low-power embedded system design improves reliability and sustainability due to less power waste/heat as well as reducing the operating expenditure (OPEX) of the local tin-based manufacturing industry. Overall, the project goal is related to Goal 12: Responsible Consumption and Production, which achieve the sustainable management and efficient use of natural resources. Apart from that, the research can also help UTAR to identify more industry challenges especially in the local tin industry and to provide innovative solutions that benefit society as a whole,” explained Dr Tham.

He added, “We aim to bridge the gap between AI research and practice. Existing works have focused on evaluating the feasibility of a neural network model in an ideal environment with a large number of computing resources and storage resources. How to put AI models into sustainable production remains a key challenge. Therefore, it will be interesting to deploy AI models for the applications in the tin industry.”

The Tin Industry (Research and Development) Board (Tin Board) was established under Section 4 of the Tin Industry (Research and Development) Fund Ordinance No.58 of 1953, and has been in operation since 1954. In March 1991, the Ordinance was amended and published as a revised legislation titled Tin Industry (Research and Development) Fund Act No.455 of 1953. Its objectives are to promote, stimulate and publicise the tin industry in Malaysia as well as to help increase its usage in existing or new applications of tin locally.