Mfanasibili Nkonyane, branch general manager (left), thanks Werner Das of SICK Automation.
Artificial intelligence is changing the way industrial vision systems approach quality control, and the Durban branch September 2026 technology meeting looked at how deep learning is moving from the realm of research into practical, everyday manufacturing applications.
Presenting ‘Machine Vision Beyond Rules: Leveraging Deep Learning for Everyday Industrial Quality Challenges’, Werner Das of SICK Automation took attendees through the evolution of industrial vision and the growing role of AI in solving problems that can be difficult to address with traditional rule-based systems.
Conventional machine vision techniques, such as thresholding, edge detection and blob analysis, remain highly effective for applications requiring precision, including dimensional verification and barcode reading. However, they can become increasingly challenging to apply when faced with natural variations in materials, complex assemblies or unpredictable defects.
Deep learning offers a different approach. Rather than requiring engineers to develop rules for every possible variation, systems can be trained using examples, allowing them to recognise patterns and make decisions in applications such as complex assembly verification, multi-class part classification and deformable packaging inspection.
Das also explored what could be next for industrial vision, highlighting the potential of hybrid systems that combine the precision of traditional vision rules with the adaptability of neural networks.
With an electronics engineering degree from the University of Pretoria and extensive experience in automation, vision systems, PLCs and industrial digitalisation across Africa, Das brought a strong practical perspective to the discussion. The presentation provided insight into how AI is becoming a practical addition to the industrial engineer’s toolbox, helping manufacturers tackle quality challenges that were previously difficult to automate.
SAIMC: From the office of the CEO SAIMC
SAIMC
The INSPIRE Conference 2026 brought industry and academia together to explore advances in automation, AI, mechatronics and engineering. It also marked a new direction for the SAIMC as it works to build future-ready engineering talent and strengthen South Africa’s automation industry.
Read more...SAIMC Durban takes to the skies SAIMC
SAIMC
SAIMC Durban members went behind the scenes at the ATNS facility at King Shaka International Airport. The visit offered a fascinating look at the sophisticated technology and highly trained people keeping South Africa’s skies safe.
Read more...SAIMC: From the office of the COO SAIMC
SAIMC
The recognition of mechatronics and computer engineering has created both an opportunity and a responsibility. Recognition means little without participation, professional registration without professional conduct, and technical knowledge without the mentorship, implementation and measurable value that industry needs.
Read more...SAIMC: Johannesburg branch SAIMC
SAIMC
Hendrik Spies of SICK Automation Southern Africa presented at the SAIMC Johannesburg Technology Evening on 12 August 2026, covering the digital journey from condition monitoring to predictive maintenance.
Read more...SAIMC: Durban branch SAIMC
SAIMC
The SAIMC Durban branch hosted S’yabonga Langa, systems performance engineer at eThekwini Electricity, who presented on applying substation automation to improve operational efficiency and service quality in modern power distribution networks.
Read more...SAIMC INSPIRE 2026 brings together industry, robotics and research
SAIMC
The inaugural SAIMC INSPIRE Conference will bring together the industrial automation, robotics and academic communities for two days of presentations, panel discussions and networking, and will explore developments in smart manufacturing, mining, robotics
While every effort has been made to ensure the accuracy of the information contained herein, the publisher and its agents cannot be held responsible for any errors contained, or any loss incurred as a result. Articles published do not necessarily reflect the views of the publishers. The editor reserves the right to alter or cut copy. Articles submitted are deemed to have been cleared for publication. Advertisements and company contact details are published as provided by the advertiser. Technews Publishing (Pty) Ltd cannot be held responsible for the accuracy or veracity of supplied material.
printer friendly version