Jamie from 42 Technology (https://www.42technology.com) showcases their application for automated line clearance at Embedded World 2025. Utilizing embedded edge AI technology, this demonstration addresses real-time quality control and anomaly detection specifically designed for pharmaceutical manufacturing processes. The system employs industrial-grade cameras and Synaptics’ Astra SL1680 edge AI processor to monitor pill production lines, accurately detecting contaminants and process irregularities.
—
Synaptics is my Embedded World 2025 video coverage sponsor, check out my Synaptics videos here: https://www.youtube.com/playlist?list=PL7xXqJFxvYvhAbQoe9YN4c84SqXxIY3fQ
—
The demonstration specifically captures images of pills moving on a conveyor belt, analyzing these visuals instantly to identify anomalies based on color differentiation. Though color is the demonstrated differentiator here, the system’s AI capabilities are flexible, capable of identifying various anomalies or contaminants that pose significant risks in pharmaceutical production and other industrial applications.
42 Technology has spearheaded the design and conceptualization of this demonstration. Leveraging their expertise in manufacturing processes, instrumentation, and systems engineering, they collaborated closely with Synaptics, using the Astra SL1680 processor at the system’s core. The Astra platform, recognized for its edge AI performance, integrates seamlessly with Bala industrial camera systems to deliver precise imaging data for analysis.
A critical partner in this collaboration is Arcturus Networks, part of Synaptics’ ecosystem. Arcturus provided essential support in developing the machine learning aspects of the project, enabling accurate anomaly detection and enhancing the overall reliability and responsiveness of the embedded AI solution.
The showcased conveyor system replicates real-world pharmaceutical manufacturing environments. Pills are automatically ejected onto a production line, where real-time image analysis is conducted. The edge AI technology quickly counts pills, identifies irregularities, and logs any contaminants, providing immediate visual feedback and timestamped records. This capability allows operators to review anomalies rapidly and take prompt corrective action.
This solution addresses a common challenge in industrial manufacturing: ensuring line clearance and eliminating safety risks associated with contaminants or leftover packaging materials. The demonstration specifically illustrates detection of anomalies such as residual pill bottles or cartons that could otherwise remain unnoticed. Timely identification of these irregularities helps maintain safety standards, regulatory compliance, and operational efficiency.
This project highlights the effectiveness of collaborative development. It combines 42 Technology’s industrial expertise and integration skills, Synaptics’ specialized edge AI hardware, Bala’s robust vision systems, and Arcturus Networks’ sophisticated machine learning algorithms. Such multidisciplinary collaboration ensures that the final system is both reliable and versatile, suitable for diverse manufacturing contexts beyond pharmaceuticals.
As showcased at Embedded World 2025, this technology represents an effective use of edge AI in enhancing industrial safety and quality assurance. Its flexibility and adaptability to various anomaly detection scenarios illustrate a practical implementation of machine learning in manufacturing, marking a notable development in embedded industrial automation.
Check out all my Embedded World 2025 videos in this playlist: https://www.youtube.com/playlist?list=PL7xXqJFxvYvjgUpdNMBkGzEWU6YVxR8Ga
This video was filmed using the DJI Pocket 3 ($669 at https://amzn.to/4aMpKIC using the dual wireless DJI Mic 2 microphones with the DJI lapel microphone https://amzn.to/3XIj3l8 ), watch all my DJI Pocket 3 videos here https://www.youtube.com/playlist?list=PL7xXqJFxvYvhDlWIAxm_pR9dp7ArSkhKK
Join https://www.youtube.com/charbax/join for Early Access to my videos and to support my work, or you can Click the “Super Thanks” button below the video to send a highlighted comment!
