In the fast-paced world of manufacturing, ensuring product quality whilst maintaining efficiency is a significant challenge. Traditional human-led inspection processes are often prone to errors and inefficiencies. However, ADLINK’s Automated Optical Inspection (AOI) solution is changing the game. Powered by Intel CPU, GPUs, and AI models optimised by the OpenVINO toolkit. This solution offers a highly accurate and scalable approach to visual inspection. Therefore, significantly reducing the overhead costs associated with quality assurance. So much so that ADLINK took the opportunity to install a system on their Embedded Computing line.
The Need for Automated Inspection:
Human inspectors, despite their best efforts, struggle with the high-speed, repetitive nature of product inspection on manufacturing lines. This often leads to decreased accuracy and reliability. The integration of computer vision and AI in inspection processes has been a game-changer. Thereby, reducing fatigue and boosting accuracy and productivity. With the advent of AI and neural networks, automated optical inspection systems are now more efficient than ever. Therefore, reducing the burden on human inspectors.
How ADLINK’s AOI Solution Works:
ADLINK’s AOI system uses camera sensors to capture images of products from multiple angles. These images are then algorithmically analysed to detect defects and abnormalities. This technology is versatile and can be applied across various manufacturing industries. Enhancing quality control and reducing costs. The global market for such technology is expected to grow significantly in the coming years.
The Technology Behind the Solution:
ADLINK’s AOI solution is implemented using the MVP-5200 Series compact industrial computer. Firstly, it features an Intel Core i7-12700E processor and an ADLINK MXM-AXe GPU module based on Intel Arc Graphics. This setup provides a flexible mix of CPU, integrated GPU, and discrete GPU resources. Making it suitable for handling neural network models of varying complexities. The solution’s accuracy is its most crucial aspect. Initial tests showing a remarkable inspection rate of one product per second with 98.8% accuracy.
Streamlined Integration and Development:
To facilitate the development and deployment of machine vision applications, ADLINK incorporates the Edge Vision Analytics (EVA) SDK into its AI vision products. This platform offers a runtime and integrated development environment (IDE) that supports Windows, Linux, and Docker containers. It simplifies the development process. Achieved with open-source plugins, low-code/no-code GUI, and support for multiple camera protocols.
Leveraging OpenVINO for Enhanced Performance:
The Intel Distribution of OpenVINO toolkit, included with the EVA SDK, plays a crucial role in the AOI solution. It enables the creation of fast and accurate deep learning pipeline. Therefore, optimising edge-optimised models for Intel architecture. This results in improved accuracy and accelerated development.
Conclusion:
ADLINK’s AOI solution, with its open and extensible platform, is not just a tool for quality inspection but a stepping stone towards broader AI implementations in manufacturing. As AI continues to integrate with robotics and other automated systems on the factory floor, solutions like ADLINK’s AOI are essential for building efficiency, reducing costs, and paving the way for a more profitable future in manufacturing.
Click here to read the full application note on the ADLINK AI Automated Optical Inspection Solution or contact us for more information.