In the last decade, the pace of technology change has been breathtaking. From mobile devices, big data, artificial intelligence (AI), and internet of things, to robotics, blockchain, 3D printing, and machine vision, industries have been thrust into a transformative era. Strategically planning for the adoption and leveraging of some or all these technologies will be crucial in the manufacturing industry. The companies that can quickly turn their factories into intelligent automation hubs will be the ones that win long term from those investments.
But AI, specifically deep learning-based image analysis or example-based machine vision, combined with traditional rule-based machine vision can give a factory and its teams superpowers. Take a process such as the complex assembly of a modern smartphone or other consumer electronic devices. The combination of rule-based machine vision and deep learning based image analysis can help robotic assemblers identify the correct parts, help detect if a part was present or missing or assembled incorrectly on the product, and more quickly determine if those were problems. And they can do this at an unfathomable scale.
The combination of machine vision and deep learning are the on-ramp for companies to adopt smarter technologies that will give them the scale, precision, efficiency, and financial growth for the next generation. But understanding the nuanced differences between traditional machine vision and deep learning and how they complement each other, rather than replace, are essential to maximizing those investments.
At a fundamental level, machine vision systems rely on digital sensors protected inside industrial cameras with specialized optics to acquire images. Those images are then fed to a PC so specialized software can process, analyze, and measure various characteristics for decision making. Traditional machine vision systems perform reliably with consistent, well-manufactured parts. They operate via step-by-step filtering and rule-based algorithms that are more cost-effective than human inspection at scale. They can be executed at extremely fast speeds and with great accuracy. On a production line, a rule-based machine vision system can inspect hundreds, or even thousands, of parts per minute. The output of that visual data is based on a programmatic, rule-based approach to solving inspection problems.
Deep learning is a subset of artificial intelligence and a part of the broader family of machine learning. Instead of humans programming task-specific computer applications, deep learning uses data and then trains it via neural networks to make more accurate outputs based on that training data. Simply put: deep learning allows for solving specific tasks without being explicitly programmed to do so.
Rule-based machine vision and deep learning-based image analysis are a complement to each other instead of an either/or choice when adopting next generation factory automation tools. In some applications, like measurement, rule-based machine vision will still be the preferred and cost-effective choice. For complex inspections involving wide deviation and unpredictable defects—too numerous and complicated to program and maintain within a traditional machine vision system—deep learning-based tools offer an excellent alternative.
More at www.cognex.com or in the QR code