When machine vision in manufacturing emerged decades ago, it seemed science fiction came to life. Despite the power of machines that could “see” and interpret, manufacturers were hesitant to embrace machine vision. A general lack of awareness plus steep investments fueled this hesitancy to leverage machine vision as a solution.

Times have changed. With decades of experience working with machine vision, I’ve witnessed interest soar in manufacturing machine vision solutions along with transformative generational changes. The question used to be “Can it be done?” Now, it’s “How will we do it?” Let’s take a look at the shift.

The Manufacturing Game Has Changed

Today, machine vision is integral to manufacturing automation and industrial imaging as top inspection and quality control solutions. Its capabilities have evolved dramatically with more innovative technology and more widely accessible costs.

For example, 30 years ago, a 1.4-megapixel machine vision camera was over $10k. Today a 5-megapixel industrial camera is less than $1k. The computing power behind machine vision has also dramatically increased over the last three decades. While early systems had only a few megabytes of RAM, today’s smartphones possess far more power than those early systems.

Improving beyond more powerful cameras and advanced algorithms, the integration of artificial intelligence (AI) gave machine vision systems a superpower jolt. AI brought the analysis of images, parts, or components to new heights. While traditional machine vision always had the advantage of accuracy and speed compared to human capability, it was limited to applications that could be analyzed through advanced mathematical algorithms. AI shifted the paradigm to include applications that historically only humans could do.

Precision is non-negotiable in manufacturing processes and AI has completely changed options for manufacturers, enabling automated inspection to enhance quality control. Take auto manufacturing for instance: AI-powered machine vision systems can inspect every component and “see” anomalies or defects that are not readily discernible to the human eye or easily missed. This process ensures every piece or part that goes into assembling a vehicle meets quality standards to help produce a safe and functional vehicle. Automated inspection processes also help manufacturers address issues before they escalate, greatly reducing waste and rework.

Machine vision will continue to evolve. Stay on top of current trends and tools, but remember, it’s also important to leverage the past several decades of experience with the latest innovations to advance.

AI Moves Machine Vision Forward

Many of today’s latest machine vision advancements are around AI, but an AI system is only as good as its training. Some AI deep learning machine vision systems are pre-trained based on a large data set and can work extremely well. For example, our latest Deep Learning OCR (Optical Character Reading) tool works in this way. Edge-based training or single-product line training can be easy and quick but may not capture AI’s full potential.

Cloud-based AI training uses cloud computing and images collected from different production lines to generate a neural network model that automatically categorizes images that can be used at scale across manufacturers’ plants. It puts every user and site across an operation on the same page in terms of model performance, having collaborated on its training, a huge benefit. The best system deployments can use a combination of all three training methods – pre-trained, edge, and cloud – to align with the needs of the application.

Cloud-based AI training is the next phase of AI within machine vision and will continue to strengthen the machine vision market. Together, the power of the cloud, AI, and advances in computing and capture hardware have shifted the machine vision conversation from “Can it be done” to “Let’s do it.”

For example, there’s a stark difference between using a 3D and 2D camera, but both solutions bring a different perspective to solving different manufacturing problems. However, advancements in 3D cameras will impact the machine vision market for the next decade. Both the rise of 3D and AI have allowed us to solve complex manufacturing challenges that couldn’t have been entertained previously.

If you’re using a machine vision system without 3D or AI, it’s time to consider a second look and modernize to a machine vision system that is powered by AI and deep learning. It will be critical along the path to digital transformation and operating a connected factory.

When Worlds – And Machine Vision Users – Collide

As technology evolves, so do the people who use and purchase it. The face of machine vision buyers is changing, starting with millennials and even more so with Gen Z who grew up in a digital world – with smartphone megapixel cameras.

They’ve moved into their careers along with technological advances including in digital cameras – from small pixels to today’s powerful cameras. They have a generational advantage, meaning they understand technology and can adapt technologies like machine vision to their careers naturally.

There are also other important differences with our newest workforce. Gen Z’s values around work differ greatly from previous generations. A McKinsey study found that while Gen Z is open to working in manufacturing, they want an environment focused on people, not one that is optimized for machines. Compensation is less important to Gen Z than meaningful work, flexibility, and career advancement, according to McKinsey.

Use automated machine vision systems to attract Gen Z to new job opportunities including training, maintaining, and overseeing automated systems, for example, as well as skills and higher-value work. Technology like machine vision helps to advance the workforce overall and offers Gen Z manufacturing workers the chance to pursue what’s most important to them.

Generation Next

Think about how rapidly technology has evolved. It will continue to do so, bringing even more sophisticated solutions that impact everyday life, industries, and the people who work in them.

I can’t predict exactly what the next innovation in machine vision will be, but I expect it to continue to bring dramatic changes for the better.

Learn more about machine vision and industrial automation.