Skip to main content

How Deep Learning Improves Machine Vision in Factories

Factory automation is evolving, once again. Machine vision technology remains popular in the manufacturing environment, due to its proven track record of results. However, the introduction of artificial intelligence (AI) technology and machine learning applications will transform many conventional factories.

The emergence of deep learning apps creates expanded capabilities and flexibility, leading to more cost efficiency and higher production yield. Deep learning-based machine vision techniques within smart manufacturing will experience a CAGR of 20 percent between 2017 and 2023, with revenue that will reach $34 billion by 2023, according to the latest market study by ABI Research.

Machine Vision Market Development 

Manufacturers need to improve their production yields and workflow efficiency. Legacy machine vision is easy to implement but is somewhat limited. Current solutions that are widely deployed in quality control, safety inspection, predictive maintenance, and industrial monitoring rely upon preprogrammed rules and criteria, supporting limited ranges of functions.

In contrast, AI deep learning-based machine vision is highly flexible due to its ability to be trained and improved using a new set of factory data, enabling manufacturers to incorporate updates and upgrade quickly.

"This is in part driven by the democratization of deep learning capabilities. The emergence of various open source AI frameworks -- such as TensorFlow, Caffe2, and MXNet -- lowers the barrier to entry for the adoption of deep learning-based machine vision," said Lian Jye Su, principal analyst at ABI Research.

These AI frameworks can be deployed using on-premise IT infrastructure and vendor software suites. In the past, the choice of machine vision solutions was limited to a handful of companies that performed relatively simple image processing operations. With deep learning-based machine vision, manufacturers can now develop their own deep learning-based machine vision systems.

In addition to cameras, deep learning-based machine vision can also incorporate data collected from various sensors, including LiDAR, radar, ultrasound, and magnetic field sensors. The rich set of data will provide further insight into other aspects of production processes.

Compared to conventional machine vision, which can only detect product defects and quality issues which can be defined by humans, deep learning algorithms can go even further. These AI algorithms can detect unexpected product defects, providing flexibility and valuable insights to manufacturers.

According to the ABI assessment, manufacturers are encouraged to work with a wide range of vendors, including industrial cloud platform, camera and sensor suppliers, and public cloud vendors. Deep learning-based machine vision requires a robust cloud platform that will enable condition-based monitoring, sensor data collection, and analytics.

Outlook for Machine Vision Application Growth

Unlike conventional machine vision which relies on line-by-line software coding, deep learning-based machine vision models can be deployed by users without significant developer experience, as these models undergo unsupervised learning based on data gathered.

"Manufacturers are opening up to adopting AI capabilities into their workflow. Deep learning-based machine vision will serve as the right catalyst to drive progress. Startups that launch as deep learning-based machine vision solution providers are also beginning to enable big data processing, process optimization, and yield analytics on their platform," concluded Su.

Popular posts from this blog

GenAI Can Supercharge Economic Recovery

The Economic Recovery Corps (ERC) is a new, collaborative initiative designed to accelerate recovery from the COVID-19 pandemic in communities and regions throughout the U.S. by connecting organizations with the talent and capacity needed to advance new ways of doing economic development. However, it's unknown if new technology will be a key component. For example, less than 25 percent of government organizations will have Generative Artificial Intelligence (GenAI) enabled citizen-facing services by 2027, according to the latest worldwide market study by Gartner. Furthermore, fear of public failure and a lack of community trust in government use of the technology will slow adoption for external use by a nation's citizens. Government GenAI Market Development Like many organizations over the past 15 months, federal and regional governments have been exploring the opportunities and risks associated with the emergence of GenAI tools. Gartner’s annual global survey of over 2,400 CIO