Three things to consider when choosing a smart camera for an embedded vision application
Users choosing smart cameras for embedded vision applications should think about the camera's processor, vision software and automation system integration.
Smart cameras are more popular than ever, allowing for more facilities to take advantage of embedded vision in a wide range of applications. If you’ve been considering smart cameras as a possible automation solution, you may be struggling with which camera you should pick.
Embedded vision offerings come in different shapes and sizes with a variety of features. Even the experts find it difficult to agree on the technical definition of a smart camera. It’s generally agreed that a smart camera includes a lens, image sensor, and processing chip.
A smart camera differs from a conventional camera in that the primary output of a smart camera isn’t the image, but rather a decision or some other form of information. This is all handled within the camera and doesn’t need input from an external computer system.
Smart cameras fill the needs of a huge market, and the options seem to be limitless. The following characteristics will help you make an informed decision about the smart camera you choose to implement in your embedded vision application.
1. Smart camera processors
The evolution of processor performance is a major contributor to the growth of smart cameras and embedded vision. Processors are more compact and faster than they were, even just a few years ago. As a result, they can now process and interpret images in real-time instead of being held to simple tasks like reading barcodes. Today, there are a number of processors available to choose from with varying degrees of capabilities, flexibility, and speed.
2. Embedded vision software
To help simplify manufacturing, smart camera producers will often build models with the same image sensor, processor, and form factor. The differentiator is sometimes only the software loaded onto the camera. This allows manufacturers to offer simple-to-use software for specific embedded vision applications. Smart camera users should keep in mind the flexibility of the software needed for future applications, not just its plug and play capabilities.
3. Automation system integration
Smart cameras are usually just one part of a larger automation system. An embedded vision device must be able to send a decision to another component. These include robot controllers, programmable logic controllers (PLCs), cloud servers, and hardware interfaces. Be sure to choose a smart camera that has outputs compatible with existing and future devices. Devices will also need to support the same protocols, including DeviceNet, EthernetIP, and serial.
This article originally appeared in Vision Online. AIA is a part of the Association for Advancing Automation (A3), a CFE Media content partner. Edited by Chris Vavra, associate editor, CFE Media, firstname.lastname@example.org.