Machine Vision – A Guide to Understanding the Basics

Machine vision (also referred to as “vision systems” or “industrial vision”) is the use of digital sensors (wrapped in cameras with dedicated optics) that are connected to processing software and hardware algorithms to visually inspect… pretty much anything. Machine vision is a true multi-disciplinary field, covering computer science, mechanical engineering, optics and industrial automation. While historically the tools of machine vision were concentrated on manufacturing, that aspect is rapidly changing, moving into research, medical applications and even movie making.

A machine vision system integrates image capture and processing systems (computer software and hardware) with digital input/output devices (cameras) and computer networks (to save and share the image data) to drive real-time quality control systems, or to direct equipment, such as manufacturing robots. This is marginally different from “computer vision,” which is mainly about image processing. Instead, machine vision systems are constructed for visual inspection and control under challenging industrial applications that require high-magnification, high-speed, 24-hour operation, and/or repeatable measurements.

Doing a Thankless Job Really Well

No machine can beat human vision for versatility; however, other human weaknesses restrict their productivity in a manufacturing environment. As these kinds of jobs are “no fun”, distraction, boredom, and fatigue normally lower human performance in vision-related factory operations like inspection, which require constant operation, high speeds and a lot of monotony.

Factory automation employing a machine vision system in such operations, then, provides a number of benefits. Compared to humans, machine vision systems can perform repetitive operations faster and more accurately, with greater consistency over time. They can lower labor costs, boost production yields and eliminate expensive errors related with incorrect or incomplete assembly. They can help automatically identify and correct manufacturing issues on-line by forming part of the factory control network.

This can be a part of the inspection procedure itself (for example examining a measurement or identifying whether a character string is printed properly) or through some other input required for control (for example robot control or type verification).

The machine vision system can include nearly any number of cameras, all capturing, deducing and signaling individually to a central control system according to a prearranged tolerance or requirement.

This makes it simple to automate difficult multi-stage visual inspection operations, or just perform easy inspections at speed and scale. These kinds of applications include identification, positioning, measurement, verification and flaw detection. Machine vision system has driven ever higher standards in manufacturing — higher yields, better product quality and lower production costs. The net result is higher productivity and value-added customer satisfaction through the reliable delivery of quality products.

No machine beats human vision for versatility, but other human weaknesses limit their productivity in a manufacturing environment. Because these kinds of jobs are “no fun.”

Where Do You Want to Go Next?

Since machine vision systems are used almost everywhere — on the road, in factories and in space — the pieces they are constructed from need to be standardized but customizable. Basically, each component of a machine vision system does one thing. This makes it easier to switch parts to incorporate a new capability or level of performance without replacing parts pointlessly. It also makes it easy to change a part that has failed. Cameras have gone from the weakest link in the chain to the pixel pushing drivers of innovation in the rest of the system. This is contrary to a consumer-level digital camera or even digital film camera, which will record light, but also perform varying degrees of processing to enhance image quality or decrease file sizes as well as record out to a transferrable medium.

Elements of Your Awesome Vision System

A characteristic machine vision system will be part of an automated production process comprising of the following components:

Camera

Systems will have one or more digital cameras that record monochrome, color or wide-spectrum images. Similar to consumer cameras, the optics/lenses attached provide a specific field of view and available light. To choose the proper lens, one will first need to know the field-of-view (FOV) and the working distance. The FOV is the size of the area one wants to capture. The working distance is approximately the distance from the front of the camera to the part being examined. A more precise definition takes into consideration the structure of the lens.

Trigger

This component makes certain that specific action occurs in response to something the camera sees. A synchronizing sensor for part detection (usually an optical or magnetic sensor) to trigger image acquisition and processing, and some form of actuators to sort, route, or reject defective parts.

Image Processor

A computer program (generally running on a version of Microsoft Windows) to process images, detect, measure, compare etc. so as to confirm a quality criteria has been met or to provide type verification or robot control to another control system.

Cables

Input/output hardware such as digital I/O or communication links such as Ethernet, CameraLink, USB, Firewire, etc. to report results and to automatically reject components.

Building Your Own System

Any machine vision system comprises of a few essential components. The image sensor/ camera captures light and changes it into an electrical signal that computers can comprehend and transmits it to a processing engine that renders and communicates the result to a computer that can do something with it.

Figure 1. A basic machine vision system. The target is lit as it moves across the camera’s field of view, and the data is transferred to the vision processing system.

While the additional features and operational requirements depend on the application, all machine vision systems share some important behaviors and attributes. Systems all have a need to image or inspect an object or scene, functioning on a continuous basis at the fastest practical speed.

Figure 2. An inspection system that not only visually inspects the target, but feeds the acquisition system to react based on the scanned information. Objects that don’t meet specifications are rejected from the assembly line.

The vital elements of an inspection system, illustrated in Figure 2, include the vision system, the response system, a delivery vehicle and sensors to trigger image capture and system response. The delivery vehicle positions the object to be inspected. The vision system, which includes camera, lighting, optics and image processor, captures and processes the object image to establish a pass/fail response. The response system takes the necessary action as well as communicates the results to operators or other systems. The sensor triggers a reaction, identifying when the object is positioned correctly for the system to do their tasks.

The first, and questionably most crucial, step in building an inspection system is positioning or “staging”. After all, no amount of processing can recover information that a sensor did not capture at the scene. This is the determination of how a subject will be positioned in front of the camera for imaging. Figuring out how objects will be delivered to the camera and sensor can also be the toughest part of machine vision design. The delivery choice will frame all the remaining system choices, including camera, sensors, lighting and response systems. And it looks like the camera can go nearly anywhere. Here are a few extreme examples. Instead of objects being supplied via conveyer belt in an orderly and predictable fashion, they could run or even fall in.

Figure 3. The conventions of an inspection system get turned on their head. Multiple cameras are ringed around the target. The multiple visual information feeds are joined together to build a more complete picture - whether it’s a winning touchdown or a secret spy skydiving.

In the case of systems designed by Replay Technologies for their stadium systems, the target is in the middle. It can be used to capture the whole basketball or football game, or it could be a skydiving scene in James Bond’s Quantum of Solace. In the latter, the cameras were positioned in a ring around a wind tunnel to guarantee that action was captured from several angles.

Figure 4. Full motion picture machine vision. Instead of the target moving at high speed on an assembly line, the front end of the vision system is in motion.

The system shown in Figure 4, applies for filming movie stunts, reversing the motion relationship that characteristic vision assumes: here the camera is moving. That which is being filmed, specifically the motorcycle and rider, does not move in relation to the camera. The other subject being recorded, the scenery is stationary. It is the camera that is moving at 240 km/hour.

Getting Better All the Time

It is not just what is possible — it is what one does with it. Robust industry innovation and competition have driven vivid improvements in what processing, cameras and automation can achieve. Visual inspection makes it possible to exactly guide robots and other machinery to produce more adaptable manufacturing processes. Custom products and more diverse production environments are possible when machines can see as well as construct and much of this is taking place on line, making for a more effective and faster inspection and preventing wastage or scrap material as they function.

With an endless quest for better quality and the availability of higher resolution cameras with faster and more effective processing power, increasing bandwidth and access to cheaper cabling and storage, there really is no limit to where one might find what began as machine vision. The sky (and the universe) truly is just the beginning.

This information has been sourced, reviewed and adapted from materials provided by Teledyne DALSA.

For more information on this source, please visit Teledyne DALSA.