Many tasks in the manufacture of products, including inspection, orientation, identification, and assembly, require visual feedback. Human vision and response, however, can be slow and error-prone due to boredom and fatigue. Replacing human interaction with machine vision can help automate factory operations and is worth considering in a growing number of applications.

Factory automation using machine vision offers many benefits. Vision systems can perform repetitive tasks faster and with greater consistency over time than humans. They can also reduce labor costs, increase production yields, and eliminate costly errors associated with incorrect assembly. The net result? Greater productivity and consistent delivery of quality products.

Implementing a successful machine vision system, however, is not a casual task. Selecting components and system programming must accurately reflect the application's requirements. Decisions need to consider more than initial component costs. Factors such as the time required for system development, integration with the factory system, operator training, maintenance, and software upgrades all should be evaluated before investing in a specific system design.

Define the requirements

The first step in selecting a machine vision system is to closely define the requirements:

What task does the system need to perform? Inspection requires an ability to examine objects in detail and evaluate images to make pass/fail decisions. Assembly, on the other hand, requires the ability to scan an image to locate reference marks and use those marks to determine placement and orientation of parts. A vision system designed for one task may be poorly suited to the other.

What are the key visual performance criteria? The vision system's camera and lens must perform at the right levels. Factors such as the smallest object or defect to detect, required measurement accuracy, image size, speed of image capture and processing, and the need for color discrimination all affect camera and lens choices.

What are the environmental factors? Some camera choices better suit stationary views while others are more suitable for handling linear object motion. Other factors, such as temperature, humidity, and vibration, may call for special system fabrication and assembly practices. The physical space available for installing the system can also influence camera and lens choices.

Beyond the system's physical requirements, developers should consider operational requirements:

Who will program the system? If configuration expertise is not available in-house, the user must depend on third-party support to make changes and correct errors. If the system needs periodic changes, such as to inspect a new product line, the question of programming becomes particularly important. A system set up for a single task that requires a system integrator to reconfigure it can interrupt production for extended periods. A system with enough flexibility to allow factory personnel to make adjustments may cost more to create, but will save production time later.

What equipment must the vision system interface with? A vision system that only activates a solenoid to eject failed parts from a production line is easier to implement than one that also reports results to a quality control network. Likewise, systems that inform and enable human operators have different needs than those that interface only to other machines.

What information must the system provide? Machine vision systems in factory automation seldom operate in a standalone mode. Instead, they must send information to other parts of the factory enterprise for various purposes. Quality traceability, for instance, requires that the vision system either log or report inspection results to enterprise-level computers. Highly controlled operations such as pharmaceutical manufacturing may also require the logging of access to the vision system, sending such data to a secure drive.

What are the operator requirements? If operators are required to periodically change inspection criteria, such as acceptable tolerances, the software must support such manipulation. System software may also need to provide security to prevent unauthorized access and include safeguards to avoid the introduction of erroneous values. Software design can affect the type of training that operators require as well as the ease of system maintenance and modification.

Building a vision system

While the answers to these questions depend on the application, all machine vision systems for factory automation share some fundamental attributes and behaviors. In particular, all vision systems strive to image or inspect a scene or object, operating continuously at the fastest practical speed. In doing so, they usually cycle through the following steps:

  • Position the object or camera so that the camera can view the object
  • Capture an image with a camera
  • Process the image
  • Take action based on the image processing results
  • Communicate results to operators and other factory systems

Because of this commonality, examining a specific application such as inspecting objects on an assembly line will help illustrate the method by which developers can build a suitable machine vision system.

Essential elements of an inspection system include a delivery vehicle, vision system, response system, and sensors to trigger image capture and system response. The delivery vehicle positions the object for inspection. The vision system (camera, optics, lighting, and image processor) captures and processes the object image to determine a pass/fail response. The response system takes the required action and communicates results to operators or other systems. Sensors trigger the vision and response systems, identifying when the object is positioned properly for the systems to perform their tasks.

A first step in developing an inspection system is to determine how parts are to be placed in front of the camera. In this example, the delivery vehicle is a conveyer belt that carries objects past the vision system at a constant speed. Other delivery vehicles include part feeders, robotic arms, or a human placing an object in a station for inspections. Choosing a delivery system is often the hardest part of a factory automation design because it places restrictions on the remaining system choices.

Next, developers must determine the most appropriate method for triggering the vision system (to capture the image) as well as the response (to take action). In the case of conveyer belt delivery, an appropriate sensor might be electronic photo-eyes that produce a signal when an object passes between them. With other delivery vehicles, sensors such as proximity switches or programmable limit switches could work.

Image capture, processing, and results evaluation are tasks for the vision system, which determines if the object being inspected is within acceptable quality tolerances and then tells the response system what to do. A separate vision controller such as a vision appliance may handle the image processing, or those functions may be integrated into a smart camera.

Parameters to be evaluated and object characteristics will influence the choice of camera, optics, lighting, and image processing software. For example, reading an identification number requires close-up imaging, front lighting, and optical character recognition software. Inspecting the fill level in a soda bottle requires back lighting and the ability to detect the position of the liquid's surface.

The final step is to define how the system will respond to its decisions. In this example, the vision controller triggers a PLC to push rejected parts off the conveyer to another delivery system, allowing acceptable parts to continue undisturbed. The controller may also send decision results to higher level networks for quality control and traceability purposes.

Special thanks to Steve Geraghty, director of Dalsa's Industrial Products, for this month's tips. For more information, visit www.dalsa.com or call (978) 670-2000.