- THE MAGAZINE
- WEB EXCLUSIVES
Selecting a machine vision camera is not a simple task. With a large number of industrial cameras and camera vendors to choose from, and an array of interface, imaging software and sensor choices, it is easy to become overwhelmed.
In today’s crowded marketplace, multiple vendors offer what appear to be identical cameras-all at varying price points and degrees of quality.
With the range of cameras available to today’s buyer, it is necessary to look beyond the specs and evaluate the quality of the physical camera, its software and the practices of the camera vendor.
Industrial cameras are designed to handle a range of environmental stresses that include heat, g-forces and electrostatic discharge (ESD), to name but a few. The degree to which an imaging system can handle these stresses depends on the quality of the camera boards used in the manufacture of the camera. Machine vision cameras also need to be properly calibrated to meet the unique needs of industrial environments.
When seeking out a machine vision camera:
- Trust but verify. Every vendor worth its salt will perform some degree of camera testing. Ask the camera vendor specific questions about its board testing and inspection processes. Are all boards tested in-house before they go out the door? Are they inspected by automated and human inspection? The vendor’s answers to these questions will indicate the quality of testing being performed.
- Inquire about quality assurance. Ask the vendor whether its cameras go through automated individual calibrations and if image quality assurance procedures are in place. If not satisfied with the answers, consider another vendor.
Machine vision camera software should function as expected, be intuitive and easy to use and been developed following sound software engineering and design principles. How, though, can one evaluate whether the embedded software within a camera is adequate?
- Ensure that the software is developed in-house or by a reputable dealer. Ideally, the camera software should be developed in-house by the company creating the camera hardware. If, however, the vendor does not use its own software, inquire about the software manufacturer and do some due diligence. Is the company that developed the software well known? Does it specialize in machine vision camera software? The answers to these questions will help you determine the quality of the software in place.
- Ensure that the hardware and software staff work closely together. Even if the software is developed in-house, it is important to make sure that the lines of communication between hardware and software staff are open. Ask the vendor about their relationship with the software designers and whether hardware and software staffs have knowledge of each other’s domains.
Are the software designers familiar with the field-programmable gate array (FPGA) circuits that will be used in the camera? Are the hardware designers intimately familiar with the software to be used? These questions will uncover the relationship between the vendor’s hardware and software operations.
- Make sure that the software is tested regularly. Just as with hardware, a solid testing process should be in place for software. This testing should include automated testing, stress testing and human testing.
All machine vision camera vendors are not created equal. As with any industry, some vendors will be reputable and others will not. The quality of the vendor will directly affect the quality and reliability of the camera, so choosing the right vendor is one of the most important decisions in the search for an industrial camera.
- Look for a vendor that balances price and quality. In some cases, a low-cost camera may be perfectly adequate for one’s needs. Be wary, though, of vendors who offer cameras at prices that are too good to be true. Nonreputable vendors will often “cheap out” on parts-leading to unreliable cameras that perform poorly or have a short shelf life.
To determine the quality of a company’s workmanship, ask the vendor about their return material authorization (RMA) rate. A company that is receiving a high number of returns is likely putting out subpar products.
- Choose a vendor with a clear support program. No two industrial applications are the same and machine vision cameras often need to be customized, modified or replaced. Be sure to inquire about the vendor’s support processes. Do they offer a dedicated support phone number? If purchasing through a distributor, who is contacted for support-the distributor or the vendor?
If evaluating a camera in-house, use the evaluation period to gage the quality of support offered by the vendor. Do they respond quickly to questions? Do they brush off requests or sound annoyed when receiving a call? The actions of the vendor during the evaluation period are a good predictor of how they will interact after the camera is purchased.
- Ask if the vendor uses its own cameras. There is no better way for an engineering team to know its cameras than to work with them on a daily basis. The best vendors are often the biggest consumers of their own products. Ask the vendor how it uses its own cameras and whether staff are permitted to use, test and suggest improvements to products.
- Look for an experienced team. The best camera vendors will have staff with a wide range of hardware and software experience. Look for a staff with electrical engineering, FPGA, firmware, device driver and application layer experience.
- Look for a diverse application portfolio. No two industries are the same and no two companies will use a machine vision camera in the same fashion. Look for a vendor with customers across a range of industries and of various sizes. Good vendors learn from their customers and apply best practices from one application to another.
- Seek out evidence of a constantly evolving product line. A vendor with multiple product lines and frequent product releases is likely on top of market trends and continually improving its cameras.
While the array of machine vision vendor and camera options on the market may appear daunting, narrow the choices by applying the rules of thumb above. By evaluating the quality of a physical camera, its software and the practices of the camera vendor, one can significantly increase their chances of imaging success. V&S
- Ask the vendor whether its cameras go through automated individual calibrations and if image quality assurance procedures are in place.
- Ensure that the software is developed in-house or by a reputable dealer.
- The best vendors are often the biggest consumers of their own products.