Machine vision requirements for better performance and higher resolution continue driving developers to incorporate digital cameras into their solutions. This trend will likely accelerate as the price and performance of digital cameras improves. This article will provide you with information on digital camera technology and key factors to consider when choosing a digital camera and associated frame grabber — assuming that all upfront analysis has been performed and that a digital imaging solution is required.
Why Use a Digital Camera?
Analog cameras continue to be the dominant choice in most machine vision applications for several reasons:
- Huge installed base,
- Mature technology with well-known standards,
- Performance is often adequate for the application, and
- Inexpensive, readily available cabling.
Until recently, there were several significant disadvantages to using a digital camera. Digital cameras typically were priced higher than similarly performing analog models. And, in addition to being expensive and bulky, digital camera cabling often was not easily interchangeable between different types of cameras and frame grabbers. However, with the introduction of the Camera Link™ and GigE Vision standards, digital camera cabling issues have largely been eliminated, while their performance continues to steadily improve, in many cases far exceeding that of any analog camera.
So why use a digital camera? Digital cameras can deliver higher data rates, higher resolution, and higher bit depths than analog cameras. Digital transmission is also inherently less susceptible to noise than is analog — a key consideration for plant environments.
One of the ways that digital cameras achieve these high data rates is by providing data on multiple taps, or channels. While this has been done with analog cameras, synchronizing the taps is more difficult with analog, as the frame grabber must be able to lock with subpixel accuracy from multiple sources. Since a digital camera performs the data alignment, synchronizing and interfacing to multiple taps using these sources is assured.
Digital Camera Standards
When they were first introduced, commercial digital cameras used parallel TTL (Transistor-Transistor Logic) level signal outputs. These single-ended, 5V signals were prone to noise interference and could not be transmitted over long distances. As frame rates and resolutions increased, it became necessary to transition to a signaling standard that could accommodate the higher data rates.
RS-422 was the next step in the evolution of commercial digital camera standards. RS-422 parallel output cameras provide potentially higher data rates than TTL by using a lower voltage and higher clock frequencies. The differential signal strategy provides better noise immunity, and because it is a differential signal, the number of data lines doubles over that of a single-ended TTL signal and, therefore, cabling becomes an issue. Cables for RS-422 digital cameras typically are unique to the camera/frame grabber pair, so cable costs are high.
Parallel signal standards next moved to LVDS (Low-Voltage Differential Signaling), which has a lower signal voltage than RS-422. LVDS also accommodates higher transmission frequencies (>40 MHz) and consumes less power than TTL or RS-422. However, the same issues with cable costs and pin counts remain.
With the availability of serial digital standards like Camera Link, many of the problems inherent in parallel signaling were alleviated. Because it transmits data serially, the number of required signals per tap is greatly reduced. And, since Camera Link is a standard, cables are interchangeable between cameras and frame grabbers. Camera Link allows up to four independent control signals for events such as reset and integration, as well as integrated serial communication for camera configuration and status, so that Camera Link cameras incorporate many of the control features required for machine vision applications. The ability to dynamically reconfigure and query the status of the camera under software control, while sometimes available, is rare with analog cameras.
What to Consider When Selecting a Digital Camera
The initial step in any imaging application is determining what to acquire and what features within that image should be highlighted. As an example, let’s consider an application in electronics manufacturing that defines an object to be acquired, such as a populated circuit board. The overall image will contain many features of interest for inspection. The electronic component features — such as size, identity, quantity, location, and orientation — are all critical for the correct and complete assembly of a circuit board. Other features may include markings on the board and characteristics of the bare board itself, such as material density and routing/continuity of the circuit board traces. The features of interest within the object’s image become the inspection points, and the characteristics of those features are the inspection criteria.
The resolution of the image required for inspection is determined by two factors: the field of view required and the minimal dimension that must be resolved by the imaging system. Using a basic example, if a beverage packaging system must verify that a case is full prior to sealing, it is necessary for the camera to image the contents from above and verify that 24 bottle caps are present. It is understood that since the bottles and caps fit within the case, the caps are then the smallest feature within the scene that must be resolved.
Once the parameters and smallest features have been determined, the required camera resolution can be roughly defined. It is anticipated that, when the case is imaged, the bottle caps will stand out as light objects within a dark background. With the bottle caps being round, the image will appear as circles bounded by two edges with a span between the edges. The edges are defined as points where the image makes a transition from dark to light or light to dark. The span is the diametrical distance between the edges.