Switching to Digital Has Clear Advantages

"Advantages such as increased stability and better noise immunity make digital cameras attractive to current analog users"

Digital Vision Camera Index 

With the recent introduction of the A600 series, Basler has emphasized the appeal of digital cameras to customers who have only used analog units in the past. Since the compact A600 cameras use inexpensive PC adapter cards and standard IEEE 1394 cables, the system cost for an A600 digital camera is quite competitive with the system cost for a typical analog camera solution. And when an A600 camera is combined with a DCAM compliant driver in the PC, the A600 is essentially plug and play. But while these are key advantages, they don't really highlight many of the more technical aspects of why the switch to digital is so attractive. In this article, we will take a closer look at the full array of reasons why analog users should consider switching to today's digital cameras.

Increased stability over time and temperature

Analog cameras use circuits based on potentiometers, capacitors and amplifiers to perform such functions as white balance, gain, black level offset, and other functions critical to video performance. Digital cameras are calibrated and referenced in the digital domain. Once calibrated at the factory, the calibration and reference values are permanently stored in the camera's ROM. These values remain constant and minimize the drift over time and temperature associated with the discrete components used to perform the same functions in analog cameras.

Much better noise immunity

Analog cameras are usually based on a video output signal that has only a small voltage change from black to fully saturated white. Although coaxial cable does provide some level of noise resistance, only a few millivolts of external noise is relevant compared to the overall video signal level and can begin to affect the signal. Digital signals are either voltage or current driven around a threshold value. Anything above the threshold is seen as a digital "high" and anything below the threshold as a "low". External noise must reach a level capable of forcing the video signal above or below the threshold value before it can even begin to  interfere with the video signal. This level of noise rejection is inherently hundreds or even thousands of times greater than analog systems. In addition, RS-644, Camera Link, Channel Link, and other digital standards use differential signaling methods that further increase noise immunity. Once a signal is digitized, there is very little chance that it will be affected by electrical noise. With analog signals however, a small amount of electrical noise can significantly degrade the quality of the transmitted images. Both analog and digital cameras may be high quality video sources, but in noisy industrial environments, the captured images from a digital system will have the highest signal-to-noise ratio.

High performance video

Today's video customers have ever-increasing demands for their applications. Speed and resolution have become the foundation of video performance. Cabling and capture technology limit analog cameras to a maximum pixel clock of about 40 MHz. This ultimately limits the performance of the camera. Current digital standards allow for pixel clocks of 85 MHz and above. In addition, multiple taps and multiplexing methods for digital data transmission increase the capabilities of digital cameras even further. As video frame rates venture into the hundreds or thousands of frames per second and resolutions go to two megapixels, four megapixels and beyond, only digital solutions offer the speed and bandwidth necessary to handle video demands into the future. Cameras such as the Basler A504k with 500 frames per second at 1280 x 1024 (1.3 MPixel) resolution would not even be possible with analog only systems.

Ease of use

Digital standards such as IEEE 1394 (Firewire) offer users unprecedented access to industrial cameras and high performance video. PCs equipped with DCAM compliant software and drivers now directly support any IEEE 1394 based industrial digital camera. A user can connect an industrial spec IEEE 1394 camera to their system and live video appears on the default viewer with just a few mouse clicks. A  variety of software packages such as National Instruments LabView, Matrox Mil, and NorPix StreamPix are available to provide powerful tools for image processing and storage with IEEE 1394 cameras.

Advanced features

At the heart of any digital camera is a microprocessor, FPGA, ASIC, or other processing device. In recent years, these devices have become not only smaller and more efficient, but more powerful as well. In addition to simply operating the camera, manufacturers of digital cameras can use this power to provide advanced features such as shading correction, on-the-fly subtraction of a reference image, custom Gamma or other look-up tables, and more. For OEM customers, custom features can be added to their specification, making the camera an even more cost effective and efficient imaging device. These types of features and performance are not possible in cameras that are strictly analog.

No pixel jitter

Analog cameras use a "pixel clock" to determine when the value of each pixel within the analog video signal should be read. In many cases, the frame grabber generates the pixel clock using an electronic device called a Phase Locked Loop (PLL). During the exchange of the pixel clock signals between the frame grabber and the camera, small shifts in the timing of the pixel clock can occur. This constant shifting is referred to as "pixel jitter" and it leads to inaccuracies in the reading of pixel values. Even with high quality PLLs, jitter can be between 5 ns and 20 ns. Though this seems like a small number, it  can become significant when sampling pixel signals that may only be 25 ns long themselves. As shown in the graph above, pixel jitter can cause significant errors in the sampling of the video signal in analog systems.

In digital cameras, the video signal is digitized at the imager. A single clock is used both for sampling and for the pixel clock. No PLL is used and the pixels are clocked on and off in unison. This results in a higher sampling accuracy for digital cameras. Author: Anthony Pieri Technical and Marketing Communication Specialist

In summary

In the past, digital camera systems were significantly more expensive than comparable analog systems and were seen as more difficult to use. While advantages such as increased stability, better noise immunity, and no pixel jitter were attractive to analog camera users, they were not enough to convince these users to make the switch to digital. With the arrival of standardized interfaces for industrial camera systems and the compact Basler A600 camera series, the price difference has become an insignificant issue and ease of use is now a key feature. When you couple this with the higher performance range and the on-board processing capabilities of digital cameras, you see why the "analog goes digital" trend is picking up more and more speed in the marketplace.