Looking at transfer bandwidth from another perspective

One thing I’ve learned from my years in Machine Vision is that performance and bandwidth are never enough to satisfy leading edge applications; hence the constant quest to increase the transmission speed of camera interfaces. Teledyne DALSA has always been at the forefront of this chase and the GigE Vision standard is no different: GigE Vision 2.0’s primary mandate was to find ways to increase the bandwidth between camera and PC.

For obvious reasons, people looking to increase bandwidth take a straightforward approach to the problem: simply increase clock speed on the cable! For GigE Vision, this leads to the usage of 10 gigabit Ethernet. This is certainly a valid approach – although an expensive one, may I say. But one can look at this problem from another perspective: if the goal is to transfer more images in a shorter period of time, why not consider image compression?

You are going to say image compression leads to loss of information. And you are right. Compression means throwing away “less important” information, typically the high frequency content. But a study from Ebrahimi, Chamik and Winkler (see reference below) teaches us that the most advanced image quality authority, the human eye, does not see any visible degradation when using JPEG with a compression ratio less than 5. And if your eyes cannot see compression artifacts when they can definitively see sensor-induced noise, wouldn’t you consider image compression a valid option to reduce bandwidth? Additionally, most image compression algorithms, to a greater or lesser degree, filter out high frequency content, and noise typically fits in this category.

GigE Vision 2.0 (a free download from the AIA web site) is the first Machine Vision camera standard to natively enable usage of image compression. It introduces support for JPEG, JPEG 2000 and H.264. I admit data compression is not suitable for all applications, but this is an interesting low cost way to look at increasing the frame rate between the camera and the PC, with the added benefit that data archiving frequently relies on compression. So the next time you need to increase transmission bandwidth, you might open your wallet or you might take a look at alternative strategies, such as image compression.


Reference: JPEG vs JPEG2000: An Objective Comparison of Image Encoding Quality, Farzad Ebrahimi, Matthieu Chamik, Stefan Winkler, http://stefan.winkler.net/Publications/adip2004.pdf

About Eric

Eric is in charge of R&D activities at the Montreal office of Teledyne DALSA where he is surrounded by talented people working on the technologies of tomorrow. Chair of the GigE Vision committee, he enjoys reading and writing machine vision standards, especially the thicker ones.
Posted on by Eric. This entry was posted in Cameras, Frame grabbers, Interface Standards, Machine Vision. Bookmark the permalink.

One Response to "Looking at transfer bandwidth from another perspective"

  1. stephane francois says:

    Hi Eric,
    The machine vision market has effectively been preferring uncompressed format for transfer of images.
    The first reason was that compression was another step in the chain and required special handling, so before FPGA libraries and special chips, we had to integrate it. For machine vision, that meant more work and more cost.
    The second was that compression was not as advanced as it is today. As you mentioned, in those days, jpg was the compression, by hardware, because software processing power couldn’t offer a decent solution; it was too slow and the hardware solutions were too basic so it meant poor image quality. And in machine vision we process the images, not for a human to look, having image alterations would impair the ability to extract the information we need in the image.
    The third is that in most application, we are getting a lot of images but as compression is mostly dependent on the image content, we would have fluctuating needs for the streaming. And that would mean more difficulty managing the time constraints and possible overflow in the system.

    So it was easier, and safer to use uncompressed images.

    Nowadays, compression is an affordable option as many cameras have pretty good processing power already.

    It is great that GigE vision would integrate it to make it more accessible. It also places the machine vision market on par with sister industries such as the document imaging, security, video broadcast industries, all manipulating images and using compression.
    You have also companies such as Norpix and IO Industries that have specialized for years in offering compression mostly for writing images to disk, where the bottleneck is in such applications. They of course offer lossless compression as an option.

    So the technique is not new but it will be easier to make use of it with the help GigE Vision 2.0 compliant products.

    Stephane Francois @ Computed Vision