GigE Vision 2.0, the fourth version of the popular interface standard for high-performance industrial cameras, released in 2011. This release improved connection to faster sensors which is done in a number of ways.
The GigE Vision 2.0 offers a new and advanced support for 10 Gigabit Ethernet by increasing the transfer throughput through augmentation of the clock speed on the cable. This can be done for a standard copper cable (CAT-6a) or using fibre optic. This can also be done through link aggregation, combining more than one cable but keeping the clock rate the same. So by using 2 gigabit/sec cables, a GigE Vision camera can double its maximum throughput from 125 to 250 MB/s. While this type of camera is not a new phenomenon, the GigE Vision 2.0 paves the way for smoother interoperability and system integration.
One of the main objectives of GigE Vision 2.0 has always been to increase the amount of information for transfer, which as described, can be done through augmentation of the total physical bandwidth. An alternative approach would be to reduce the amount of data to be transferred. To do this, GigE Vision 2.0 offers several data compression schemes: JEPG, JPEG 2000 and H.264. Although some information can be lost in the compression step, these compression standards have demonstrated good performance and are certainly a way to improve transmission bandwidth while simultaneously keeping system costs down. Whilst applications require archiving the images, native data compression within the transmission standard removes this extra burden.
As GigE Vision uses some packets to announce the format of the proceeding image, transfer overhead is higher for small images. As Ethernet allows for jumbo packets, GigE Vision 2.0 includes an “All-in Transmission” mode where a full image can be transmitted using a single packet. This gets rid of many packetisation overheads, further streamlining the transfer. This is particularly effective when small regions of interest must be transferred at a high rate from CMOS sensors and could also be used for some line scan applications.
By leveraging IEEE 1588 Precision Time Protocol to allow each camera on the network to be synchronised up to 1 µs, the GigE Vision 2.0 improves real-time synchronisation of multi-camera systems. It is now possible to schedule actions to be carried out at the same time by the various devices attached to the network. The 64-bit timestamp is attached to the image to facilitate the correlation of data coming from multiple cameras.
Sensor vendors use multiple taps to increase the readout throughput. In previous GigE Vision versions, the camera had to reconstruct the image locally in raster-scan format before transmission. GigE Vision 2.0 uses “multi-zone Image” to simplify this task. By dividing an image into horizontal bands, the camera can now transmit packets from different bands in any order.
The GigE Vision standard supports various payload formats and since version 1.2, can be used by non-streamable devices, such as I/O boxes that provide a GenICam™ interface to read and write registers. This speeds up development because the same function calls can be used to configure any GigE Vision device. GigE Vision also improves the device discovery process by allowing usage of multicast DNS (mDNS), whereas previous versions only allowed for a polling mechanism to enumerate devices.
Over the years we have seen the introduction of many new machine vision camera standards. This emphasises the need to ensure coherency between standards to allow them to function together in machine vision systems. To do this, the team at GigE Vision generalised the definition of its pixel formats and created a separate document titled “Pixel Format Naming Convention”. The document is managed by the AIA and provides the layout of the supported pixel formats, enabling re-usability across camera interface standards.
By offering performance unseen in typical machine vision cameras, GigE Vision 2.0 offers the potential to expand beyond the traditional markets of the machine vision industry. With improved connection to faster sensors, smoother system integration and optimised information transfer, the GigE Vision 2.0 brings new and improved functions to interface standards.
