Standards in Machine Vision explained

There are several standards available in machine vision products, each with its own specifications and requirements. Some of the most common standards include CoaXPress (CXP), CameraLink High Speed (CLHS), HD-SDI (High Definition SDI, or SDI-12), Camera Link, GigE Vision, and USB3.

CoaXPress

CoaXPress is a high-speed communication standard that uses coaxial cables to transmit data between cameras and computers. CoaXPress supports data rates of up to 12.5 Gbps and can transmit data over distances up to 40 meters.

If a longer distance of cable is required there are also solutions available. The KAYA Instruments’ CoaXPress Range Extender over Fiber is industry’s first CoaXPress range extender which provides a high-resolution stream interface for distances up to 80km in single-mode and up to 300m in multi-mode. Optic fiber is electrically isolated, hence it does not radiate nor is it susceptible to electromagnetic interference, also eliminates the problems associated with grounding. The fiber cable is not easily tapped, providing more secure communication. The system is constructed of two convertors, one on the camera side and one on the host side. The converter on the camera side can provide power to the camera over CoaXPress link, while the converter on the host side can sink power from the frame grabber. The converters use flexible SFP+ modules for optical connection that can be easily changed. The range extender is able to provide an uplink of up to 12.5Gbps and downlink at 20.83Mbps.

CoaXPress is a high-speed communication standard for machine vision applications. It uses coaxial cables to transmit image and control data from a camera to a computer. CoaXPress was developed by the Japan Industrial Imaging Association (JIIA) and first introduced in 2008. The standard has since been updated to include additional features and capabilities.

CoaXPress is designed to provide a high-speed, low-latency interface between cameras and computers for high-performance image processing applications. It supports data rates of up to 12.5 Gbps over a single coaxial cable and can transmit data over distances of up to 40 meters. CoaXPress also supports power-over-coax, which allows cameras to be powered directly from the CoaXPress interface.

The features of various CoaXPress frame grabbers can be compared using a table.

CoaXPress includes several features that make it well-suited for machine vision applications. These features include:

• Multi-channel support: CoaXPress supports multiple data channels, each capable of transmitting image and control data. This allows multiple cameras to be connected to a single interface, reducing the number of cables required.
• Real-time communication: CoaXPress provides low-latency communication between the camera and the computer, which is essential for real-time image processing applications.
• Bidirectional communication: CoaXPress supports bidirectional communication between the camera and the computer. This allows for control signals to be transmitted from the computer to the camera, enabling advanced features such as trigger control and synchronization.
• Triggering support: CoaXPress supports advanced triggering features, such as precise timing control and multiple trigger modes. This is useful for applications such as high-speed inspection and motion analysis.
• High resolution and color depth: CoaXPress supports high-resolution images and high color depths, allowing for detailed and accurate image capture.
• CoaXPress has several advantages over other communication standards in machine vision applications, such as Camera Link and GigE Vision. It provides a higher data rate, lower latency, and more reliable communication between the camera and the computer. Additionally, CoaXPress is a mature standard that is widely used in the industry and has a large ecosystem of compatible products.

The various attributes of some CoaXPress cameras can be compared in table format.

In summary, CoaXPress is a high-speed communication standard for machine vision applications that uses coaxial cables to transmit image and control data from a camera to a computer. It provides high-speed, low-latency communication, bidirectional communication, advanced triggering features, and support for high-resolution and high-color depth images. CoaXPress is widely used in the industry and is well-suited for high-performance image processing applications.

CoF “CXP over Fiber”

CXP over fiber (Cof) refers to the use of fiber optic cables for transmitting data using the CoaXPress (CXP) standard. CoaXPress is an industrial communication standard that is widely used for high-speed image and data transfer in machine vision, medical imaging, and other applications.

Cof (CXP over fiber) offers several advantages over traditional copper-based CXP connections. Fiber optic cables have much higher bandwidth capabilities and can transmit data over longer distances without signal degradation or electromagnetic interference. Additionally, fiber optic cables are lightweight and flexible, making them easier to install and maneuver in tight spaces.

The use of fiber optic cables also allows for the implementation of remote power over the same cable, which can simplify system designs by reducing the need for separate power cables.

A frame grabber with optical fiber SFP connectors is required to receive these images before transmission, usually via PCIe, to the host.

In summary, the use of CXP over fiber can provide higher bandwidth, longer transmission distances, and greater flexibility and convenience in installation for industrial communication applications that require high-speed data transfer.

CLHS “CameraLink High Speed”

Camera Link HS (High Speed) is an industrial communication standard for transmitting high-speed image data from digital cameras to image processing systems or computers. It was developed by the Automated Imaging Association (AIA) and was first released in 2011.

CameraLink HS is a successor to the original Camera Link standard, and it offers higher data transfer rates and longer cable lengths than the original CameraLink. The standard defines a point-to-point connection between a camera and a frame grabber or other image processing system, with data rates of up to 50 Gbps over fiber optic cable.

CameraLink HS is designed to support a wide range of industrial applications that require high-speed image acquisition and processing, such as machine vision, medical imaging, and military and aerospace imaging.

CameraLink HS X-Protocol is an extension to the Camera Link HS standard that provides a standardized method for configuring and controlling multiple cameras in a synchronized manner.

The CLHS X-Protocol defines a master-slave relationship between cameras, where one camera acts as the master and the others act as slaves. The master camera sends synchronization signals to the slave cameras, which use these signals to synchronize their image acquisition and transfer processes. The X-Protocol also allows the master camera to configure various camera parameters, such as exposure time and gain, on the slave cameras.

The CLHS X-Protocol is particularly useful in applications where multiple cameras need to be synchronized, such as in stereo vision or multi-camera inspection systems. It simplifies the process of synchronizing cameras and ensures that all cameras are operating at the same settings and producing images at the same time.

CLHS Frame grabbers receive and pre-process images from CameraLink HS cameras before presentation to the host, usually via PCIe, for further processing by the CPU or even NVIDIA Jetson GPU.

HD-SDI (High-Definition Serial Digital Interface)

HD-SDI stands for High-Definition Serial Digital Interface. It is a standard for transmitting uncompressed high-definition video signals over a coaxial cable. HD-SDI is commonly used in the broadcast and professional video industries, where high-quality video signals are required for live broadcasting, video production, and post-production work. It is also widely used in industrial machine vision applications.

HD-SDI cameras support video resolutions up to 1080p at 60 frames per second and is capable of transmitting high-quality video over long distances without signal degradation. It uses a 1.5 Gbps serial data stream to transmit uncompressed video, which results in high-quality images without the loss of detail or color information that can occur with compressed video formats. However, 3G-SDI (standardized in SMPTE 424M) consists of a single 2.970 Gbit/s serial link that allows replacing dual link HD-SDI. The 6G-SDI and 12G-SDI standards refer to 6 Gbps and 12 Gbps transmission rates.

HD-SDI is widely used in professional video cameras, switchers, routers, and other equipment used in video production and broadcasting, as well as machine vision in industrial applications. Its advantages include high video quality, low latency, and compatibility with existing coaxial cable infrastructure. However, it is limited to a maximum resolution of 1080p and requires dedicated cabling and connectors.

CameraLink

Camera Link is a popular standard for high-speed machine vision applications. It uses a serial communication protocol to transmit data between the camera and the computer. CameraLink supports multiple data formats, resolutions, and frame rates, and can transmit data over distances up to 10 meters, although with extra 3rd party products such as the CameraLink Range extender over fiber from Kaya Instruments, distances of up to 80km can be achieved in single mode, and 300m in multimode.

Camera Link is a digital communication standard used in machine vision applications to transfer image data from a camera to a computer. The standard was developed by a consortium of companies in the industry, including National Instruments, Matrox, and DALSA, and was first introduced in 2000. CameraLink is designed to provide a high-speed, reliable, and flexible interface between cameras and computers for high-performance image processing applications.

CameraLink consists of three components: a camera interface, a cable, and a frame grabber interface. The camera interface is located on the camera and includes a set of pins that transmit image data, control signals, and timing information. The cable is a high-speed, shielded twisted-pair cable that connects the camera to the frame grabber. The frame grabber interface is located on the computer and receives image data from the camera.

There are three main types of CameraLink interfaces: Base, Medium, and Full. The bandwidth provided by CameraLink allows for high-resolution images and high frame rates to be transmitted from the camera to the computer.

Camera Link also supports different configurations of the camera interface, such as the number of taps, bit depth, and pixel format. The number of taps refers to the number of parallel data channels used to transmit the image data. The bit depth refers to the number of bits used to represent each pixel in the image, and the pixel format refers to the order of the color components in the image data.

CameraLink has several advantages over other digital communication standards in machine vision applications, such as FireWire and USB. It provides a higher data rate, lower latency, and more reliable communication between the camera and the computer. Additionally, CameraLink is a mature standard that is widely used in the industry and has a large ecosystem of compatible products.

In summary, CameraLink is a digital communication standard used in machine vision applications to transfer image data from a camera to a computer. It provides high-speed, reliable, and flexible communication between the camera and the computer and is widely used in high-performance image processing applications.

GigE Vision

GigE Vision is a communication standard for machine vision applications that uses Gigabit Ethernet networks to transmit image and control data from a camera to a computer. The standard was developed by the Automated Imaging Association (AIA) and first introduced in 2006. GigE Vision is designed to provide a high-speed, low-cost, and reliable interface between cameras and computers for machine vision applications.

GigE Vision uses standard Ethernet cabling and network infrastructure to transmit image and control data from the camera to the computer. It supports data rates of up to 10 Gbps over a single Ethernet cable and can transmit data over distances of up to 100 meters. GigE Vision also supports power-over-Ethernet, which allows cameras to be powered directly from the GigE Vision interface.

GigE Vision includes several features that make it well-suited for machine vision applications. These features include:

• Standard Ethernet infrastructure: GigE Vision uses standard Ethernet cabling and network infrastructure, which is readily available and easy to set up. This makes it a cost-effective solution for machine vision applications.
• Large data bandwidth: GigE Vision provides a high data bandwidth, which allows for high-resolution images and high frame rates to be transmitted from the camera to the computer.
• Real-time communication: GigE Vision provides low-latency communication between the camera and the computer, which is essential for real-time image processing applications.
• Bidirectional communication: GigE Vision supports bidirectional communication between the camera and the computer. This allows for control signals to be transmitted from the computer to the camera, enabling advanced features such as trigger control and synchronization.
• Multi-camera support: GigE Vision supports multiple cameras to be connected to a single interface, which reduces the number of cables required.
• GigE Vision has several advantages over other communication standards in machine vision applications, such as CameraLink and FireWire. It provides a lower cost solution, supports longer cable lengths, and can use existing network infrastructure. Additionally, GigE Vision is a mature standard that is widely used in the industry and has a large ecosystem of compatible products.

In summary, GigE Vision is a communication standard for machine vision applications that uses Gigabit Ethernet networks to transmit image and control data from a camera to a computer. It provides a high-speed, low-cost, and reliable interface between cameras and computers, with a large data bandwidth, real-time communication, and bidirectional communication. GigE Vision is widely used in the industry and is well-suited for machine vision applications that require high-resolution images and real-time processing.

USB3 Vision

USB3 Vision is a communication standard for machine vision applications that uses USB 3.0 ports to transmit image and control data from a camera to a computer. The standard was developed by the Automated Imaging Association (AIA) and first introduced in 2013. USB3 Vision is designed to provide a high-speed, low-cost, and easy-to-use interface between cameras and computers for machine vision applications.

USB3 Vision uses USB 3.0 ports to transmit image and control data from the camera to the computer. It supports data rates of up to 5 Gbps over a single USB cable and can transmit data over distances of up to 5 meters. USB3 Vision also supports power-over-USB, which allows cameras to be powered directly from the USB3 Vision interface.

USB3 Vision includes several features that make it well-suited for machine vision applications. These features include:

• Plug-and-play operation: USB3 Vision provides plug-and-play operation, which means that the camera can be automatically detected and configured by the computer. This makes it easy to set up and use.
• Standard USB infrastructure: USB3 Vision uses standard USB cabling and infrastructure, which is readily available and easy to set up. This makes it a cost-effective solution for machine vision applications.
• Large data bandwidth: USB3 Vision provides a high data bandwidth, which allows for high-resolution images and high frame rates to be transmitted from the camera to the computer.
• Real-time communication: USB3 Vision provides low-latency communication between the camera and the computer, which is essential for real-time image processing applications.
• Bidirectional communication: USB3 Vision supports bidirectional communication between the camera and the computer. This allows for control signals to be transmitted from the computer to the camera, enabling advanced features such as trigger control and synchronization.
• Multi-camera support: USB3 Vision supports multiple cameras to be connected to a single interface, which reduces the number of cables required.
• USB3 Vision has several advantages over other communication standards in machine vision applications, such as CameraLink and GigE Vision. It provides a lower cost solution, supports longer cable lengths than CameraLink, and is easier to set up than GigE Vision. Additionally, USB3 Vision is a mature standard that is widely used in the industry and has a large ecosystem of compatible products.

In summary, USB3 Vision is a communication standard for machine vision applications that uses USB 3.0 ports to transmit image and control data from a camera to a computer. It provides a high-speed, low-cost, and easy-to-use interface between cameras and computers, with plug-and-play operation, a large data bandwidth, real-time communication, and bidirectional communication. USB3 Vision is widely used in the industry and is well-suited for machine vision applications that require high-resolution images and real-time processing.

GenICam

GenICam (Generic Interface for Cameras) is a software standard that provides a generic, application-independent interface for machine vision cameras. It was developed by the European Machine Vision Association (EMVA) and is now maintained by the Automated Imaging Association (AIA). The goal of GenICam is to simplify the integration of cameras into software applications by providing a common interface that can be used with different types of cameras and different software development environments.

GenICam defines a set of standard features and functions for cameras, such as image resolution, frame rate, exposure time, gain control, and image format. These features are represented by a set of registers and control commands that can be accessed through the camera interface. GenICam also defines a set of standard programming interfaces (APIs) that can be used to access these features from software applications.

The benefits of GenICam include:

• Compatibility: GenICam provides a common interface for different types of cameras, which simplifies the integration of cameras into software applications. This allows users to choose the best camera for their application without worrying about software compatibility issues.
• Flexibility: GenICam provides a flexible interface that can be customized to meet the needs of different applications. This allows users to create custom software applications that take advantage of the features and functions of their cameras.
• Standardization: GenICam provides a standardized interface for machine vision cameras that is widely adopted in the industry. This makes it easier for camera and software vendors to develop and support their products.
• Interoperability: GenICam ensures interoperability between different camera vendors and software applications. This allows users to mix and match cameras and software applications from different vendors without worrying about compatibility issues.
• GenICam is typically implemented through a software development kit (SDK) provided by camera vendors. The SDK includes the GenICam-compliant camera driver, which provides access to the camera features and functions through the GenICam interface. The SDK also includes example code and documentation to help users develop their own software applications.

In summary, GenICam is a software standard that provides a common, generic interface for machine vision cameras. It simplifies the integration of cameras into software applications by providing a set of standard features and functions that can be accessed through a common interface. GenICam provides compatibility, flexibility, standardization, and interoperability, and is widely adopted in the industry.

EMVA 1288: 

EMVA 1288 is a standard for characterizing the performance of machine vision cameras. The standard provides a framework for measuring camera parameters such as sensitivity, dynamic range, and noise, and helps users to compare the performance of different cameras.

These are just a few examples of the standards available in machine vision products. The choice of standard depends on the specific application requirements, such as data transfer speed, cable distance, and compatibility with other equipment.