Glossary of Machine Vision and Vision Systems Terms
1D - One dimensional.
2D - Two dimensional.
3D - Three dimensional.
Aperture - The size of the opening in the lens. A wide open aperture will let more light into the image for a brighter photo, while a smaller aperture lets in less light. Aperture is measured in f-stops; a small f-stop like f/1.4 is a wide opening, a large f-stop like f/22 is a very narrow one. Aperture is one of three camera settings that determine an image’s exposure, or how light or dark it is. Aperture also affects how much of the image is in focus - wide apertures result in that creamy, unfocused background while narrow apertures keep more of the image sharp.
Area Scan Camera - A solid state imaging device with both rows and columns of pixels, forming an array which produces a 2D image.
Bayer Conversion - Conversion of Bayer color (obtained from a Bayer matrix or color filter array) into RGB color.
Bitmap - A raster graphics image, digital image, or bitmap, is a data file or structure representing a generally rectangular grid of pixels, or points of color, on a computer monitor, paper, or other display device. Black and white require one bit, while high definition color up to 32 bits.
C-Mount - Standardized adapter for optical lenses on industrial cameras. C-Mount lenses have a back focal distance 17.5 mm vs. 12.5 mm for CS-mount lenses. A C-Mount lens can be used on a CS-Mount camera through the use of a 5 mm extension adapter. C-mount is a 1" diameter, 32 threads per inch mounting thread (1"-32UN-2A).
Camera - Imaging device, a device that acquire images.
Camera Link - A robust communications link using a dedicated cable connection and a standardized communications protocol introduced by the Automated Imaging Association or AIA, the global machine vision industry's trade group. Camera Link is a serial communication protocol designed for computer vision applications based on the National Semiconductor interface Channel-link. It was designed for the purpose of standardizing scientific and industrial video products including cameras, cables and frame grabbers.
CCD - Charge-Coupled Device - A light-sensitive chip or image sensor used in scanners and digital cameras that converts light into proportional (analog) electrical current.
CMOS - Complementary Metal Oxide Semiconductor - A new type of sensor used in scanners and digital cameras that is based upon a semiconductor process designed for digital electronics instead of analog electronics as in the CCD.
CoaXPress (CXP) - CoaXPress is an asymmetric high speed serial communication standard over coaxial cable. CoaXPress combines high speed image data, low speed camera control and power over a single coaxial cable. The standard is maintained by JIIA, the Japan Industrial Imaging Association.
Contrast - In visual perception, contrast is the difference in visual properties that makes an object (or its representation in an image) distinguishable from other objects and the background.
CS-Mount - Standardized adapter for optical lenses on industrial cameras. Same as C-Mount but the focal point is 5 mm shorter (12.5 mm vs. 17.5 mm for C-mount lenses). A CS-Mount lens will not work on a C-Mount camera. CS-mount is a 1" diameter, 32 threads per inch mounting thread (1"-32UN-2A).
Deep learning - A class of machine learning algorithms that:
DOF - Depth of field - The distance in front of and behind the object (subject) which appears to be in acceptable focus.
DSP - Digital Signal Processor – A specialized digital microprocessor used to efficiently and rapidly perform calculations on digitized signals that were originally analog in form.
Dynamic range - Indicates the ability of a camera to reproduce the brightest and darkest portions of an image. It illustrates the brightest and darkest areas of a scene that can be captured in a single image and still be read. It is calculated as logarithmic ratio between the camera’s full well capacity and its dark noise.
EMVA 1288 - Standard introduced by European Machine Vision Association (EMVA). It defines a unified method to measure, compute and present specification parameters and characterization data for cameras and image sensors used for machine vision applications.
Firewire - see IEEE 1394
Focal length - The distance from a lens' principal point to the corresponding focal point on the object.
FOV - Field of View - The amount of area that can be seen by a camera at one time. It is a result of the size of the image sensor, the lens of the system, and the working distance between object and camera.
FPGA - Field Programmable Gate Array - A specially made digital programmable semiconductor. With an FPGA, a design engineer is able to program electrical connections on site for a specific application.
Frame grabber - An electronic device that captures individual, digital still frames from an analog video signal or a digital video stream.
Full Well Capacity - The maximum amount of electrons that can be stored in each basic sensor element (pixel) in a digital CCD or CMOS camera imaging sensor, where incident photons are recorded as electric charge (electrons).
GenICam - Standard introduced in 2006 by the European Machine Vision Association (EMVA). GenICam
is a universal configuration interface across a wide range of standard inustrial camera interfaces, such as
USB3 Vision, GigE Vision, Camera Link, CoaXPress and IEEE 1394 IIDC, regardless of the camera type and image
format. GenICam standardizes the user application programming interface by defining how to configure the
camera, grab images, send extra data, and perform events. It also includes a standard feature naming convention
(SFNC) for camera features. The core of the GenICam standard is the XML Descriptor File inside the camera, which
maps the cameras internal registers to a standardized function list. This standard consists of 3 main standards:
GigE Vision - Industrial camera communication protocol introduced in 2006 by the Automated Imaging Association
(AIA). The standard works to unify integration and communication protocols over Ethernet between cameras, hardware
components, and software packages. For example, 3rd party software that is GigE Vision compliant will work with any
camera that is also GigE Vision compliant. The GigE Vision contains 4 components:
Global Shutter - A global shutter is analogous to a mechanical shutter, in that all pixels are exposed and sampled simultaneously, with the readout then occurring sequentially; the photon acquisition starts and stops at the same time for all pixels.
Histogram - A function plotting the frequency of occurrence of an intensity value as a function of those intensity values. As such, a histogram illustrates the distribution of intensity values in a given region of interest in an image.
IEEE 1394 - A high-speed data protocol involving an external bus capable of throughput up to 1.6 Gb/s (3.2 Gb/s) in version B with 8/10b coding and control of up to 63 devices. Also know as "FireWire" and was initially launched by Apple Computer. This standard was implemented to use for digital output of cameras.
Image Processing - The transformation of an input image into an output image with desired properties.
JPEG - JPG - A method of compressing an image to reduce file size. JPEG is an acronym for the Joint Photographic Experts Group, which created the standard for this compression method. The degree of compression can be adjusted, allowing a selectable tradeoff between storage size and image quality. JPEG typically achieves 10:1 compression with little perceptible loss in image quality.
LED - Light Emitting Diode - A special type of semiconductor diode that emits incoherent narrow-spectrum light.
Line Scan Camera - Camera that uses sensors that consist of a single or few rows of photodectors, which gathers an image on a line-by-line basis. It requires that either the object is moving or the camera is moving.
Machine vision - MV - The technology and methods used to provide imaging-based automatic inspection and analysis for such applications as automatic inspection, process control, and robot guidance, usually in industry. Machine vision is a term encompassing a large number of technologies, software and hardware products, integrated systems, actions, methods and expertise. Machine vision as a systems engineering discipline can be considered distinct from computer vision, a form of computer science. It attempts to integrate existing technologies in new ways and apply them to solve real world problems. The term is the prevalent one for these functions in industrial automation environments but is also used for these functions in other environments such as security and vehicle guidance.
MOD - Minimum Object Distance - The closest focusing distance of a lens. This does not mean that a lens at shorter working distances no longer works. Focus on shorter working distances can be achieved by increasing the distance between sensor and lens (= screw type design of the lens and possible use of distance rings)
MTF - Modulation Transfer Function – The ability of a lens or optical system to reproduce (transfer) various levels of detail (modulation) of an object to the image as the frequency (usually sinusoidal) increases.
Neural network - An interconnected group of artificial neurons that uses a mathematical or computational model for information processing based on a connectionist approach to computation. In most cases a neural network is an adaptive system that changes its structure based on external or internal information that flows through the network.
OCR - Optical Character Recognition - The process by which a string of unknown characters is digitally imaged and converted into machine-encoded text by a machine vision system.
Pattern Recognition – A process which identifies an object based on analysis of its features.
Pixel – An acronym for "picture element" (i.e. an individual element of a digitized image array). The smallest distinguishable and resolvable area in an image or CCD and CMOS imager. The discrete location of an individual photo-sensor in a solid state camera.
QE - Quantum Efficiency - A quantity parameter defined for a photosensitive device such as photographic film or a charge-coupled device (CCD) which is the percentage of photons hitting the photoreactive surface that will produce an electron–hole pair. It is a key measurement of the device's electrical sensitivity to light.
Resolution - as feature – The smallest object or feature in an image which may be sensed.
Resolution - as image resolution - The number of rows and columns of pixels in an image.
Resolution - as measurement resolution - The smallest movement measurable by a vision system.
Rolling Shutter - A rolling shutter exposes, samples, and reads out sequentially; it implies that each line of the image is sampled at a slightly different time. Intuitively, images of moving objects are distorted by a rolling shutter. This effect can be minimized with a triggered strobe placed at the point in time where the integration period of the lines overlaps.
Smart Camera - A complete or near complete vision system contained in the camera body itself. Lighting and optics may or may not be included. At a minimum a Smart Camera combines a camera with image processing and MV related programs within the same housing. A smart camera is functionally equivalent to an Embedded Vision Processor. Sometimes smart cameras are called "intelligent cameras" and "vision sensors". The term "vision sensor" tends to apply to a lower-end Smart Camera.
SNR - Signal-To-Noise Ratio - A physical measure of the sensitivity of a digital imaging system. CCD and CMOS cameras produce several forms of noise as variations in pixel charges not generated by the light striking the imager. These can be caused by thermal conditions, electronics, or simply the fundamental physical laws of how photons are converted to electrons. Noise can appear as random graininess, horizontal or vertical lines that become visible in low signal areas of the image, blotchy gradients between darker and lighter regions, and other manifestations. The signal-to-noise ratio is a measure of how much a typical image is corrupted by these noise sources. It is generally expressed in decibels (dB) – the higher number means the cleaner image.
Telecentric Lenses - Parallax corrective lenses maintaining within a certain range of working distance a constant viewing angle at any point across the clear aperture of the objective lens, thus allowing the machine vision system to generate dimensionally accurate images for measurement.
USB3 Vision - A communication interface for vision applications based on the USB 3.1 technology. All cameras supporting USB3 Vision interact the same way with software also supporting USB3 Vision. The USB3 Vision standard defines required elements for camera identification, control, and output. It uses GenICam, a programming interface for camera attribute control. GenICam allows camera vendors to define features and attributes in an XML file stored inside the camera. The file is parsed by the host application when the camera is initially discovered. One of the key benefits of GenICam is the ability for camera vendors to introduce new camera-specific features without needing to update the host application. Each camera attribute, such as exposure time, is controlled by a specific GenICam feature. Most USB3 Vision applications provide a Graphical User Interface (GUI) to control these attributes.
Vision Sensor - A lower-end smart camera. A smart camera with less flexibility and programmability that is usually intended for less demanding applications.
WD - Working distance - The distance between the front element of a lens and the object (subject) when it is in focus.
Zoom Lens - A mechanical assembly of lenses whose focal length can be changed, as opposed to a standard lens, which has a fixed focal length.