Basics of cameras

Industrial cameras are a key component of any image processing system. They capture the light coming through the lens and convert it into digital image data, which can then be analysed. Numerous factors, such as the sensor, resolution, frame rate and interface, determine how accurately and reliably a system operates. This overview explains the most important terms and functions relating to industrial cameras and helps you to better understand the right solution for different applications. Translated with DeepL.com (free version)

The camera is the "eye" of any industrial vision system. It converts incoming light into a digital electrical signal and provides the raw image for further processing. At its core, an industrial camera consists of an image sensor, signal processing electronics, and a housing with interfaces. The key technical characteristics of a camera include:

• Sensor types:
o Area scan sensor: Captures a complete image at once, ideal for most standard applications (e.g. inspection, positioning).
o Line scan sensor: Captures a single line per exposure, used for continuously moving objects (e.g. webs, paper, bottles on a conveyor).

• Sensor sizes: The physical size of the sensor (e.g. 1/2", 2/3", 1", 4/3" or APS-C) determines, together with the lens, the field of view and light sensitivity.

• Sensor architecture:
o Single-chip camera: One sensor with a color filter (Bayer pattern) for color imaging or monochrome (black and white).
o Three-chip camera (3-CCD / 3-CMOS): Three separate sensors for red, green, and blue, providing maximum color accuracy and sharpness (mainly used for demanding color inspection).

• Resolution: From VGA (0.3 megapixels) up to 100+ megapixels – the higher the resolution, the finer the details that can be detected.

• Color mode:
o Monochrome (B/W): Higher light sensitivity, better contrast, and faster frame rates.
o Color: Used for color inspection, sorting tasks, or when color is a quality criterion.

• Bandwidth / frame rate: The number of images the camera can deliver per second (fps = frames per second). Critical for high-speed applications.

• Interfaces: Common industrial interfaces include:
o GigE Vision (Gigabit Ethernet)
o USB3 Vision
o Camera Link / CoaXPress (very high bandwidth)
o HD-SDI / 3G-SDI (video transmission)

• Lens mounts: C-mount (most common), CS-mount, F-mount, M42, M58, etc. The mount defines mechanical compatibility with the lens.

Summary: The camera is much more than just a “digital camera.” Its selection largely determines the image quality, speed, and robustness of the entire vision system.

Terminology

Control

In machine vision, control refers to how camera or lens functions are managed. This can include exposure, gain, focus, or iris control. Control can be manual or automatic, depending on the system. Many industrial cameras allow control via software. Reliable control is essential for stable and repeatable imaging.

Frames per second (FPS)

Frames per second describes how many images a camera captures each second. Higher FPS allows faster motion to be captured clearly. Lower FPS may cause motion blur or missed events. FPS depends on sensor speed, interface, and resolution. High frame rates are important in applications such as high-speed inspection and motion analysis.

Shutter type

Shutter type defines how the sensor captures light. Global shutter captures the entire image at once. Rolling shutter captures the image line by line. Rolling shutter can cause distortion with moving objects. Global shutter is preferred for fast machine vision tasks.

Sensor size

The sensor size describes the physical dimensions of the sensor. It is often specified in inches (e.g. 1/2", 1"). Larger sensors capture more light. They influence the field of view and lens selection. The lens must be compatible with the sensor size.

Camera interfaces

Interfaces determine how image data is transmitted. GigE uses Ethernet and allows for long cable lengths. CoaXPress supports very high data rates. The interface impacts both speed and cost. Choosing the right one is crucial for overall system performance.

Gain

Gain is the electronic amplification of the image sensor signal. Increasing gain brightens the image but also increases noise, which can reduce image quality. It is typically used in low-light conditions or when short exposure times are needed. Proper adjustment balances brightness and noise.

Spectral response

It describes which wavelengths a sensor can detect. Some sensors are optimized for visible light, others for IR or SWIR. The lens and sensor must be spectrally matched. This affects sensitivity and image quality. It is especially important for multispectral applications.

Sensor type

Distinguishes between monochrome, color (Bayer), and three-chip sensors (3-CCD/3-CMOS). Monochrome sensors offer higher light sensitivity, while color sensors provide additional color information. The sensor type therefore influences image quality and application suitability.

Dynamic range

Indicates the range of brightness levels the camera can capture simultaneously. A high dynamic range allows both very bright and very dark areas to be clearly visible in the same image, which is especially important in high-contrast scenes.

Trigger modes

Defines how the camera can be triggered (free-run, hardware trigger, software trigger). They enable precise synchronization of image capture with external processes or motion.

Power over Ethernet (PoE)

Enables the camera to be powered directly via the network cable. This simplifies installation and reduces cabling effort, as no separate power supply is required.

Region of Interest (ROI)

Allows limiting the image output to a specific area of the sensor. This increases frame rate and reduces data volume, as only relevant image data is processed.

Pixel binning

Combines neighboring pixels into larger pixels. This increases light sensitivity and can improve frame rates, but reduces image resolution.

ADC bit depth

Describes the resolution of the analog-to-digital conversion (e.g. 8, 10, or 12 bit). Higher bit depths allow finer brightness gradations and improve detail recognition in images.

GenICam compliance

Standardized interface for camera control and image transmission. It enables easy integration into common machine vision software and ensures manufacturer-independent compatibility.

Form factor

Describes the size and design of the camera (e.g. compact, box, or board-level). It determines how easily the camera can be integrated into machines or confined spaces.

Software Development Kit (SDK) & software support

Describes the scope of included programming libraries and sample code. A well-designed SDK simplifies the integration of the camera into custom applications and reduces development time.

Digital inputs and outputs (I/O ports)

I/O ports are digital inputs and outputs used to connect sensors, lighting, or PLC controllers. They enable communication with external devices, for example to trigger the camera or to control processes.