FAQ topic08 / 12

Vision, Video and Capture FAQs

Vision and video systems capture, encode and analyse camera feeds at the edge. These FAQs explain capture cards and frame grabbers, SDI, codecs and bitrate, camera interfaces such as GMSL, machine vision and video analytics.

Search this topic

43 answers Search all FAQs

Topics 12

VIDEO · COMPRESSION AND STREAMING

What Is Bitrate?

Bitrate is the amount of data processed or sent per second, measured in bits per second (kb/s, Mb/s or Gb/s). For video and audio it sets quality, file size and the bandwidth a stream needs. A higher bitrate keeps more detail with the same codec, while a newer codec such as H.265 gives similar quality at a lower bitrate.

6 MIN

VIDEO · COMPRESSION AND ENCODING

What Is a Codec?

A codec (coder-decoder) is a method, in software or hardware, for compressing audio or video so it takes less space and bandwidth, then decompressing it for playback or analysis. Common video codecs are H.264, H.265, AV1 and MJPEG. The encoder compresses at the source and a matching decoder restores the picture at the other end.

6 MIN

VIDEO · CAPTURE

What Is a Capture Card?

A capture card is a device that takes a video signal from a camera or other source, such as HDMI or SDI, and converts it into digital frames a computer can record, stream or analyse. It can be an internal PCIe or M.2 card or an external USB device, and industrial versions feed medical imaging and edge AI systems.

6 MIN

MACHINE VISION · BASICS

What Is Machine Vision?

Machine vision is the use of cameras, lighting and a computer to inspect, measure, identify or guide objects automatically. In industry it checks products for defects, reads codes, verifies assembly and guides robots, faster and more consistently than a person can, and today it increasingly runs AI models on edge computers beside the line.

7 MIN

NETWORKING · TIME SYNCHRONISATION

What Is Precision Time Protocol (PTP)?

Precision Time Protocol (PTP), defined in IEEE 1588, synchronises the clocks of devices on a network to a single reference clock, called the grandmaster. With hardware timestamping in the network interfaces and switches, PTP typically aligns clocks to within a microsecond or better, far tighter than NTP, which is why it is used in industrial automation, substations, broadcast and telecoms.

7 MIN

ROBOTICS · EDGE AI

What Is an Autonomous Mobile Robot?

An autonomous mobile robot (AMR) is a robot that moves through a building or site on its own. It builds a map with lidar and cameras, works out where it is, plans a route and steers around people and obstacles, so it needs no tape, wires or fixed tracks. AMRs move goods, tow carts, clean, inspect and deliver.

6 MIN

NETWORKING · ETHERNET PERFORMANCE

What Are Jumbo Frames?

Jumbo frames are Ethernet frames that carry more than the standard 1,500 bytes of payload, usually up to 9,000 bytes. Fewer, larger frames cut header overhead and the number of packets a computer has to process, which helps bulk transfers such as GigE Vision cameras and storage. Every device on the path must support the same size.

6 MIN

NETWORKING · MULTICAST

What Is IGMP Snooping?

IGMP snooping is a switch feature that listens to the IGMP messages hosts send when they join or leave a multicast group. The switch records which ports want each group and forwards multicast traffic only to those ports, instead of flooding every port like a broadcast. It keeps multicast video and control traffic off links that do not need it.

6 MIN

EDGE AI · NVIDIA SOFTWARE

What Is NVIDIA Holoscan?

NVIDIA Holoscan is NVIDIA’s sensor-processing platform for real-time AI on streaming data at the edge. It combines the open-source Holoscan SDK, which builds low-latency pipelines from capture to inference to display, with the Holoscan Sensor Bridge for bringing camera and sensor data over Ethernet straight into GPU memory, on Jetson, IGX and x86 GPU systems.

6 MIN

VISION · CAMERA INTERFACES

What Is MIPI CSI-2?

MIPI CSI-2 (Camera Serial Interface 2) is the standard high-speed link between a camera’s image sensor and a processor, defined by the MIPI Alliance. It carries image data over one to four differential lanes on a short flex cable, and it is the native camera input on NVIDIA Jetson, NXP i.MX, Raspberry Pi and most smartphone processors.

6 MIN

MACHINE VISION · QUALITY CONTROL

What Is Automated Visual Inspection?

Automated visual inspection (AVI) uses cameras, controlled lighting and software to check products for defects without a human inspector. Each part is imaged on the line, the software compares it with what a good part should look like, and faulty items are rejected and logged. It is faster and more consistent than manual checks.

7 MIN

VISION · SOFTWARE

What Is OpenCV?

OpenCV (Open Source Computer Vision Library) is a free, open-source library of more than 2,500 algorithms for image processing and computer vision. Developers call it from C++ or Python to read camera feeds, clean up and measure images, find features and run trained neural networks, on anything from a laptop to an NVIDIA Jetson.

6 MIN

VIDEO · CODECS

What Is H.266 (VVC)?

H.266, also called Versatile Video Coding (VVC), is the video compression standard ITU-T and ISO/IEC published in 2020 as the successor to H.265. It aims for the same picture quality at around half the bitrate of H.265, at the cost of much heavier encoding and, for now, limited hardware support.

6 MIN

DEFENCE · COMMAND AND CONTROL

What Is C4ISR?

C4ISR stands for command, control, communications, computers, intelligence, surveillance and reconnaissance. It is the military term for the networked systems that gather information from sensors, turn it into a shared picture and pass decisions to forces. C5ISR adds cyber, and UK forces often use C4ISTAR, which adds target acquisition.

7 MIN

CONNECTORS · AUTOMOTIVE RF AND CAMERAS

What Is a FAKRA Connector?

A FAKRA connector is an automotive coaxial connector: an SMB-type 50 ohm RF contact inside a colour-coded, keyed plastic housing that clicks and locks. Standardised in ISO 20860 and USCAR-18, it carries antenna signals such as GPS, cellular and radio, and is the usual connector on GMSL cameras for edge AI and vehicle vision systems.

6 MIN

EDGE AI · TRAINING DATA

What Is Synthetic Data Generation?

Synthetic data generation is the creation of artificial training data, such as rendered images, simulated sensor readings or generated text, that mimics real data. It fills gaps where real data is scarce, costly to label, private or unsafe to collect, and it comes with exact labels. Most vision and robotics teams combine it with real data.

6 MIN

EDGE AI · VISION

What Is Computer Vision?

Computer vision is the branch of AI that lets computers interpret images and video: naming what is in a frame, locating objects, outlining them pixel by pixel and following them over time. Modern systems use deep learning models such as CNNs and vision transformers, and many run on edge computers next to the cameras they read.

6 MIN

EDGE AI · GENERATIVE AI

What Is a Vision Language Model (VLM)?

A vision language model (VLM) is an AI model that takes images or video together with a text prompt and answers in text. It joins a vision encoder to a large language model, so you can ask about a scene, caption or summarise footage, read documents or search video in plain language instead of training a fixed-class detector.

6 MIN

EDGE AI · DEEP LEARNING

What Is a Convolutional Neural Network (CNN)?

A convolutional neural network (CNN) is a deep learning model designed for grid-shaped data such as images. Its layers slide small learned filters across the input to pick out edges, textures and shapes, then combine them to classify an image, detect objects or label pixels. CNNs remain the most widely deployed vision models on edge hardware.

6 MIN

DISPLAYS · RESOLUTION

What Is 8K Resolution?

8K resolution is 7680 × 4320 pixels at 16:9, also called 8K UHD or 4320p. That is about 33 million pixels: four times 4K and 16 times 1080p. Carrying 8K at 60 Hz needs HDMI 2.1 or DisplayPort with compression, or DisplayPort 2.1, plus graphics, storage and capture hardware rated for it.

7 MIN

DISPLAYS · RESOLUTION

What Is 4K Resolution?

4K resolution is a picture about 4,000 pixels wide. On monitors, TVs and cameras it almost always means 3840 × 2160 pixels at 16:9, also called 4K UHD or 2160p. That is about 8.3 million pixels, four times full HD (1080p). Cinema 4K is slightly wider, at 4096 × 2160.

7 MIN

VIDEO · COMPRESSION

H.264 vs H.265: What Is the Difference?

H.265, also called HEVC, is the newer video compression standard that followed H.264, or AVC. It can deliver similar picture quality at up to about half the bitrate, which cuts network traffic and storage per camera. The trade-off is more processing to encode and decode, and less universal playback support than H.264.

6 MIN

VIDEO · SURVEILLANCE

What Is ONVIF?

ONVIF is an open industry standard that lets IP cameras, network video recorders and video management software from different manufacturers work together. A camera and an NVR that support the same ONVIF profile can find each other on the network, share live video and recordings, and exchange settings without a custom driver for each brand.

6 MIN

MARITIME · SURVEILLANCE

What Is Maritime Domain Awareness?

Maritime domain awareness is the understanding of activity at sea that could affect security, safety, the economy or the environment. It combines data from radar, AIS ship transponders, cameras, satellites and patrol vessels into one picture, so authorities and ship crews can spot unusual behaviour and respond early.

6 MIN

VIDEO ANALYTICS · BASICS

What Is People Counting?

People counting uses sensors, most often overhead cameras with AI, to count how many people pass through entrances and exits. Subtracting exits from entrances gives live occupancy, so operators know how busy a space is at any moment. Stations, shops, offices and venues use it for safety limits, staffing and planning.

6 MIN

VIDEO ANALYTICS · SECURITY

What Is a Virtual Fence in Video Analytics?

In video analytics, a virtual fence is a line or zone drawn on a camera’s view that raises an alert when a person or vehicle crosses or enters it. It is also called line crossing detection or a tripwire. It can watch one direction only, and AI classification means it reacts to people and vehicles, not shadows or rain.

6 MIN

SURVEILLANCE · BASICS

What Is an NVR?

An NVR (network video recorder) is a recorder that receives video streams from IP cameras over a network, stores them, and lets users watch live and recorded footage locally or remotely. PoE NVRs also power the cameras. An AI NVR adds a GPU or AI accelerator so it can analyse the video, detecting people, objects and events as it records.

6 MIN

SURVEILLANCE · COMPARISON

NVR vs DVR: What Is the Difference?

A DVR (digital video recorder) records analogue cameras connected by coaxial cable, converting their video to digital inside the recorder. An NVR (network video recorder) records IP cameras over a network, with each camera encoding its own video. NVRs support higher resolutions and flexible camera placement, and PoE NVRs power cameras over the same cable.

6 MIN

VIDEO · COMPARISON

SDI vs HDMI: What Is the Difference?

SDI and HDMI both carry uncompressed digital video, but for different settings. SDI runs over coaxial cable with locking BNC connectors and covers long distances, around 100 m for 3G-SDI, making it the standard in broadcast, defence and industry. HDMI is a consumer interface with short cables and push-fit connectors, common on monitors, PCs and cameras.

6 MIN

VIDEO ANALYTICS · COMPARISON

ANPR vs LPR: What Is the Difference?

ANPR and LPR are two names for the same technology: reading vehicle plates automatically from camera images. ANPR (automatic number plate recognition) is the term used in the UK, Europe and Australia, while LPR (licence plate recognition) or ALPR is used in North America. The cameras, software and uses are essentially the same.

5 MIN

VIDEO ANALYTICS · BASICS

What Is ANPR?

ANPR (automatic number plate recognition) is technology that reads vehicle number plates from camera images. A camera, usually with infrared lighting for night use, captures the vehicle; software finds the plate in the image and reads its characters; and the system checks the result against a list, such as permitted vehicles, payments or alerts.

6 MIN

AI VISION · BASICS

What Is Video Analytics?

Video analytics is software that analyses video automatically to detect, classify and track people, vehicles and events, so operators do not have to watch every screen. Modern systems use AI models and often run at the edge, on a computer beside the cameras, to react in real time and avoid sending every stream to the cloud.

6 MIN

AI VISION · BASICS

What Is Object Tracking?

Object tracking is a computer vision task that follows an object from one video frame to the next, keeping track of where it is and where it is going. Detection finds objects in each frame; tracking links those detections over time, so a system can follow a person, vehicle or drone and act on its movement.

6 MIN

VIDEO · CAMERAS

What Is AHD?

AHD (Analog High Definition) is a CCTV standard that sends high-definition video over coaxial cable as an analogue signal. It lets existing coax and BNC cabling carry HD cameras, with long cable runs and very low latency, because the video is not compressed or sent over a network. HD-TVI and HD-CVI are competing analogue HD formats.

6 MIN

SENSORS · PERCEPTION

LiDAR vs Camera: What Is the Difference?

LiDAR measures distance directly with laser pulses, giving accurate 3D depth in any light. A camera captures colour, texture and fine detail at low cost, but needs light and has to work out depth indirectly, through stereo vision or AI. Many autonomous machines use both, combining the camera’s detail with LiDAR’s depth.

6 MIN

MACHINE VISION · CAMERAS

What Is a GigE Vision Camera?

A GigE Vision camera is an industrial camera that sends images and takes commands over standard Ethernet, following the GigE Vision interface standard. It runs over long, low-cost network cables, can be powered by PoE, and lets many cameras share one network through a switch, which makes it a common choice for machine vision on lines and vehicles.

6 MIN

SAFETY · AI VISION

What Is a Pedestrian Detection System?

A pedestrian detection system uses sensors on a vehicle, such as AI cameras, radar, LiDAR or worker-worn tags, to spot people nearby and warn the operator or slow and stop the vehicle. In cars it supports automatic emergency braking. On forklifts and AGVs it reduces collisions in busy warehouses and factories.

6 MIN

CAMERAS · INTERFACES

What Is GMSL2?

GMSL2 (Gigabit Multimedia Serial Link 2) is a high-speed serial link that carries uncompressed camera video, power and control data over one coaxial cable of up to around 15 metres. Developed for automotive cameras, it is now widely used on robots, autonomous machines and drones with NVIDIA Jetson computers.

6 MIN

SURVEILLANCE · CAMERAS

What Is a PTZ Camera?

A PTZ (pan-tilt-zoom) camera has motors that turn it left and right, tilt it up and down and zoom its lens, so one camera can cover a wide area and close in on detail. PTZ cameras are used in surveillance, security and broadcast, and can follow targets automatically when paired with AI.

6 MIN

RETAIL · DATA

What Is Retail Analytics?

Retail analytics is collecting and analysing data from a retail business, such as sales, footfall, dwell time, queues and stock levels, to make better decisions about layout, staffing, promotions and inventory. In physical stores, edge AI on camera feeds now adds live insight into how shoppers move, wait and buy.

6 MIN

DEFENCE · SENSORS

What Is an EO/IR Camera?

An EO/IR camera combines electro-optical (visible light) and infrared (thermal) imaging in one stabilised payload, so an operator can see by day and at night. EO/IR gimbals are standard on ISR drones, patrol aircraft, border security and maritime systems, and their video increasingly feeds onboard AI computers for detection and tracking.

6 MIN

DEFENCE · ISR

What Does ISR Mean in the Military?

In the military, ISR stands for intelligence, surveillance and reconnaissance: the collection, processing and analysis of information about an area or target to support decisions. It is carried out by satellites, aircraft, drones, ships and ground sensors, and increasingly processed on board with AI so that commanders get an accurate picture faster.

6 MIN

MACHINE VISION · STEREO CAMERAS

What Is a Stereo Camera?

A stereo camera is a camera with two image sensors a fixed distance apart, like a pair of eyes. By finding the same point in both images and measuring how far it shifts between them, it calculates the distance to every pixel, so one device gives a robot or vehicle a 3D depth map as well as a colour image.

6 MIN

Products

Shop this topic

Products for vision, video and capture

All video capture cards

CategoryVideo Capture CardsCapture cards for HDMI, SDI and analogue video sources.View products

CategoryFrame GrabbersFrame grabbers for GMSL, USB and GigE machine vision cameras.View products

CategoryIndustrial NVRRecorders for IP cameras in transport, security and industry.View products

CategoryVideo Encoders and DecodersEncoders and decoders that move video over IP networks.View products