Image Acquisition and Machine Vision: My First Vision System

"Ah, a little Satori for me today after consuming some Roborant."

Why is a frame grabber required?

Maximum performance (to avoid loosing frames)
High bandwidth cameras
For integral preprocessing
For on board processing – to reduce CPU load and also conduct processing at a higher speed
Food or fruit inspection
Line synchronization for web inspection
For certain applications such as web inspection of paper, wood, steel, textiles, etc.

Those of you novice to the field of machine vision or image acquisition systems, frame grabber is definitely a new term. The reason behind this is because in commercial photography the terms cameras, lens, memory card and USB cable are commonly used. Consumer cameras like Canon, Nikon, Sony, etc. do not require a specialized hardware to transfer images from cameras to the computer. Photos or images from a consumer cameras can be transferred easily using a USB cable or inserting the memory card directly into the SD port of the laptop, desktop or printer.

When I first heard the term “frame grabber”, like most of us, I Googled it to understand what it exactly meant. Even though a lot of documents are written on frame grabbers due to the marketing competition, except Wikipedia information, no other information is available. I guess a lot of these documents are submerged in the latter pages of the Google results pages.

This document provides a brief yet good understanding of frame grabbers. After reading this you will understand what a frame grabber is, the different types of frame grabbers, and the advantages of using a frame grabber.

Frame grabbers may be defined as an interface between an industrial camera (analog or digital) and a computer or laptop.

A frame grabber can also be considered as an electronic device that captures individual still frames from an analog video signal or a digital video stream.

The important functionalities of a frame grabber include:

Real time image compression to MPEG2, H.264, JPEG or other related formats
Temporarily stores multiple images on the “on board memory” of the frame grabber

Advanced frame grabbers can do on board processing, thereby reducing the CPU usage.

Frame grabbers can also be classified based on the camera interface:

Analog frame grabber which includes NTSC/PAL/RS170
Camera link frame grabber which includes base, medium and full
GigE Vision frame grabber
IEEE-1394 frame grabber

Advantages of using a frame grabber

On Board Compression: Uncompressed raw images from the camera can be compressed on the frame grabber before being transferred to the host PC either for saving or processing.
On Board Processing: Advanced frame grabbers have the capability of processing images on board. This reduces the usage of the CPU memory. In many applications a single computer does both image processing and motion control therefore onboard processing is an added advantage.
GigE Vision Frame Grabber: Most of the computers have only an Ethernet port. Many applications require the sue of more than one camera. In these cases GigE Vision frame grabbers play a critical role, in addition, using GigE Vision frame grabbers reduces bottle neck bandwidth at the Ethernet port. Bottle neck bandwidth results in frames being lost.
Synchronization: if two or more cameras need to be synchronized for a particular application, then a frame grabber would be necessary.
Triggering: Frame grabbers through its auxiliary IO can be used to trigger the camera. This is referred to as “Hardware Triggering.” Even though software triggering is possible. Hardware triggering is precise.

If you have any questions about frame grabbers or whether or not your application can benefit from using one, please leave a comment or call 510 657 4000.