MACHINE VISION AND ARTIFICIAL VISION

The goal of the computer vision is to extract informations from digital images. In this way it is a part of the instrumentation. The complex technologies used by machine vision are evolving continuously and become more and more affordable, and consequently relatively popular.

The images computing operated by these machines is used to measure and to analyse objects or actions with the objective to control them, providing the results to robots, programmable logic controllers ( PLC ) or any other type of controllers. The capabilities of the processors are increasing continuously, and the users express the need to use color processing, especially for sight control and look control of the final products and their packing.

The important amount of solutions available on the market of industrial machine vision has led some providers to standardize the exchanges of data between the vision systems and their environment. As the constraints imposed by some application in term of volume of informations to exchange and in term of response time, they have inclined manufacturers to propose some solutions to adapt networking to vision process.

This complete and very useful course - downloadable in « pdf » - carries on the fundamental knowledge about image processing and machine vision which needs to be mastered before designing an application. This document can be considered as a reference for these matters.

• Basics and concepts.

  ( Digital images, display, system setup and calibration )

• Image analysis and image processing.

  ( Histograms, line profile, intensity measurements, lookup table, spatial filtering and greyscale morphology )

• Operators.

  ( Arithmetic and logical )

• Particle analysis.
• Detection, measurement and control functions.

  ( Edge detection, pattern matching, geometric matching, dimensional measurements, colour inspection, optical character recognition )

National Instruments - www.ni.com

This teaching of almost 120 pages precisely describes the basic concepts and the fundamental knowledge to get in order to master digital processing of images as well. All the explanations are associated with many illustrations and with all their mathematical equations.

• The authors tackle the definitions of digital image, the visual perception, the image sampling, the noise, the cameras, the display, the basic algorithms, then they present various techniques and some tools.
• You will discover notions as diversified as filtering, photonic noise, signal phase and magnitude importance, contour representation, brightness sensitivity, sampling density, basic convolution computing, dilatation operations and so on …

Shortly, to get a good overview about the content, the better is still to quickly read this document from the beginning down to the bibliography.

Ian T. Young, Jan J. Gerbrands, Lucas J. van Vliet

( Delft university )

• Intelligent vision systems for Industry (   ~20 Mb )

  In complement of the National Instrument's document above, a so complete and very useful course about image processing and machine vision, especially related with the industry.

  It includes basic machine vision techniques, pattern recognition, intelligent image processing, lighting, multi-camera systems, controlling external devices ( robots, tools ), colour image recognition, examples of intelligent vision applications, proverbs and more.

  Bruce G. Batchelor et Paul F. Whelan

  ---

• « Proverbs, opinions and folklore » about image computing ( )

  « This is a list of observations, comments, suggestions, etc based upon our direct and our colleagues' experiences.

  It is offered in a light-hearted manner, but encapsulates some important lessons that we have learned but which are unfortunately not universally acknowledged.

  We hope it will bring enlightenment and promote discussion among our colleagues. »

  Bruce G. Batchelor and Paul F. Whelan

Vision Systems Laboratory

Dublin City University - www.dcu.ie

• Machine vision resources.
• Algorithms, Approximations and Heuristics. ( )

Pr. Bruce G. Batchelor

The « computer vision group » focuses on digital images acquisition and manipulation, particularly from a research and development point of view. It lists many resources on these subjects.

• Publications.
• General informations.

• Related links.
• Test images libraries.

• Demos.

The Computer Vision Group

This amazing website looks like an interactive museum joining science, technique and art.

• Causes of colour
  • Introduction : overview, mind, eye.
  • Light bulb, fire, lighting, aurora.
  • Water, plant, gold, emerald.
  • Rainbow, sky and sunset, butterfly, opal.
  • Colour blind, animal, theory, neurology.

  ---

• Colour art
  • Introduction.
  • Light, colour and vision.
  • Colour interactions: Simultaneous contrast.
  • Luminance and equiluminance.
  • Peripheral vision.

  ---

• History of Pigments through the ages and the civilizations.

www.webexhibits.org

This exhibit is a public service of the Institute for Dynamic Educational Advancement.

• FAQ, articles, links.
• Technical aspects of video.

Charles Poynton [ Archives ]

« The goal of GenICam is to provide a generic programming interface for all kinds of cameras. No matter what interface technology ( GigE Vision, Camera Link, 1394 DCAM ) they are using or what features they are implementing, the application programming interface ( API ) should be always the same. »

• GenICam standard overview.
• Introduction to machine vision standardization.
• GenICam standards and specifications.

www.emva.org

« EMVA has launched an initiative to define a unified method to measure, compute and present specification parameters for cameras and image sensors used for machine vision applications. Application of this standard will be at the benefit of the customers of vision components component manufacturers.

It will avoid misunderstanding and reduce pre and post support cycles. The chosen parameter definitions are based on physical properties and units. The standard comprises the mathematical model behind the parameters, the measurement setup and the matching of the measurement data to the model as well as the data presentation. »

• Standard for Measurement and Presentation of Specifications for Machine Vision Sensors and Cameras.
• EMVA 1288 standard downloads.

EMVA - The European Machine Vision Association

www.emva.org