January 2, 2015


Machine vision is an evolving technology used to replace or complement manual inspections and measurements. The technology uses digital cameras and image processing software. This technology is used in a variety of different industries to automate production, increase production speed and yield, and to improve product quality. One primary objective is discerning the quality of a product when high-speed production is required.  This industry is knowledge-driven and experiences an ever- increasing complexity of components and modules of machine vision systems. In the last few years, the markets pertaining to machine vision components and systems have grown significantly.

Machine vision, also known as “industrial vision” or “vision systems”, is primarily focused on computer vision in the perspective of industrial manufacturing processes like defect detection and in non-manufacturing processes like traffic control and healthcare purposes. The inspection processes are carried by responsive input needed for control; for example, robot control or default verification. The system setup consists of cameras capturing, interpreting and signaling individual control systems related to some pre-determined tolerance or requirement. These systems have increasingly become more powerful while at the same time easy to use. Recent advancements in machine vision technology, such as smart cameras and embedded machine vision systems, have increased the scope of machine vision markets for a wider application in the industrial and non-industrial sectors.


Let’s take a very quick look at several components and systems used when applying vision to specific applications.


You can see from the graphic above products advancing down a conveyor past cameras mounted on either side of the line.  These cameras are processing information relative to specifications pre-loaded into software.  One type of specification might be a physical dimension of the product itself.  The image for each may look similar to the following:


In this example, 55.85 mm, 41.74 mm, and 13.37 mm are being investigated and represent the critical-to-quality information.  The computer program will have the “limits of acceptability”; i.e. maximum and minimum data.  Dimensions falling outside these limits will not be accepted.  The product will be removed from the conveyor for disposition.

Another usage for machine vision is simple counting, and the following two JPEGs will indicate.




One example of a non-industrial application for machine vision is facial recognition.   This technology is generally considered to be one facet of the biometrics technology suite.  Facial recognition is playing a major role in identifying and apprehending suspected criminals as well as individuals in the process of committing a crime or unwanted activity.  Casinos in Las Vegas are using facial recognition to spot “players” with shady records or even employees complicit with individuals trying to get even with “the house”.   This technology incorporates visible and infrared modalities face detection, image quality analysis, verification and identification.   Many companies use cloud-based image-matching technology to their product range providing the ability to apply theory and innovation to challenging problems in the real world.  Facial recognition technology is extremely complex and depends upon many data points relative to the human face.



A grid is constructed of “surface features”;those features are then compared with photographs located in data bases or archives.  In this fashion, positive identification can be accomplished.

One of the most successful cases for the use of facial recognition was last year’s bombing during the Boston Marathon.   Cameras mounted at various locations around the site of the bombing captured photographs of Tamerian and Dzhokhar Tsarnaev prior to their backpack being positioned for both blasts.  Even though this is not facial recognition in the truest since of the word, there is no doubt the cameras were instrumental in identifying both criminals.



Remember that last ticket you got for speeding?  Maybe, just maybe, that ticket came to you through the mail with a very “neat” picture of your license plate AND the speed at which you were traveling. Probably, there was a warning sign as follows:


OK,so you did not see it.  Cameras such as the one below were mounted on the shoulder of the road and snapped a very telling photograph.


You were nailed.


There are five (5) basic and critical factors for choosing an imaging system.  These are as follows:

  • Resolution–While a higher resolution camera will help increase accuracy by yielding a clearer, more precise image for analysis, the downside is slower speed.
  • Speed of Exposure—Products rapidly moving down a conveyor line will require much faster exposure speed from vision systems.  Such applications might be candy or bottled products moving at extremely fast rates.
  • Frame Rate–The frame rate of a camera is the number of complete frames that a camera can send to an acquisition system within a predefined time period, which is usually stated as a specific number of frames per second.
  • Spectral Response and Responsiveness–All digital cameras that employ electronic sensors are sensitive to light energy. The wavelength of light energy that cameras are sensitive to typically ranges from approximately 400 nanometers to a little beyond 1000 nanometers. There may be instances in imaging when it is desirable to isolate certain wavelengths of light that emanate from an object, and where characteristics of a camera at the desired wavelength may need to be defined.  A matching and selection process must be undertaken by application engineers to insure proper usage of equipment relative to the needs at hand.
  • Bit Depth–Digital cameras produce digital data, or pixel values. Being digital, this data has a specific number of bits per pixel, known as the pixel bit depth.  Each application should be considered carefully to determine whether fine or coarse steps in grayscale are necessary. Machine vision systems commonly use 8-bit pixels, and going to 10 or 12 bits instantly doubles data quantity, as another byte is required to transmit the data. This also results in decreased system speed because two bytes per pixel are used, but not all of the bits are significant. Higher bit depths can also increase the complexity of system integration since higher bit depths necessitate larger cable sizes, especially if a camera has multiple outputs.


Machine vision technology will continue to grow as time goes by simply because it is the most efficient and practical, not to mention cost effective, method of obtaining desired results.  As always, I welcome your comments.

What do you think?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: