Mr. Ford would be proud of us.
Widely regarded as the father of industrial mass production, Henry Ford had a habit of personally inspecting every automobile that rolled off the belts of his factory. He would sign it with his own signature, well illustrating his almost obsessive attention to the quality of the legendary Model T.
When production began to accelerate, the brilliant visionary was obviously forced to abandon his practice, but customers appreciated the quality of this simple car.
Thus the brand legend was born, thanks to which America shifted from horses to cars and entrepreneurs became convinced that a mass product could be offered at a low price and with good quality.
After more than a hundred years, the method based on the owner's inspections has been replaced by an electronic eye aided by artificial intelligence,
and we know what to do so that you can sign every product that leaves the walls of your company with no hesitation.
The beginnings of machine vision.
The idea of machine vision was born in the late 1940s/early 1950s with research into artificial intelligence, when the US Army began using image analysis. However, the practical application of the nascent concept of applying it to industry had to wait until the 1960s.
The 1970s
The 1970s saw a breakthrough in the field of machine vision. The Massachusetts Institute of Technology developed an image analysis system that would ultimately control a robotic arm for industrial applications. In 1966, MIT's Marvin Minsky asked his student Gerald Jay Sussman to spend his summer holiday trying to connect a camera to a computer, and so a mathematical description of what the eye of a camera captures was born.
1980s
In the 1980s, image pyramid linking made it possible to search for congruent parts of images, and subsequent versions of pyramids also made it possible to locate scaled objects. Thanks to algorithms that reduced image analysis time, machine vision could be applied on an industrial scale.
The repeatability of measurements, and the increasing quality requirements of products, made machine vision the basis of many production lines.
In addition to constantly developing algorithms for existing methods of creating and programming machine vision, new methods were created, such as:
-
photo-stereoscopy, which involves analysing several photographs under varying lighting conditions,
-
the ‘shape from shading’ method, which makes use of the relationship between the amount of reflected light and the third dimension, the ‘shape from focus’ method, which makes use of the relationship between image sharpness and the third dimension.
1990s
The 1990s saw tremendous growth in the machine vision industry, driven by technological advances in computer design. Integrated circuit processing made it possible to make smart cameras that could not only collect image data, but also extract information from the resulting images, without the use of a computer or other external processing device.
With more powerful computers, it was possible to incorporate developed mathematical formulas into the analysis of images to help derive models of three-dimensional images.
Towards the end of the 20th century, scientists speculated on the most likely direction for the development of machine vision
and the assumptions of the researchers at that time still seem to be valid today.
David Marr in ‘Vision: A Computational Investigation into the Human Representation and Processing' included three levels of development of machine vision:
-
computational theory: what is the purpose of computing? What are the constraints and how to deal with them?
-
representation and algorithms: what is the input, output, intermediate information, and which algorithm to use for computation, to get the desired results?
-
hardware implementation: How to transfer representations and algorithms to existing hardware?
21st century
We now know that the problem of machine vision is much more complex than researchers at MIT believed in the 1970s. Early attempts to obtain 3D models were based on extracting edges in 2D space, and using a block model. By the end of the 20th century, research into edge detection algorithms was a milestone in the field of machine vision. There was also research into modelling objects with shapes other than polyhedrons, such as cylinders. It was also noted that 3D modelling could be used to create a 3D model:
-
stereoscopic vision,
-
shadows,
-
and light intensity.
Due to technological advances, machine vision on production lines has ceased to be just an idea in R&D and is now widely implemented and used in the manufacturing industry for applications including quality verification and product inspection.