Despite the fact that they have been around for several decades, line scan cameras remain perplexing to many in the machine vision industry. Two-dimensional “area” cameras are still the norm in many circles – likely because they are more like human vision and, therefore, more intuitive – so much so that I have seen area cameras “shoe-horned” into applications where a line scan camera would have served nicely. Web inspection applications are a good example of where line scan cameras offer benefits over an area camera.
Why line scan in web inspection?
In cases where the web material does not have repeating patterns at regular intervals (steel, aluminum, glass, blank paper, non-woven materials, etc.), using an area camera to get 100% coverage of the web can be extremely difficult. An area camera, by its nature, takes snapshots of a two-dimensional piece of the web. Once two successive snapshots are taken, the two images need to be aligned, so that the top line of image #2 begins exactly where the bottom line of image #1 left off. With no repeating patterns on the web to assist in alignment, you can see how this task can be very difficult and, more than likely, will have to rely on the time between frames, compared against the speed of the web.
This is precisely why the line scan camera was designed. The line scan camera will acquire one-dimensional image lines at a speed dictated by an encoder. An encoder is a device that is mechanically attached to the web and can tell the line scan camera to increase or decrease its line acquisition speed based on the speed of the web – thereby keeping the distance between lines constant (something an area camera cannot do). The image lines can be divided into any number and at any position for the host computer to process and, no matter what data set the computer looks at, the images will always be in perfect alignment and provide 100% coverage of the web.
How it all works
Since the line scan camera acquires only a single, one-dimensional image line at a time, it relies on the motion of the web to build a series of these one-dimensional image lines, taken at different locations in the direction of travel of the web, thus creating a two-dimensional image, which, after all, is all a two-dimensional image is – a collection of one-dimensional lines taken at slightly different locations and stacked one on top of another.
The encoder is usually a wheel or drive shaft which makes physical contact with the web. As the web moves, the wheel of the encoder is pushed by friction like the rollers on the web. Every millimeter of movement in the direction of travel on the web translates to an angular distance on the wheel of the encoder. The encoder will generate some number of pulses per revolution (1,000 and 2,000 pulses per revolution are common), and these pulses tell the line scan camera when to grab the next line. In this way, the line scan camera is in perfect synchronization with the movement of the web and, no matter whether the web speeds up or slows down, the line scan camera will always grab the same number of one-dimensional lines for a given distance that the web has traveled.
Try it out.
The best way to get a feel for how line scan technology works in web inspection situations is to try it. With decades of experience applying line scan cameras to web inspection (among other types of inspection) Teledyne DALSA is uniquely positioned to show you how easily and efficiently this technology can be integrated into your next application.