Lighting Equipment Questions & Answers
Click a question to reveal the answer. Don't see the information you're looking for? Take a look at the archived questions or ask a new question by clicking the button to the right.
"Sheet of Light" as used in machine vision terminology is another name for structured line light illumination. It is used to extract a 3D profile from an object. It requires an area scan camera and often uses a laser line projector (although other illumination sources are possible). The line (or sheet) of light illuminates an object and creates a geometric structure related to the height of features on the object. To create a full image, the part is moved in front of the camera and illumination (or the imaging system is moved over the part), and many pictures are taken of the line geometry. Through proper calibration, image reconstruction and analysis, an accurate 3D point cloud of the object is obtained. Many vendors offer either components or software tools to do this type of imaging including SICK, Halcon, Scorpion, Cognex and several others. This is only a brief overview; please feel free to contact me directly and I would be happy answer any questions you may have.
Answered by:
David Dechow
President
Aptúra Machine Vision Solutions
ddechow [at] aptura [dot] com
(517) 655-7860, x11
Hi Bill! The type of lighting you will need for your imaging application depends on several factors, such as the surface structure of your parts and what you would like to inspect. The term sheet of light is commonly used for 3D imaging using laser triangulation. We have some explanations on our website as well: http://www.movimed.com/moviSCAN_3D_Cameras.htm However, the best way to go about your application is to get a few sample parts in our imaging lab and to figure out the best way to enhance the contrast of the features of interest you are trying to examine or measure. If you can email me some photos of the parts I can comment on a possible approach. We have solved a lot of medical/health care type imaging applications.
Answered by:
Markus Tarin
President and CEO
moviMED
m [dot] tarin [at] movimed [dot] com
(949) 699-6600
What is the light you are replacing? Can you get me your details on what is needed? Pattern of light? Working distance? Wavelength?
Answered by:
Matt Pinter
Design Engineer
Smart Vision Lights
matt [at] smartvisionlights [dot] com
(231) 722-1199 x-101
Lance, We are a machine vision system integrator, please feel free to call me at your convenience to discuss.
Answered by:
Andrew Long
CEO
Cyth Systems, Inc.
andy [dot] long [at] cyth [dot] com
888 508 7355
A good question. Lighting creates contrast, and, in machine vision, that is our signal. However, contrast is the difference between two things. You mention sheet metal as one of the things, but did not tell us what you are trying to contrast against the sheet metal. Also we would need to know something about the surface characteristics of the sheet metal (e.g., polished, fine grained, galvanized, etc.). For mosaicing, I'm assuming you are working on a wide area with multiple cameras (images). That imposes some constraints about achieving uniformity. So, knowledge of the area being illuminated is also necessary.
Answered by:
Perry West
President
Automated Vision Systems
perry [at] autovis [dot] com
(408) 267-1746
What aspect of the sheet metal, what are you tying to observe/measure? Creating contrast is the goal, contrast of what is the question. Depending on the feature(s) you are trying to observe or measure, you will be pushed to different lighting technologies/configurations.
Answered by:
Wallace Latimer
Product Line Manager
Coherent Inc
wallace [dot] latimer [at] coherent [dot] com
(513) 454-5826
Although we could supply a darkfield ringlight in any LED color, specular reflections are caused by the angle of the light more than its color. Perhaps a set of four (bar lights) arranged into a square where you could adjust each one's angle would help you find. In addition polarizing filters on both the light and lens could also knock out specular effects. Give us a call to discuss further.
Answered by:
Scott Bouchard
Sales Manager
Innovative Illumination & Optics
sbstang [at] yahoo [dot] com
(866) 570-5736
John -- Scott is right when he said that wavelength will likely not have any effect on the specular reflections. A lot depends on the microtexture. If it's truly irregular like sandpaper, there is not going to be any illumination angle that will eliminate all glints. If it's a manufactured texture about which you know the geometry, there may be an illumination angle or range of angles that will work to keep the specular reflection out of the camera. The challenge may be to get the illumination within this angle range across the entire field-of-view. The other option Scott offered is the polarizer-analyzer approach. Using a polarizer over the light source and an analyzer (2nd polarizer) over the camera lens, you might be able to tune the analyzer position to remove the glints. However, if the texture results in multiple reflections off the texture facets, even this might not work.
Answered by:
Perry West
President
Automated Vision Systems
perry [at] autovis [dot] com
(408) 267-1746
I agree with Scott and Perry, Since the surface you are inspecting is white it will reflect all visible wavelengths equally. Therefore, simply changing the color of the illumination is unlikely to solve the specular reflection problem. What you will have to do is change the geometry of the illumination, perhaps abandoning the idea of a ring light. One thing you did not mention is what kind of features you are trying to detect. That will have a bearing on what lighting geometry will give the best results.
Answered by:
Robert McConnell
WAY-2C Color Machine Vision
rkm [at] way2c [dot] com
(781) 641-0605
John, Materials reflect/absorb so different when tested with various wavelengths. In my experience, the best way is to test the different wavelengths (IR, Red, Blue, Green, Cyan, White and even UV) on your material. We can test your sample with the different wavelengths and filters if you would like. We also have a hand held tester that has 6 wavelenths in it. I also recommend filters to be tested with the wavelengths. Filters are very helpful on applications in conjunction with wavelengths.
Answered by:
Matt Pinter
Design Engineer
Smart Vision Lights
matt [at] smartvisionlights [dot] com
(231) 722-1199
I do not know of anything generic that can be used other than to maybe say that the area where machine vision inspection is to occur should be enclosed to block out ambient light and only light needed for the inspection should be employed. I hope this is of some help to you.
Answered by:
Rex Lee
CEO/President
Pyramid Imaging, LLC
rlee [at] pyramidimaging [dot] com
(813) 984-0125
The most common ways to reduce ambient in a vision application are the use of a filter or overpowering the ambient with the camera lighting. A bandpass filter is an excellent way of reducing ambient to less than 10% of your light in the camera image. A band pass filter will block 90% plus of ambient. Midwest Optical is a good source for information and product on machine vision filters. A band pass will only allow the a color or specific wavelengths to pass through the lens into the camera. A LED light with a band pass filter is a popular way of getting rid of ambient light. Another option is make the light in the vision system bright enough that the ambient is not a factor. If the light for the vision is bright enough the ambient will not have as great of effect. I recommend a bandpass filter for monochromatic lighting applications but if white light is needed then overpowering the ambient or a shroud maybe the only option.
Answered by:
Matt Pinter
Design Engineer
Smart Vision Lights
matt [at] smartvisionlights [dot] com
(231) 722-1199
Steve -- From your question, you obviously know the following that I'll restate as an introduction to my remarks. Considering light sensed by the machine vision system, you have the engineered illumination and the ambient light. Engineered illumination creates the signal; ambient light creates the noise. The problem, then is signal-to-noise ratio. To improve the ratio, you need to either increase the signal or decrease the noise or some of both. I don't have template language for you as each application is very unique. It's difficult to specify ambient light (e.g., intensity, direction, spectrum, etc.), and this can change for several reasons. I suggest your buy-off test incorporate a provision to compare the vision system's operation under different lighting conditions (e.g., lights off and lights on) and require that the performance not change by more than some modest delta.
Answered by:
Perry West
President
Automated Vision Systems
perry [at] autovis [dot] com
(408) 267-1746
Consider near infrared illumination for your engineered application lighting and a visible light (long pass) filter on your camera. Ambient lighting is seldom in this wavelength area and cameras are usually quite sensitive. As a bonus, LED's usually have quite strong output in near infrared. We can furnish illumination, filter, and lens solutions for your next application. Give us a call or e-mail info@i2optics.com.
Answered by:
Scott Bouchard
Sales Manager - Innovative Illumination & Optics
Innovative Illumination & Optics
sbstang [at] yahoo [dot] com
(866) 570-5736
We do not have anthing at this time. But we are working on the high speed light for the high speed camera market that will be out late fall.
Answered by:
Bobby Segraves
Sales Manager
Smart Vision Lights
bobby [at] smartvisionlights [dot] com
(231) 722-1199
No light source that I know of produces an exactly uniform light over the visible spectrum. Some come close but still are not totally uniform. One suggestion would be to use a fiber optics light with an appropriate illumination source. The lamps for that type of set-up may come closer to uniform than an LED or fluorescent, depending upon the lamp used. Fiber optic illumination is available from Schott Lighting, Dolan-Jenner, Volpi and others. There also are companies who make constant feedback illumination sensors that will monitor the lighting and make adjustments to the intensity to provide a more consistent output. In the end, you likely will have to compensate for color offset/non-uniformity by color-calibrating your camera using your best-case light source and a known (white) target. Hope this helps, good luck with your project. David
Answered by:
David Dechow
President
Aptúra Machine Vision Solutions
ddechow [at] aptura [dot] com
(517) 272-7820, x11
There are 3 common white LED's available now. Cool, Neutral and Warm white. Cool do not have as much yellow and red in them while neutral and warm each have more. Warm white LED's have the most yellow and red wavelengths. You can find the spectral response curves on any of the high power LED manufacturers web sites. Cool White LED's were used in the past with majority of applications and the camera was calibrated with the "white balance" on the light to equalize the color spectrum. There in not true white light LED source that has an equal amount of wavelengths (Red, Green, Blue's) so white balance is needed. The warm white LED's have a more equalized spectral response so many have started to use them in white LED lighting applications.
Answered by:
Matt Pinter
Design Engineer
Smart Vision Lights
Matt [at] smartvisionlights [dot] com
(231) 722-1199
LED Lights can operate at 1k to 10k. Until now strobing LED lights at high speeds was not an option. High speed cameras use lights in a constant on state because speeds were beyond the capabilities of lights. The advantage to strobing a light is to increase the output. If the light does not increase for the strobe then the light could be left on. The old saying "strobing freezes motion" does have some merit but freezing motion is the capture of the image. A short image acquisition will freeze motion. If the image can not be controlled to a short enough duration THEN a short burst of light for part of the image acquisition will help to freeze motion. We have developed a LED light for high speed cameras. It has been 2 years on the making. Light will be available 4th quarter 2010. This LED light can strobe at speeds of up to 5k (we say SPS - Strobes Per Second) while outperforming a Xenon strobe. This new high speed, high output light is a break through in camera lighting.
Answered by:
Matt Pinter
Design Engineer
Smart Vision Lights
matt [at] smartvisionlights [dot] com
(231) 722-1199




















