Vision & Sensors / Vision & Sensors Exclusives

Case Study: Customer Requirements Define 3CCD

January 5, 2012
/ Print / Reprints /
/ Text Size+
The proof is in the inspection. Two images are captured with the same lens and light and were white balanced with the same target. The only difference is that the image on the left (A) is a single CCD camera and the image on the right (B) is a 3CCD camera. At first glance, the two images look very similar-it is not until you look closer at the enlarged view in the top left corner that you can see the difference.

The enlarged section in Figure A, or the single CCD camera, shows the evidence of interpolation. The colored pixels have become bunched in the image unlike the 3CCD camera, which has a much better resolution. A camera with 3CCD technology is best suited for high accuracy color applications, such as color verification, flaw and contaminate detection and quality control. Source: PPT Vision

It was a call from a European pharma organization with a particularly challenging inspection that led to PPT Vision’s investigation of three charge-coupled-device (3CCD) camera technology.

Within weeks of that call, the results were so impressive that the company recently launched a 3CCD camera, with the potential to respond to similar inspection challenges when absolute color accuracy is required.

“I was first contacted by Simac Masic & TSS, a specialist in turnkey vision applications and long-standing partner, with a customer need that they couldn’t meet,” says Ivar Keulers, PPT Vision European director. “Where past inspections on surface contamination used high-resolution, 2-megapixel greyscale cameras, the customer needed to also detect color changes in the surface, or small particles with small color changes in regard to the background.

“While some color applications can be handled with the correct light and filters on greyscale cameras, a feasibility study was done and the defects were too small and too less in contrast and with it a wide range of colored products the solution could not be found with greyscale cameras.”

Next, a 5- megapixel color camera was put to the test, but the resolution (detectable defect size) gave the same result as the 2-megapixel greyscale camera with not enough color depth to see minimal changes in the surface color, according to Keulers. Finally, the feasibility study tested cameras that use 3CCD technology.

“We reviewed several GigE color cameras for these applications, using the images with our PPT Vision Impact software, and confirmed that, indeed, 3CCD technology would solve the problem,” says Keulers. “And, we were convinced that there could be a growing need for 3CCD to provide high resolution and superior light gathering in more complex inspections.”

The Bayer mosaic camera uses a pattern of color filters and an interpolation process to estimate the approximate RGB value of a given pixel. 3CCD technology, on the other hand, produces higher color precision because a specific RGB value is captured for each pixel. Source: JAI


In simplest terms, CCD is the means for transmission of electrical charge, typically from within the device, to an area where the charge can be manipulated, such as converting it to a digital value. This conversion occurs by shifting the signals between stages within the device, one at a time, by moving charge between capacitive bins.

CCD was invented over 40 years ago at AT&T Bell Labs by Willard Boyle and George Smith, who were working on semiconductor bubble memory. They designed “charge bubble devices” that could be used as a shift register and linear and area imaging devices to transfer charge along the surface of a semiconductor from one storage capacitor to another. Boyle and Smith were awarded the Nobel Prize for Physics in 2009 for their work on the CCD.

3CCD in Machine Vision

Conventional color cameras in the machine vision industry process color by using a singular CCD. The CCD is a silicon chip that converts light into a digital signal by the use of photo sensors. Photo sensors are filtered using a color filter array (CFA), where each of the photo sensors is assigned a color filter: red, green or blue.

Single CCD cameras often use the Bayer filter mosaic, named after inventor Bryce E. Bayer of Eastman Kodak, as their CFA. The Bayer filter uses a pattern of 50 percent green filters, 25 percent red and 25 percent blue filters. The majority of the photo sensors are filtered green, due to the fact that the human eye is most sensitive to the color green during daylight.

Using color cameras with Bayer filter mosaics for machine vision applications can present limitations. In a monochrome camera, one photo sensor on a CCD represents one pixel of information in the output image. A color camera with a singular CCD uses the same principle, but the camera takes data from the filtered photo sensors and uses interpolation to fill in the missing data for the red, blue and green color planes. This leads to a lower resolution image on each of the three color planes.

These limitations led to the development of the 3CCD color camera, with its high-color precision that provides distinct advantages over both Bayer mosaic. In a 3CCD RGB color camera, each CCD is independently filtered for each value of red, green and blue. The three CCDs use one or multiple prisms, called trichroic prism assemblies, to direct the light towards each CCD. A trichroic prism assembly splits the incoming white light into red, green and blue light.

The advantage of having three independently filtered CCDs is that there is no reduction of image quality on the output image. A camera with three independent CCDs no longer has to use a filter mosaic to create color. Instead it combines the values from the each one of the filtered CCDs to create a very accurate pixel color on all three color planes.

A 3CCD color camera is typically more expensive than its single CCD counterpart due to the advanced infrastructure, with the multiple CCDs and prism. In order to justify the extra cost, the application should require the improved color accuracy of a 3CCD camera.

The 3CCD spectral curves that result from the soft dichroic prism coatings are much steeper than the curves resulting from the soft polymer dyes used in Bayer filters. Thus, the 3CCD camera produces exceptionally accurate color data without the uncertainty that comes from the overlap regions. Source: JAI

When to Use 3CCD

“We recognized a need for a camera with both enhanced color fidelity and resolution, and we were driven by our customers for a camera to detect more subtle color differences than what was possible with our standard color cameras,” says Steve Maves, PPT’s manager of application engineering. “Working with JAI, our strategic partner, we developed our 3CCD camera, with full 2 MP resolution for gauging applications, as well as the highest color fidelity of any other camera in the PPT lineup.

“When we are evaluating a color application with a standard color camera that involves measurement, we de-rate the camera resolution by one half. With 3CCD, this isn’t necessary because we get higher accuracy than is otherwise possible, even with our highest resolution, 5MP color camera,” says Maves. “Although the 3CCD color camera can’t deliver the kind of calibrated color information of a photo spectrometer, it is useful in a wide variety of industries where appearance is critical, including; medical devices, printing, packaging and automotive. Any application that requires accurate color and high resolution will benefit greatly from the 3-CCD color camera.”

Success with 3CCD

Tim Moonen, sales support engineer at Simac, agrees.

“The main reason for choosing the 3CCD camera is to update an existing monochrome inspection to a color inspection. A Bayer-filter camera is less light sensitive because of the color filters, which absorb light. With the 3CCD, the image light goes through a prism and is then split into three spectral bands, with very little loss of light intensity,” he says.

And, resolution is lower in Bayer filter cameras, according to Moonen.

“A 1600 x 1200 resolution Bayer filter camera has an effective resolution of 800 x 600 because of the color filter array,” he says. “Since the 3CCD camera, with 1620 x 1236 resolution, has almost the same CCD size as a standard 1600 x 1200 monochrome camera, it is possible to change the camera without changing the lens, while keeping the same field of view [FOV].”

Moonen expects that other inspections will be upgraded with this 3CCD application in the near future. “Simac has provided machines to this customer for more than 10 years-all with PPT Vision. While this is our first project with the 3CCD application, we know that the customer is happy to work with PPT.”

PPT Vision

(952) 996-9500


The advantage of having three independently filtered CCDs is that there is no reduction of image quality on the output image.

A camera with three independent CCDs no longer has to use a filter mosaic to create color.

Since the 3CCD camera has almost the same CCD size as a standard monochrome camera, it is possible to change the camera without changing the lens, while keeping the same field of view.

Did you enjoy this article? Click here to subscribe to Quality Magazine. 

You must login or register in order to post a comment.




Charles J. Hellier has been active in the technology of nondestructive testing and related quality and inspection fields since 1957. Here he talks with Quality's managing editor, Michelle Bangert, about the importance of training.
More Podcasts

Quality Magazine


2014 December

Check out the December 2014 edition of Quality Magazine for features!

Table Of Contents Subscribe

The Skills Gap

What is the key to solving the so-called skills gap in the quality industry?
View Results Poll Archive

Clear Seas Research

qcast_ClearSeas_logo.gifWith access to over one million professionals and more than 60 industry-specific publications,Clear Seas Research offers relevant insights from those who know your industry best. Let us customize a market research solution that exceeds your marketing goals.


facebook_40.png twitter_40px.png  youtube_40px.pnglinkedin_40px.png