3 Chip Vs 1 Chip Camera
Three-chip beamsplitter cameras provide image acquisition that is simultaneous in time and space and perfectly registered. This allows them to handle tasks that are difficult or impossible for LFSS or single-chip color camcorders to do.
A one-chip camera relies on a Bayer or other mosaic optical filter to separate the incoming light into its component colors. The resulting luma and chroma signals are combined via pixel interpolation to produce the final video image.
Improved Sensitivity
Three-chip cameras can capture images that are simultaneously and accurately registered in both horizontal and vertical planes. The image-acquisition process can be realized in analog as resistor networks, or it may be computed digitally in a DSP. In microscopy applications the advantages of a three-chip camera are clear.
In a one-chip camera, the pixel values of adjacent pixels must be interpolated to derive the red, green and blue signals for each pixel, which can result in artifacts. This is a problem that can be mitigated by using dual-row readout, but it still limits the maximum resolution of a single-chip camera.
A three-chip imager uses a prism to separate the light into red, green and blue wavelengths for each sensor. Each sensor can then be optimized for the color it will “see”, either through chemical doping or by a combination of sensor design and electronic settings. This results in a superior image that is ideally suited for most microscopy, machine vision and broadcasting applications.
Better Color Accuracy
Using separate red, green and blue image sensors allows each chip to be chemically doped to provide full-bandwidth R, G, B signals. This allows the camera to achieve a high degree of color differentiation.
By contrast, a single-chip camera must perform the color discrimination on its own through an array of color filters on the image sensor itself or via a prism. The resulting signal requires more processing power and may introduce artifacting.
In our testing of three-chip cameras (see Table 2), all had accuracy better than 1 CIE LAB difference, which is able to be discerned by 50% of observers for opaque colors. Those results are impressive, especially since the three-chip cameras used in this test required more expensive lenses and a bigger camera body than a single-chip video camera with a prism.
Of course, this extra expense and complexity translates into higher prices for 3-chip cameras. Nevertheless, the advantages of improved sensitivity and greater color accuracy are compelling for many video applications.
Lower System-Development and Maintenance Costs
A three chip camera uses an optical splitter block to land the red, green and blue image components onto separate sensors – traditionally CCDs but increasingly CMOS devices. This allows a camera to capture and control each of the colour channels separately resulting in better, more accurate, colour.
Moreover, because each sensor has its own photosites (not shared as in a Bayer pattern), there is no mingling of pixels which could result in undesirable aliasing. This makes it much easier to optimise low-pass filters for maximum sharpness without objectionable artefacts.
While it is possible to build a large traditional Bayer pixel camera, and many cameras do, these are often not suited for use with high resolution video. For example, they are unlikely to provide enough depth of field control for professional applications unless the camera is made very large. The three-chip approach provides a more practical solution, with the added benefit of lower system-development and maintenance costs.
Higher Resolution
Once upon a time any camera that didn’t have three separate imaging sensors for red, green and blue was viewed as third rate. Those days are long gone, however, as single sensor cameras have become the workhorses of modern audiovisual production.
A three chip camera works by using an optical splitter block to land a red, green and blue image component on each of three completely separate sensors (usually CCD but nowadays CMOS is more common). The images from these are then combined digitally to produce the video output.
This design allows manufacturers to produce a camera with higher resolution than a one CCD camera without sacrificing any of the other important attributes like lower noise, sensitivity or dynamic range. For example, the Karl Storz Endovision TRICAM NTSC HD endoscopic 3-Chip Camera has an incredible horizontal resolution of more than 750 lines. This means that even the most subtle variations in tissue structure are discernible.