JAI’s new 2-CCD camera provides HDR color video at 1.3 megapixel resolution
August 2, 2012
SAN JOSE, CA – August 2, 2012 - JAI today announced the release of the AD-132GE, a new 2-CCD camera designed
to produce high dynamic range (HDR) color images by intelligently combining different exposures from the camera’s dual sensors
in real time. The new camera builds on the concept originally introduced by JAI in late 2009 in the AD-081, which featured two
0.8 megapixel monochrome sensors configured for HDR imaging. The new AD-132GE not only adds the capability of color imaging,
but also increases full frame resolution to 1.3 megapixels (1296 x 966 pixels) by utilizing two Sony ICX447 CCDs. Output from
the camera is via a standard GigE Vision digital interface.
The two precisely-aligned 1/3” CCDs are mounted to a high performance beam-splitter prism which enables the camera to provide
an identical field of view to each CCD at a full resolution speed of 31 frames per second. The shutter speed and/or gain for
each CCD can be calibrated independently such that one imager captures details in the brighter areas of a scene, and the second
CCD captures the same image, but with an emphasis on the details in the darker areas. The two images or video streams can then
be processed with in-camera image fusion algorithms to produce a dynamic range nearly double the normal CCD response.
Alternatively, users can choose to send the camera’s raw video output to a host computer for post-processing using their own
algorithms. Either way, JAI’s high performance multi-imager technology enables the AD-132GE to achieve dynamic range levels
as high as 20-bits per pixel (~120 dB) in a linear fashion that avoids the noise, rolling shutter, and compression issues found
in comparable CMOS-based logarithmic or LinLog™ HDR cameras.
The AD-132GE is targeted at a range of applications where extreme lighting contrast affects image quality.
These include inspection tasks where incident light or bright reflections are present, such as LED inspection, welding,
glass inspection, solar panel manufacturing, and monitoring of industrial furnaces or heated metal. The camera is also ideal
for microscopy applications, as well as high-end surveillance or traffic applications where sun and shadow or indoor/outdoor
scenes make it difficult for conventional cameras to maintain exposure across the entire field of view. The parameters of each
CCD can be adjusted by the user to provide maximum dynamic range or maximum contrast/sensitivity within a narrower lighting range,
depending on the application.
Several other operating modes are provided to enable users to leverage the unique 2-CCD configuration for uses other than
HDR imaging. For example, a high signal-to-noise (S/N) mode automatically averages the video information from the two CCDs
to produce a single image with a significantly lower noise component than a typical single-CCD image of the same scene.
A double-speed mode offsets the timing of the two CCDs by one-half frame and interleaves captured images into a single
1.3 megapixel output stream at 62 fps. Because this is done without any increase in the clock frequency of the camera,
users can achieve high frame rates while keeping clock noise to an absolute minimum.
Also included is an advanced PIV mode, which leverages the camera’s two-channel operation to capture three closely-spaced
images on a single trigger instead of only two images captured by conventional PIV trigger modes. This allows 50% more data
to be collected to better analyze ultra-fast events such as vortex forming in artificial heart chambers, combustion analysis
in engines, and air flow studies in wind chambers, to name a few.
The AD-132GE’s GigE Vision output is provided via two configurable RJ-45 ports, offering users a choice of single-cable or
dual-cable operation depending on the mode used. Output can be either 24-bit or 30-bit RGB with in-camera interpolation,
or is available in raw Bayer form (8, 10, or 12-bit) for host-based interpolation. Other features include partial scanning,
blemish and shading compensation, as well as analog video output to support auto-iris lenses.
More Stories in Imaging and Machine Vision