The MDCS3 is a an all-digital megapixel camera for machine vision tasks. It is an upgrade from the previous MDCS model, incorporating a new low-noise, high-sensitivity CMOS imager. High performance and low power in a compact package, with a simple IEEE 1394 (FireWire) interface. Full synchronization of all MDCS3 cameras on the same bus. Uncompressed video at megapixel resolution (15 fps) or VGA (60 fps), color or monochrome. System includes standard lenses and DCAM software for image acquisition.
- Maximum resolution: 1280H x 960V pixels
- Color and monochrome versions
- Direct digital information from the imagers, bypasses analog noise and frame grabber jitter
- Conforms to Digital Camera Specification version 1.30.
- Single cable interface for power, data, and control
- External synchronization - all MDCS3 devices on the same IEEE 1394 bus are synchronized to capture images at the same time
- Subsampling modes (decimation) for smaller frame sizes, faster frame rates, lower noise
- Electronic gain and exposure
- All modes fully under user control via the IEEE 1394 bus
- Milled solid aluminum allow body for extreme rigidity
- Accepts standard C/CS-mount lenses
MDCS3 Kit Contents
- MDCS2 stereo head (monochrome) or MDCS3 stereo head (color)
- 6.0 mm C/CS mount lenses standard; other lenses available as options
- DCAM image acquisition software for MS Windows 2000/XP or Linux 2.4/2.6
- Hardware and Software Manuals
MDCS Requirements
- Pentium-compatible PC (Pentium MMX, AMD K6-2 or better) running MS Windows 2000/XP or Linux 2.4 kernel
- IEEE 1394 PCI or PCMCIA card with a free port, and a 6-pin cables
NOTE 1: IEEE 1394 card must supply 1 W of power through the 6-pin cable. Most PCI cards do this; PCMCIA cards must have a separate transformer power input.
Technical Info
Manuals
Data Sheet
Overviews
Imager and Lens Characteristics
Imager size - The imager on the MDCS3 cameras (either stereo or monocular) has square pixels, 5.2 um on a side. The standard active area (1280x960) has size 6.66 x 4.99 mm. This is just slightly larger than the nominal 1/2" imager size (6.4 x 4.8 mm). Most 1/2" format lenses will be adequate, and there should be no noticeable darkening of the corners of the image ("vignetting").
C vs. CS mount - The STOC-9cm, STH-DCSG-9cm, STH-DCSG-VAR/-X, STH-MDCS3/-VAR/-X, MDCS, and DCSG take either C or CS-mount lenses. Any C-mount lens can be converted to a CS-mount lens by the addition of a 5 mm adapter, which is included with the devices.
Field of View (FOV) - This is the most important parameter for most applications. The FOV of the image depends on the focal length of the lens. The table below lists FOV for some common focal lengths. The formula for FOV of an ideal (pinhole) lens is
HFOV = (180/pi) * 2 * tan-1(H/2f), H = 6.66 mm
VFOV = (180/pi) * 2 * tan-1(V/2f), V = 4.99 mm
where H is the horizontal size of the imager in mm, V is the vertical size (mm), and f is the focal length (mm). Note that the horizontal and vertical FOV are generally different, because the imager frame is rectangular. Also, the FOV for shorter focal lengths is generally greater than the pinhole model, because the image is compressed towards the edges. Finally, for stereo applications, there will be erosion of the output disparity image, because some amount of margin is needed on the left and the right of the image for running the stereo algorithm.
Focus and Iris - The lenses we supply for the MDCS3 devices come with a manual focus and iris adjustment. The depth-of-field for a lens varies with a number of factors, including the iris setting (aperture) and the focal length. Generally speaking, unless the application demands extreme close-ups, it is sufficient to set the focus of the lens so that objects are in focus from some point near the camera out to infinity. For example, with 6.0 mm lenses and a wide-open iris, a single focus setting suffices for about 0.5 m to infinity.
The manual iris controls the aperture of the lens, the amount of light that is passed. In most situations, the iris is left wide open to allow the maximum amount of light through. The MDCS3 cameras have electronic exposure control to compensate for lighter and darker conditions. The electronic exposure can be set under program control, or via an auto adjustment algorithm. For some situations, where there is very bright light, the iris can be closed down to prevent the imager from over-saturating. Also, if a very large depth-of-field is required, shutting down the aperture can help.
Because of electronic exposure control, a motorized iris is not necessary on the MDCS3 cameras.
Varifocal lenses - If the FOV for an application is known, then the correct lens can be chosen based on the FOV formulas above. For experimentation, it is possible to use varifocal lenses. These lenses have an additional setting to change the focal length of the lens. In stereo applications, changing the focal length requires a re-calibration of the stereo parameters. Since the calibration procedure is fairly simple, varifocal lenses provide an easy way to experiment with various FOV settings. Varifocal lenses must have a locking screw to preserve the focal length setting (see below).
Locking vs. non-locking lens - Lenses with manual controls (focus, iris, focal length) may have optional locking screws. These locking screws allow the desired settings to be rigidly kept, once the lenses are tuned for the application. Locking screws are not required (except on varifocal lenses, see above), since friction will generally hold the iris and focus settings. If the camera is moved around, subject to vibration, or can be touched, then locking screws are a good way to insure that the desired settings are kept.
Lens Selection
These are standard lenses that we offer for the STOC-9cm, STH-DCSG-9cm, STH-DCSG-VAR/-X, STH-MDCS3/-VAR/-X, MDCS, and DCSG cameras. Please enquire about other focal lengths. All are locking except for the 16 mm.
| Locking Lens |
FOV
(H deg x V deg) |
Weight
Grams per pair |
3.5 mm, F1.6, 1/2" C/CS mount
Manual iris and focus |
87 x 71 |
180 g |
4.5 mm, F1.4, 1/2" C/CS mount
Manual iris and focus |
59 x 41 |
142 g |
6.0 mm, F1.4, 1/2" C/CS mount
Manual iris and focus |
58 x 45 |
180 g |
12 mm, F1.2, 1/2" C/CS mount
Manual iris and focus |
31 x 24 |
180 g |
16 mm, F1.4, 2/3" C/CS mount
Manual iris and focus (No lock) |
26 x 18 |
120 g |
4.5-12.5 mm, F1.2, 1/2" CS-Mount
Manual iris, focus, focal length |
Variable |
200 g |
Range Resolution
The resolution of range images produced from stereo is a function of the stereo head optics, imager characteristics, and distance to the object. The chart below show the range resolution for three of the lenses above.
The range resolution was calculated based on a pixel size of 6.0 um, sub-pixel interpolation of x16, and a stereo baseline of 90 mm. The range resolution grows with distance squared. Even at moderate distances, though, the range resolution is excellent. For example, a wide-angle 4.8 mm lens still has a precision of about 2 cm at 5 m.
The formula for computing range resolution is:
dR = .0052 * R**2 / (16*90*L),
where R is the range to the object, and L is the lens focal length. All dimensions should be in mm.
Color, Monochrome, and IR
MDCS3 and DCSG devices come in two varieties, color and monochrome. The basic imagers are the same, but the color type adds a striped color pattern on top of the imager photo sites. While color information can be useful in an application, it degrades the spatial resolution of the luminance signal, which is used by the stereo algorithms.
The CMOS imagers are infrared sensitive out to about 900 nm (near IR). For some applications, near-IR response is useful. In most cases, however, near-IR response degrades the sharpness of the image, since the IR wavelengths focus at a slightly different place. The MDCS3 includes IR cut filters to eliminate near-IR wavelengths.
|
Color vs. Monochrome
The MDCS3 and DCSG devices come in either color or monochrome versions. They both use the same underlying CMOS megapixel imaging array. The color version incorporates a Bayer pattern color filter on the pixels. Every set of four pixels has the color pattern:
A color image is reconstructed from this pattern by assigning R, G, and B values at every pixel of the color image. A luminance (monochrome) value can also be assigned by a suitable function of the four pixels. If the color image is the same size as the original Bayer image, then the colors and luminance must be interpolated. Because of the high resolution of the MDCS3, non-interpolated color images at 640 x 480 are also possible.
By contrast, in a monochrome imager each element of the imager array records just the luminance at that pixel. There is no color information, but the luminance information is recorded at the full resolution of the imager.
There are several important differences between the two types of stereo heads, in terms of image quality and stereo algorithms.
- Because the color imagers have filters on the pixels, they receive less light, and thus have less sensitivity than the monochrome imagers.
- The luminance spatial resolution of the color imagers is less than that of the monochrome imagers. Since the stereo algorithms use luminance information, the monochrome head is better for stereo.
Although the monochrome imagers give better spatial resolution than the color imagers, the large number of pixels means that there is almost always adequate spatial resolution for an application. Hence we do not recommend monochrome over color, except where low light levels may be a factor.
The response curves for the MDCS3 imagers are given just below.
Color Algorithm
Color interpolation from the Bayer pattern to RGB images is done on the host PC. The Small Vision System and the DCAM drivers incorporate three different algorithms for color interpolation.
- Color binning. This algorithm is used when the image from the device has twice the resolution of the required output image. For example, if the device is sending a 640x480 image to the PC, and the output resolution is 320x240, then the binning algorithm is used. Binning gives the best color quality.
- Edge-weighted color interpolation. This algorithm takes edges into account, and gives a good rendition of color when input and output images are the same resolution. It is a slower algorithm than (3), but the results are better.
- Color averaging. This algorithm averages colors from neighboring pixels to complete the color image. It is very fast, but produces results that are less satisfactory than (2).
Below are blow-ups showing individual pixels, of algorithms (2) and (3), on black lettering. Note how much crisper the top (2) image is. The bottom (3) image shows a "zipper" effect on the vertical lines, where alternating lines are offset slightly.
Infrared Characteristics
The CMOS imagers in the MDCS3 are highly sensitive to near-infrared radiation (up to 950 nm). The following images show a scene with just IR light admitted to the imager, just visible light, and a combination of visible and IR.
|
|
|
|
|
Visible spectrum light only (< 700 nm)
|
Visible + infrared light
|
Infrared light only (>> 700 nm)
|
Some objects reflect IR energy, e.g., vegetation and the lacquer in the Japanese tansu (the TV images were different over the three images, so ignore them).
Most lenses are optimized for visible light, and longer IR wavelengths are not focused at the same distance as visible light. This creates blur when both IR and visible light are admitted to the imager, for both color and monochrome stereo heads. If you click on the first two images above to get larger images, and look at the books on the bookshelf, you can see much more detail in them.
The Bayer color filter does not block near IR, so colors will be "washed out" if IR is admitted to the color imager.
For these reasons, both the monochrome and color MDCS3 devices incorporate a high-efficiency IR cut filter, which essentially eliminates all radiation over 700 nm.
For near-IR work, the filter can be removed, and IR pass filters placed over the lenses to eliminate visible light, as in the third image above. Please contact Videre Design if you are interested in this option.
|
