Technical Specs

set to ExposureAlternateActive upon each trigger call and reset to Low after receiving images for stereo
matching.
Note: The Single + Out1 mode can only change the out1
_
mode if the IOControl license is available
on the rc_visard.
quality (Quality) Disparity images can be computed in different resolutions: high (640 x 480), medium (320 x
240) and low (214 x 160). The lower the resolution, the higher the frame rate of the disparity image. A 25
Hz frame rate can be achieved only at the lowest resolution. Please note that the frame rate of the disparity,
confidence, and error images will always be less than or equal to the camera frame rate.
Additionally, full resolution matching with 1280 x 960 pixel is possible with a valid StereoPlus license. If
full resolution is selected, the disparity range is internally limited to 128, due to limited on-board memory
resources. Due to this limitation, it is not recommended to use the full resolution matching in parallel to the
SLAM component.
static
_
scene (Static) This option averages 8 consecutive camera images before matching. This reduces noise,
which improves the stereo matching result. The timestamp of the first image is taken as timestamp of the
disparity image. This option only affects matching in full or high quality. It must only be enabled if the
scene does not change during the acquisition of the 8 images.
disprange (Disparity Range) The disparity range always start at 0 and goes up to the maximum disparity value
a pixel in the disparity image can have. Increasing the disparity range results in a smaller minimum distance
that can be measured, because larger disparity values mean smaller distances. The disparity range is given
in pixels and can be set to a value between 32 pixels and 512 pixels. Since a larger disparity range also
means a larger search area for the matching pixel in the right rectified image, the processing time increases
with a larger disparity range and the frame rate decreases. The disparity range’s value is related to the high
quality disparity image with 640 x 480 pixels and does not have to be scaled when a different quality is
chosen. Thus, the chosen disparity range gives the same minimum distance for every image-quality option.
If full resolution quality is selected, the disparity range is internally limited to 128, due to limited on-board
memory resources.
smooth (Smoothing) This option activates advanced smoothing of disparity values. It is only available with a
valid StereoPlus license.
fill (Fill-in) This option is used to fill holes in the disparity image by interpolating a plane. Only holes smaller
than the segmentation size (see below) are selected for interpolation. The fill-in value is the maximum
allowed disparity deviation of any of the hole’s border pixels from the interpolation plane. Only if all of
its border pixels deviate less than the fill-in value from the plane, a hole will be filled. Larger fill-in values
decrease the number of holes, but the interpolated values can have larger errors. The confidence for the
interpolated pixels is set to a low value of 0.5. Their error is set to the mean deviation of the hole border
pixels from the interpolation plane. A value of 0 effectively switches hole filling off.
seg (Segmentation) The segmentation parameter is used to set the minimum number of pixels that a connected
disparity region in the disparity image must fill. Isolated regions that are smaller are set to invalid in the
disparity image. The value is related to the high quality disparity image with 640 x 480 pixels resolution
and does not have to be scaled when a different quality is chosen. Segmentation is useful for removing
erroneous disparities. However, larger values may also remove real objects.
median (Median) This value gives the window side length in pixels for the median filter, which smoothes the
disparity image. Larger values lead to oversmoothing and cost more processing time. A window size of 1
effectively turns this filter off.
minconf (Minimum Confidence) The minimum confidence can be set to filter potentially false disparity mea-
surements. All pixels with less confidence than the chosen value are set to invalid in the disparity image.
maxdeptherr (Maximum Depth Error) The maximum depth error is used to filter measurements that are too
inaccurate. All pixels with a larger depth error than the chosen value are set to invalid in the disparity
image. The maximum depth error is given in meters. The depth error generally grows quadratically with an
object’s distance from the sensor (see Confidence and error images, Section 6.2.3).
6.2. Stereo matching 38