Difference between horizontally placed steel bar and vertically placed steel bar
I have a D455 and I want to use this equipment to measure the spacing of steel bars arranged in a grid as shown below.
In the case of steel bars placed vertically during measurement, point cloud measurement works well.
However, in the case of horizontally placed steel bars, point cloud measurement is difficult.
Can we find out the cause of this?
Even when checking with just one steel bar, there is a difference in horizontal and vertical measurements.
-
Hi Dnrdyd0379 On the RGB image of the bars, the vertical bars look lighter in color than the horizontal bars. This may be because of how the light in the area is reflecting on the vertical bars when viewed from that camera angle. The horizontal bars underneath appear dark gray.
It is a general physics principle (not specific to RealSense) that dark gray or black absorbs light and so makes it more difficult for depth cameras to read depth information from such surfaces. The darker the color shade, the more light that is absorbed and so the less depth detail that the camera can obtain.
Adjusting the illumination in the area with a lamp if possible so that all bars receive equal light may help. Casting a strong light onto dark gray / black surfaces can help to bring out depth detail.
Another factor to consider when measuring bars in this arrangement is that the bars on top will appear as one long bar but the ones underneath may be perceived as multiple short bars, like the red sections on your color pointcloud image.
An alternative approach may be to pick start and end points on the image manually to measure between them. This can be done in the RealSense Viewer tool's 3D pointcloud mode with its 'Measure' feature, or in program code.
C++
https://github.com/IntelRealSense/librealsense/tree/master/examples/measure
Python
https://github.com/soarwing52/RealsensePython/blob/master/separate%20functions/measure_new.py
-
Thank you for answer.
I have additional questions.
We proceeded to measure the steel bar to be relatively brighter than the previous experiment.
Although the diameter of the steel bar is small at 10mm,
A depth map was derived from a vertically placed steel bar, while a depth map was not derived from a horizontally placed steel bar.
What causes this?
-
You can see that the horizontal bars are light blue, which shows that depth detail is actually being generated for those bars. This color shade is less visible compared to the orange / green shades.
As the color scheme is coloring the front-most bars orange / green and the further-back bars light blue, this makes me wonder if a red-to-blue depth colorization scheme is being applied to the depth data, with colors close to the red spectrum (such as orange) representing the near distance and blue representing further distance from the camera.
Using a monochrome (black to white) coloring scheme may make all of the bars more visible.
Please sign in to leave a comment.
Comments
3 comments