3D scanning maximum error SR300

[VViti]

Good morning,

I need to build a 3D scanning system. I'm thinking of using your SR300 camera and I would like to know in mm the maximum error you make in the size of the object that will be scanned.

Thank you

Vitiello

[MartyG]

I would say that the maximum object size is determined by a balance between the size of the object and how close that object is to the camera. Ideally when scanning, you want the object to fill up the camera's view as much as possible but not go outside of that view.

The larger the object is, the further back you would have to position the camera to fit it all into the camera's view. The default maximum scanning range of the SR300 is 1.5 m, though you can extend the depth range to 4 m with scripting at the expense of image quality (the further away you extend the depth scan with the scripting instruction, the more noise the image will have).

[VViti]

Thank you for your answer. You're right, I have not been clear to the question. The object I need to scan is 20cm from the camera and the object measures about 35cm.

What is the average error of the SR300?

V.Vitiello

[MartyG]

The minimum range of the SR300 is 0.2 m (20 cm), and the minimum range of the R200 is 30 cm.

The SR300 is suited to small objects and the R200 is suited to larger objects and full human bodies. 35 cm likely falls into the 'large' category that the R200 camera would be more suited to. You should bear in mind though that the R200 is older technology than the SR300 and no longer supported in the RealSense SDK, so an appropriate alternative to the R200 may be the new ZR300 camera.

The company 3D Systems, whose Sense system utilizes RealSense, had this advice: "Before scanning a small object or one without a lot of distinct features (such as a standard coffee mug), add some items around the object. This will help the scanner maintain tracking by giving it more to lock on to."

[VViti]

Ok, but the size of the object that was scanned , on average, how much it may differ from the real? Thank you.

[MartyG]

I understand now. There was a case recently where the user was concerned that their scanned object (a book) differed from the scale of the real-life object by a few cm.

After some detailed research carried out by Intel support member Sergio, it was reported by Sergio about the attempt to scan a book that "We observed that the uniformity of the object makes edge detection difficult. We did have slight improvements in performance in the measurements under certain conditions. These conditions include doing the measurements at different distances, backgrounds, and lighting conditions."

The user said that the book was a test example and they would actually be scanning more complex objects such as arm and leg anatomy. In that case, Sergio concluded with the suggestion that the user open an email support case so that further personal assistance could be provided by email instead of on the forum. The link for opening a support case was:

https://supporttickets.intel.com/%3Flang%3Den-US https://supporttickets.intel.com/?lang=en-US