Calibrated depth map?

Optical microscopes + digital photography
Post Reply
vaster04
Posts: 3
Joined: 14.10.2019 17:53

Calibrated depth map?

Post by vaster04 » 14.10.2019 18:15

Hello

I'm using microscope with stepmotor on focus knob. So, I know the distance between stack images along optical axis of microscope, stack step. After processing by Helicon Focus I can obtain depth map. Depth values are distributed between 0 and 255. Is it possible to establish relationship between these values and known stack step, to obtain calibrated depth map?

Thank you for your help

Valeriy

User avatar
Catherine
Posts: 142
Joined: 29.04.2019 22:38

Re: Calibrated depth map?

Post by Catherine » 16.10.2019 13:23

Yes, it's possible. If А is a starting point, and B is the end point, the depth map step is (B-А)/255

vaster04
Posts: 3
Joined: 14.10.2019 17:53

Re: Calibrated depth map?

Post by vaster04 » 16.10.2019 16:20

Catherine,

Thanks a lot for your valuable information. It is very important, it seems it should be placed somewhere in Helicon Focus manual, tutorial.

Please, could you inform, is it possible to calibrate depth data, extracted from vertices of *.obj file?

Best regards

Valeriy

User avatar
Catherine
Posts: 142
Joined: 29.04.2019 22:38

Re: Calibrated depth map?

Post by Catherine » 21.10.2019 12:32

Depth map can be both obtained from .obj file and saved as a black&white image (exported depth map). The min value (0) corresponds to the first image of the stack, the max value - to the last one. If you know the distance between frames, the depth map can be easily calibrated.

vaster04
Posts: 3
Joined: 14.10.2019 17:53

Re: Calibrated depth map?

Post by vaster04 » 23.10.2019 16:51

Catherine,

Thanks for your reply.

>The min value corresponds to the first image of the stack, the max value - to the last one.
Is this statement correct, if in stack a few first and a few last frames are out of focus? Perhaps, is more correctly to assign to depth step between frames value (B-А)/255?

>Depth map can be both obtained from .obj file and saved as a black&white image (export depth map).
Thank you. Situation with exported depth map file is clear, you have explained it in your reply of 16/10/2019. In this file we have 255 depth levels. But in *.obj file depth data have 4 significant digits, so depth data are quasi continuous and their calibration is not obvious.

Probably, they could be calibrated by comparison of min and max data from gray depth file and assign min and max values of *.obj file to those values, but I don't sure that this is the best calibration way.

Best regards
Valeriy

User avatar
Catherine
Posts: 142
Joined: 29.04.2019 22:38

Re: Calibrated depth map?

Post by Catherine » 01.11.2019 15:23

We recommend to use the depth map as a primary source. Data in obj file is formed based on the depth map. Even if the first and the last frames of the stack are out of focus, the depth map is still normalized by the number of files in the stack.

Post Reply