| Author |
 Replies: 10 / Views: 2,332 |
|
|
Pillar of the Community
United States
4037 Posts |
I've been using Helicon to render the 3D images of varieties that I've published recently. When Helicon does the stacking, it creates a depth map showing at which focal plan image within the stack an in-focus pixel corresponds to, and assigns that "height" to the pixel. The program of course has no idea about the step size used to generate the stack, and leaves the decision of how to scale a 3D rendering to the user. If the step is incorrectly chosen it will be either too shallow or too tall and the renderings will lack proper depth perspective. I've been using a depth amplification factor of "0.2" for all my renderings. I knew that this was a bit too much, but I thought 0.1 was too little, and since Helicon has no in-between I chose 0.2. Well, I finally captured a stack at a fairly extreme angle in order to see how far off I am with my renderings. Below I show the following images: - The straight-on stacked image of a BIE, 1954-S B624 - My usual 0.2 factor rendering at around 70-75 degrees - A stacked image of the similar BIE taken at 72-degrees tilt - A rendered image with 0.1 factor     Three things to note about these images: 1. Lighting on the 72-Deg stacked image was different by necessity, so the appearance is not the same, but for Z-calibration does not make much difference. 2. The 72-Deg stacked image required 101 images across a fairly large Z-height of around 1mm. Due to the nature of stacking, the resulting image is "telecentric", ie it has no perspective shift front to back of the stack. This makes the vertical lines in the image lack convergence and gives a bit of an unnatural look to the image, sort of like using a view camera on a building to make its walls parallel. 3. The 3D rendered images have what I consider a bit too much perspective convergence, but I have not figured out how to adjust this in Helicon. From this calibration experiment I have concluded I will render the images with an amplification factor of 0.1 for all future 3D work. I'm also going to look into how to adjust the perspective rendering in Helicon to make a less-convergent look. |
|
|
|
Pillar of the Community
2087 Posts |
interesting. I haven,t gone into stacking a deep as you have. I was a little dissatisfied with some of the results I had as the pictures just reminded me of radar topography images. What you have now informed me is probably the depth map was wrong. I suspect to get accurate( as opposed to synthesized) life-real stacked images I am going to have to set the actual increments in the stack shot ensuring they match the depth map options in Helicon. (means I will have to learn more about what the actual depth of field of the objectives are(no more guesstimates)
|
|
Pillar of the Community
United States
1247 Posts |
|
|
Moderator
Canada
10456 Posts |
"Discovery follows discovery, each both raising and answering questions, each ending a long search, and each providing the new instruments for a new search." -- J. Robert OppenheimerContent of this post is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License. See: http://creativecommons.org/licenses...0/deed.en_USMy eBay store
|
|
Pillar of the Community
 United States
4589 Posts |
You might try an experiment with some objects of known height. Something like spark plug gauges - which come in a variety of very small thicknesses, down to 0.004" or even smaller...
-----Burton 50+ year / Life / Emeritus ANA member (joined 12/1/1973) Life member: Numismatics International, CONECA Member: TNA, FtWCC, NETCC, EveryCountry (online) coin club Owned by three cats and a wife of 40+ years (joined 1983) Author: 3rd Edition of the Sample Slabs book, https://www.sampleslabs.info/ |
|
Pillar of the Community
United States
3546 Posts |
Quote:
...spark plug gauges... consistently repeatable depths, best bang for the bucks, and difficult to deform unintentionally...  |
|
Pillar of the Community
 United States
4037 Posts |
I have no problem determining the actual height of the die break, as I have calibrated my stacking unit step sizes. For example, I can tell you that the height of the BIE die break (field to top of "I") is 225 microns. The problem is getting this represented visually in the rendering so the height matches the actual coin visually. Helicon has no idea of actual heights...it can only render height proportionally to length and width.
Contact me for photographic equipment or visit my home page at: http://macrocoins.comEdited by rmpsrpms
01/05/2015 10:15 am
|
|
Pillar of the Community
United States
4589 Posts |
Contact them and ask for the feature to be added or even a special build.
As a software developer I can tell you many of these things aren't in the specs, so the developer picks a few values w/o any real understanding of the use case. In terms of the implementation, it's probably a radio button with a couple values feeding a function with that parameter, if so (especially if you can beta test to validate it), it's not a big deal to add another value.
-----Burton 50+ year / Life / Emeritus ANA member (joined 12/1/1973) Life member: Numismatics International, CONECA Member: TNA, FtWCC, NETCC, EveryCountry (online) coin club Owned by three cats and a wife of 40+ years (joined 1983) Author: 3rd Edition of the Sample Slabs book, https://www.sampleslabs.info/ |
|
Pillar of the Community
United States
4037 Posts |
I'm not sure what I could ask them to help with this problem. It's a manual process that must be done visually. What I could use is a finer step size on the slider, but I would still need to adjust the slider manually/visually.
Contact me for photographic equipment or visit my home page at: http://macrocoins.com |
|
Pillar of the Community
United States
4589 Posts |
You indicated that "I've been using a depth amplification factor of "0.2" for all my renderings. I knew that this was a bit too much, but I thought 0.1 was too little, and since Helicon has no in-between I chose 0.2." so you would be asking to have that changed/enhanced...
-----Burton 50+ year / Life / Emeritus ANA member (joined 12/1/1973) Life member: Numismatics International, CONECA Member: TNA, FtWCC, NETCC, EveryCountry (online) coin club Owned by three cats and a wife of 40+ years (joined 1983) Author: 3rd Edition of the Sample Slabs book, https://www.sampleslabs.info/ |
|
Pillar of the Community
United States
4037 Posts |
OK, I thought you were talking about some other function. I think I need to do an experiment to see if I am doing too small a step size. I do occasionally see a misinterpreted point or two on the depth map, and this may be due to the stacker thinking that a point is in focus at the wrong plane. Larger steps may mitigate this, but also may make the 3D effect less accurate. Sounds worthy of an experiment and another post. If I can double the step size, I will gain some granularity in the vertical amplifier settings.
Contact me for photographic equipment or visit my home page at: http://macrocoins.com |
| |
Replies: 10 / Views: 2,332 |
|
Posted 01/04/2015 2:11 pm
