Having said all this, I do doubt Colin can get good, shroud-like 3D from his flour-power model.
I am willing to be proven wrong.
A reader from Baltimore writes:
If you look at the illustration “3D-rendered image of plastic toy” on Colin Berry’s other blog you can see that he is incorrectly using ImageJ and as a consequence drawing false conclusions. He is using the image he is plotting as the texture. In effect he is plotting on a plot. That is a big, big NO NO. That he is doing so is obvious when you look at dark areas on the arms and legs that rise to the level of adjacent bright areas. It is also obvious because of the color we see. The best texture to use is #808080 for all pixels. Some perspective would also be helpful.
Is it a no-no? I wonder.
The image, shown below, can be found at Is a high energy laser beam really needed to model the Turin Shroud? Maybe those Italians should have tried pizza ingredients first, and a hot oven…:
Yes, I must say I agree that there is a problem here. I see it and I’ve been thinking about it for nearly a day now. I still need to do some experimenting to understand this better. But I do see that plotting without loading a texture and allowing the software to use the image itself as the texture (which seems to be the default) will lead to erroneous interpretations.
Colin follows the image in his blog with this paragraph.
Yes. one can enter 2D diagrams with no 3D history, like those concentric circls above, and they show a comparable 3D response (top left) to that of the model image OR the TS. Why is that? Look at the z scale next to the red arrow. It is on its default MINIMUM setting of 0.1. The software sets that non-zero default setting, meaning that ANY image one enters that has any kind of intensity gradient, simple stepped ones included, produce a 3D response.
What in the world is Colin saying that is not fully obvious? Back in November of 2013, I noted that:
Ray Rogers used to point out that a drop of ink on a filter paper would look like a mountain when plotted the same way (e.g. VP-8). Colin Berry is right that scorch marks and holes on the shroud produce 3D images; the scorches, obviously, are not spatial information.
Colin continues that paragraph of his, writing:
The latter is entirely artefactual unless one has evidence to the contrary. This investigator knows of no evidence to suggest that the so-called "3D properties" of the TS image are any different from those of contact imprints generally.
Colin may be right, at least to some degree. If he will post the base images (or send them to me) I will plot them with a neutral texture. If he has a paint program he can create his own; just create a rectangle the size of the image with a middle gray background, say RGB 128/128/128.
I’m looking at the following image I plotted with ImageJ and wonder if I did it right. Did I use a proper neutral texture or did I use the image as the texture? Look at the color. I probably made the same mistake. See Teaser of the Day (#3): Why many state that the Shroud is a 3D image.
I just installed Windows 10 and ImageJ won’t work for me. So be patient as I figure out what is wrong.
Having said all this, however, I do doubt Colin can get good, shroud-like 3D from his flour-power model. I am willing to be proven wrong.