How to Get Solid Tone from a pixelated Sample Area?

[rixtertrader]

I was not sure how to title this so that others could easily find this thread in the future if need be. Hopefully, I'm able to ask the question in a clear manner.

Thanks to Eggy, I now have an image that is toned based on the black value being 0. Awesome!

Please note on the image attached that there are "shades" marked 2h, hb, 2b, etc. These are pencil shades. I will call them "shade swathes" from here out.

I shaded these as completely as possible before scanning. Unfortunately, they don't scan in as a flat tone/color, but are quite pixelated with even white speckles (paper?) showing through.

This makes it difficult to determine the tone value using the eye dropper, because the eye dropper could land on any of those different shade pixels that exist within each of those shade swathes.

Is there a way, within Photoshop, where I can make these solid, flat tones?

I want to make it easier to compare these "shade swathes" with the tones on the Value Finder also shown in the image.

Thanks.

 

[thebestcpu]

Hi rixtertrader

If all you are trying to create a flat consistent tone for each swath, all you have to do is one by one select within the boundaries of the swath (the area you want averaged) and then use the filter Filter > Average

That will average out all pixells within the selection with the end result is an area that is a solid consistent tone that is the average of all the pixels from within the selection.

I am not 100% sure that the eye will "average" the swatch the same way as a mathematical average yet thought this might be a starting point.

Hope this is of some use

John Wheeler

 

[rixtertrader]

Hi rixtertrader

If all you are trying to create a flat consistent tone for each swath, all you have to do is one by one select within the boundaries of the swath (the area you want averaged) and then use the filter Filter > Average

That will average out all pixells within the selection with the end result is an area that is a solid consistent tone that is the average of all the pixels from within the selection.

I am not 100% sure that the eye will "average" the swatch the same way as a mathematical average yet thought this might be a starting point.

Hope this is of some use

John Wheeler

That's a good tip John. It's a start.

Unfortunately I think the "average" filter is also including the "white" speckles within these swaths, causing the resulting tone to be lighter than it should be.

I did a search on replacing the "white" with "transparency" in the hopes of removing the white speckles from the equation, but only managed to find information on replacing the white 'background' only. The white specks remained.

I found information on replacing a color with another color, but not with transparency.

I've attached an image with the swaths enlarged. Note the white specs.

Thank you for your comments!

 

[Eggy]

What you want isn't actually possible...(until proven wrong...)

The general tone appearence of one of those lines is a mix of dark and light pixels caused by the combination of coarseness of the used medium and pencil pressure.
That combination gives an artist the possibility to achieve shades using only one pencil on a particular medium.




You can also compare it with a two-tone print where the size variation and density of the dots size determines the shape.

 

[rixtertrader]

What you want isn't actually possible...(until proven wrong...)

The general tone appearence of one of those lines is a mix of dark and light pixels caused by the combination of coarseness of the used medium and pencil pressure.
That combination gives an artist the possibility to achieve shades using only one pencil on a particular medium.




You can also compare it with a two-tone print where the size variation and density of the dots size determines the shape.

Thank you Eggy. This is what I had suspected, but am looking for a solution in order to further help my drawing students.

Attached is one of my pencil drawings in graphite.

In the sample swaths provided in previous images, and in this drawing, the pressure is light. The density is achieved by "layering". This allows for removal of graphite for corrections and tone modifications.

With this approach, we achieve different tones by using different softness of pencils, such as 2h, 4b, etc.

With pressure being constant, each gives their own "tone".

What I'm trying to do is assign a "value number" to each tone achieved, and figured Photoshop could help here.

Unfortunately, as you have seen, the paper fiber can cause white specs to appear within the swath making determination difficult (so far).

The averaging filter was a good suggestion. If only it did not include the white specs which raises the tone value (higher value moves toward WHITE).

It seems unfortunate that there isn't (as we know) to get a better indicator of TONE VALUE as a means of reference between one pencil and another.

Again, thank you for your time and help. Cheers!

 

[Eggy]

The averaging filter was a good suggestion. If only it did not include the white specs which raises the tone value (higher value moves toward WHITE).

Maybe you could remove the white by using 'color range'...(select only the whites and delete them)
Then apply the average filter...

 

[thebestcpu]

Hi Rixtertrader

There are definitely some easy ways to remove the pure white spots in the swaths in a selection and then use the average filter.

One step before that however is the swaths you provided are not representative of the image in your last post. The image in your last post has neither any pure black points nor any pure white points. Your sample swaths (either by how they were drawn, type of paper, due to scanning, and/or post processing are full of totally black pixels and even more pure white pixels.

I suggest that having better swath examples in digital form would be the right first step if the white spots should not be there in the first place. Then using the average filter would provide what you need.

This is similar to trying to fix a bad image when sometimes the fastest approach is not by post processing yet by retaking the image.

If after that, you have samples that have a small number of pure white spots (or near white spots if desired) can easily be done using Layer "Blend If" techniques. I would be happy to share that approach yet I don't think it is the primary issue.

Just a suggestion.

John Wheeler

BTW - I love the example image you provided. Great artistry.