Jump to content

Recommended Posts

Straight-up newbie question from Christovw66 again folks,

 

I've attached two reference images to help show the nature of this how-to question of mine.

 

Pictured is a hand drawn image of a logo concept emailed to me by a friend. Using the donut shape tool I added the outer ring for contrast, and here is my question...

 

...what tool or which operation(s) can I use in AD to clean up the edge of the inner, solid black circle (seen in the close-up)? This inner black circle is also obviously not a perfect circle, which is the other thing I need to correct with this image...how best would I approach this?

 

The logo appears to already be in vector form as the above mentioned 'rough-edge' does not pixelate upon zoom in.

 

I appreciate any/all guidance on this, as I realize the questions are quite rudimentary. Honestly, I've tried to figure it out on my own - not being lazy, just have a shallow knowledge of AD ops at this point.

 

Regards,

-Christo 

post-49603-0-54240200-1487534222_thumb.png

post-49603-0-06220200-1487534259_thumb.png

Share this post


Link to post
Share on other sites

What is the file type? If it is a vector you should be able to clean it up w. a moderate amount of work. Click on the image, and switch to the node tool to see what can be cleaned up.


iMac 27" Retina, c. 2015: OS X 10.11.5: 3.3 GHz I c-5: 32 Gb,  AMD Radeon R9 M290 2048 Mb

iPad 12.9" Retina, iOS 10, 512 Gb, Apple pencil

Huion WH1409 tablet

Share this post


Link to post
Share on other sites

Thanks for the response gdenby...

 

The file type is .png and although the image opens up as a new document in my AD app, is visible, and has the blue selection square with rotation tool at the top center, the image does respond to my use of the node tool. I click on the Node tool, then attempt to create a node, and...nothing.

 

Know what I'm missing/doing wrong?

 

Thanks again,

-Christo

Share this post


Link to post
Share on other sites

I suppose the file is then only a bit map, tho' high quality. At this point, AD does not have an auto tracing feature. You can either hand trace it, which would be tedious, but doable. Or you can use a different app that does do tracing. I use an inexpensive one called ImageVectorizer. Inkscape, which is free, also does the job, but it is a large program, and the trace feature is a minor part. It would take some reading to know how to use it. In both cases, the results would probably need some cleaning up, given how many "jaggies" the original outlines have.. Depending on how exact the result must be, it might be faster to just redraw the image.


iMac 27" Retina, c. 2015: OS X 10.11.5: 3.3 GHz I c-5: 32 Gb,  AMD Radeon R9 M290 2048 Mb

iPad 12.9" Retina, iOS 10, 512 Gb, Apple pencil

Huion WH1409 tablet

Share this post


Link to post
Share on other sites

If the file type is PNG it's a bitmap and not a vector file, thus you need to vectorize it as gdenby said! - I used DragPotrace and SuperVectorizer (I think this is the same app as ImageVectorizer) the quick way to make SVG vector files out of it (see below), but you have to clean up the circle yourself as desired!

 


☛ Affinity Designer 1.7.1 ◆ Affinity Photo 1.7.1 ◆ OSX El Capitan

Share this post


Link to post
Share on other sites

For a different angle, you can draw a white "donut" shape between the two black circles to visually "trim" the inner circle (over-pasting).

 

Alternatively, you can use an ellipse as a vector mask for the bitmap layer, trimming the inner picture by hiding everything outside the inner image area (cutting), including the roughness.

 

I have vector scissors, so everything looks like vector paper.

Share this post


Link to post
Share on other sites

Inkscape, which is free, also does the job, but it is a large program, and the trace feature is a minor part. It would take some reading to know how to use it.

 

It would certainly take some reading to get to know the finer points, but basic stuff is pretty straightforward.

 

I used DragPotrace and SuperVectorizer (I think this is the same app as ImageVectorizer)

 

Inkscape uses Potrace. The attached SVG file is the output obtained via the default settings in Inkscape 0.92.

 

*sorry - meant 'v_kyr'*

 

There should be an 'Edit' link at the bottom of each of your forum posts.

ScreenShotHoundRoughEdgeFullLogo.svg


Alfred online2long.gif
Affinity Designer/Photo/Publisher 1.7.2.471 • Windows 10 Home (4th gen Core i3 CPU)
Affinity Photo for iPad 1.7.2.153 • Designer for iPad 1.7.2.6 • iOS 12.4.1 (iPad Air 2)

Share this post


Link to post
Share on other sites

Inkscape uses Potrace. The attached SVG file is the output obtained via the default settings in Inkscape 0.92.

 

 

I know though I didn't used Inkscape itself here, instead I used DragPotrace which is a small GUI frontend to the potrace command line tool...

 

  • post-49706-0-74773400-1487593492_thumb.jpg
  • post-49706-0-27471300-1487593531_thumb.jpg

☛ Affinity Designer 1.7.1 ◆ Affinity Photo 1.7.1 ◆ OSX El Capitan

Share this post


Link to post
Share on other sites

Inkscape uses Potrace. The attached SVG file is the output obtained via the default settings in Inkscape 0.92.

Do you know if all the zillions of nodes I see on opening the file in Affinity are something Inkscape created during the output to SVG, or something Affinity added on import of that file?

 

I don't use Inkscape but I do have a couple of 'one trick pony' vector tracing apps that use Potrace, & no matter what settings I use with them, when I open the SVG output with Affinity I always either get something like that or an unusable version with too few nodes to define the shapes.

 

Judging from the SVG file sizes & my very limited ability to parse SVG files with a text editor app, it seems like it is Potrace adding all the nodes, but I am not at all sure of that.

 

I suspect this is what the Affinity staff are referring to when they have explained that auto-trace won't be added to Designer until the developers are happy with the results -- if it worked like that & produced hundreds or thousands of unnecessary nodes, I would not be happy with it either.


Affinity Photo 1.7.2, Affinity Designer 1.7.2, Affinity Publisher 1.7.2; macOS High Sierra 10.13.6 iMac (27-inch, Late 2012); 2.9GHz i5 CPU; NVIDIA GeForce GTX 660M; 8GB RAM
Affinity Photo 1.7.2.153 & Affinity Designer 1.7.2.6 for iPad; 6th Generation iPad 32 GB; Apple Pencil; iOS 12.3.1

Share this post


Link to post
Share on other sites

@v_kyr: Do you get an excessive number of nodes using DragPotrace, like say dozens of sharp nodes on what is a straight line segment of a path?


Affinity Photo 1.7.2, Affinity Designer 1.7.2, Affinity Publisher 1.7.2; macOS High Sierra 10.13.6 iMac (27-inch, Late 2012); 2.9GHz i5 CPU; NVIDIA GeForce GTX 660M; 8GB RAM
Affinity Photo 1.7.2.153 & Affinity Designer 1.7.2.6 for iPad; 6th Generation iPad 32 GB; Apple Pencil; iOS 12.3.1

Share this post


Link to post
Share on other sites

Do you know if all the zillions of nodes I see on opening the file in Affinity are something Inkscape created during the output to SVG, or something Affinity added on import of that file?

 

I don't use Inkscape but I do have a couple of 'one trick pony' vector tracing apps that use Potrace, & no matter what settings I use with them, when I open the SVG output with Affinity I always either get something like that or an unusable version with too few nodes to define the shapes.

 

Judging from the SVG file sizes & my very limited ability to parse SVG files with a text editor app, it seems like it is Potrace adding all the nodes, but I am not at all sure of that.

 

My first screenshot (from Inkscape) seems to support your inference about Potrace being the guilty party.

 

 

I suspect this is what the Affinity staff are referring to when they have explained that auto-trace won't be added to Designer until the developers are happy with the results -- if it worked like that & produced hundreds or thousands of unnecessary nodes, I would not be happy with it either.

 

The second and third screenshots (from DrawPlus) allow a direct comparison between the Inkscape-generated SVG file and the output from the 'Logo Image Trace' option in DrawPlus.

post-8358-0-66345700-1487595891_thumb.png

post-8358-0-44606800-1487597071_thumb.png

post-8358-0-73138000-1487597086_thumb.png


Alfred online2long.gif
Affinity Designer/Photo/Publisher 1.7.2.471 • Windows 10 Home (4th gen Core i3 CPU)
Affinity Photo for iPad 1.7.2.153 • Designer for iPad 1.7.2.6 • iOS 12.4.1 (iPad Air 2)

Share this post


Link to post
Share on other sites

@R C-R

 

It depends here on settings and mainly also on the input base of the individual image, so to say how it recognizes and then can trace the given image. I sometimes get better results with that low cost (€ 0.99) SuperVectorizer, which seems to can smooth line segments slightly, though that (the later) doesn't offer much settings at all. - All in all I would say yes, potrace seems to have a tendency for producing such sharp nodes.


☛ Affinity Designer 1.7.1 ◆ Affinity Photo 1.7.1 ◆ OSX El Capitan

Share this post


Link to post
Share on other sites

One can also check how the autotrace tool would handle this instead (for SVG output) in contrast to potrace. This online service here uses the autotrace library instead!

That certainly looks much better than what I can get with Potrace, but it still seems to have more nodes than necessary, particularly on the peripheral parts that are basically circular. Maybe Affinity's developers are shooting for something even better than that?

 

Ideally, it would not be necessary to fiddle with any settings other than say a "detail" one to control how much fine detail is included, but I suspect some sort of artificial intelligence engine or human vision model would be required to determine what we see as important details vs. extraneous ones. Not to sell the very talented Affinity team short, but that may be more than they can accomplish, at least in the foreseeable future.


Affinity Photo 1.7.2, Affinity Designer 1.7.2, Affinity Publisher 1.7.2; macOS High Sierra 10.13.6 iMac (27-inch, Late 2012); 2.9GHz i5 CPU; NVIDIA GeForce GTX 660M; 8GB RAM
Affinity Photo 1.7.2.153 & Affinity Designer 1.7.2.6 for iPad; 6th Generation iPad 32 GB; Apple Pencil; iOS 12.3.1

Share this post


Link to post
Share on other sites

I messed around w. the file some more. The original .png is really jaggy. I would expect the trace feature to pick up the huge number of pixel steps. Inkscapes "simplify path" command does an admirable job of reducing the huge number of nodes. However, I'm insufficiently practiced w. Inkscape that I used the simplify command on the whole image, which rendered the text into blobs.

 

So I undid it, brought it into AD, and used the divide command to break the .svg into separate parts. Then back into Inkscape, where I could now select just those paths I wanted to auto-smooth. Managed to knock out over 50K of point data.

 

see example.

 

post-34886-0-87470400-1487604108_thumb.jpg


iMac 27" Retina, c. 2015: OS X 10.11.5: 3.3 GHz I c-5: 32 Gb,  AMD Radeon R9 M290 2048 Mb

iPad 12.9" Retina, iOS 10, 512 Gb, Apple pencil

Huion WH1409 tablet

Share this post


Link to post
Share on other sites

I throwed that older SV tracer version I have laying around here on the image, in order to see what that generates out of it. The resulting SVGs from that one look similar to what potrace produced in terms of lots of nodes ...

 

 

@gdenby

 

That looks much better after your refactoring !!!


☛ Affinity Designer 1.7.1 ◆ Affinity Photo 1.7.1 ◆ OSX El Capitan

Share this post


Link to post
Share on other sites

That certainly looks much better than what I can get with Potrace, but it still seems to have more nodes than necessary, particularly on the peripheral parts that are basically circular. Maybe Affinity's developers are shooting for something even better than that?

 

Ideally, it would not be necessary to fiddle with any settings other than say a "detail" one to control how much fine detail is included, but I suspect some sort of artificial intelligence engine or human vision model would be required to determine what we see as important details vs. extraneous ones. Not to sell the very talented Affinity team short, but that may be more than they can accomplish, at least in the foreseeable future.

 

I hope that the Affinity dev team have some finer grade approaches here and maybe overall better working tracing algorithms.


☛ Affinity Designer 1.7.1 ◆ Affinity Photo 1.7.1 ◆ OSX El Capitan

Share this post


Link to post
Share on other sites

I messed around w. the file some more. The original .png is really jaggy. I would expect the trace feature to pick up the huge number of pixel steps. Inkscapes "simplify path" command does an admirable job of reducing the huge number of nodes. However, I'm insufficiently practiced w. Inkscape that I used the simplify command on the whole image, which rendered the text into blobs.

 

So I undid it, brought it into AD, and used the divide command to break the .svg into separate parts. Then back into Inkscape, where I could now select just those paths I wanted to auto-smooth. Managed to knock out over 50K of point data.

 

see example.

 

HoundSimplified.jpg

Share this post


Link to post
Share on other sites

Good morning AD family,

 

For the sake of practice, as well as what seems in my case to be simplicity, I've decided to redraw this image. I'd appreciate your input on the approach I am going to attempt described below. Thanks in advance.

 

I've decided to trace around the silhouette of the hound-dog image and 'excise' or remove it (can I save as an asset?), then recreate a new black background circle with the white 'light beam' subtracted. 

 

After this new background has been created, I would then like to paste the hound-dog head image in place over the newly created background. I'm not concerned with the 'jaggies' regarding the dog's spots, etc therefore see no need to redraw those inner details.

 

I guess my question is a 2-parter:

 

1) what tool do I use to isolate and remove the hound-dog's silhouette so that I can copy/paste into an Asset file and or another document...

 

...and, 2) what would your advice be on the best way to punch out said white light-beam from a solid black circle created using the Circle tool for example. I played with the Pie tool, but wasn't sure able to figure out how to manipulate/move the center-point of the pie shape's cut-out beyond the center point in order to lengthen the pre-cur wedge. My first thought was to create a black circle, then use create an elongated triangle using the Triangle tool and simply push out the desired 'light-beam' wedge from the black circle using Boolean ops but I'm going wrong somewhere because I've been unable to switch on the 'Subtract" icon (or any of the Boolean op icons/switches) for some reason...they are grayed out in both the tool bar and the dropdown menu.

 

Thanks all for your time and expertise!

-Christo 

Share this post


Link to post
Share on other sites

AD family,

 

Can one one of y'all guide me through the process of 'cutting out' the image of the hound dog so that I can then copy/paste it into a true, perfect circle (which I've created in a different document)?

 

Thanks all!

-Christo

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×