You can import an image directly into a copper layer: ![enter image description here][1] Click `Insert image to PCB`: ![enter image description here][2] [1]: /editor/20171002/59d15e0f6c848.png [2]: /editor/20171002/59d15e7a5e320.png

The problem is that the imported image just merges into the copper layer, as it had no outline. It basically becomes an indistinguishable part of the mask.

Don't quite understand what you mean. Can you post an example?

Let's assume this is the PCB. ![Top][1] ![Bottom][2] This is the resulting PCB picture. ![Top (picture)][3] ![Bottom (picture][4] Note that both planes are Ground and set to: Pad connection: Spoke Keep island: Yes Fill style: Solid [1]: https://i.imgur.com/Ju9Yrgq.png [2]: https://i.imgur.com/SFA9ZDC.png [3]: https://i.imgur.com/fwHzeva.png [4]: https://i.imgur.com/kkG0oDv.png I hope it makes it more clear.

Have you checked the Gerbers to make sure that what the PCB photoview is showing is what the actual board will look like?

I have just now with the EasyEDA internal gerber view and it doesn't show up. It should be on the rightmost part of the PCB between LED1 and D17 https://gerber-viewer.easyeda.com/showcase/#!id=22a069eca82111e79ed5026a86b9cae7&type=bottom&layer_list=1-2-3-4-5-6 Thanks for your time.

OK, I think I understand the issue: * If an image is imported into a copper layer, there needs to be at least a `Net` attribute available to allow the copper representing the image to be assigned to a net so that it could be assigned to a *different* net from the surrounding copper. Then it will have a clearance set all the way round it. * Another possibly useful attribute would be to be able to assign it a status of `Solid`, `Cutout` or `NPTH` in the same way that can be applied to a **Solid Region**. Maybe what should happen is that an imported image is somehow converted to being represented in the PCB editor as a Solid Region? * At present, could you live with the image being on the silkscreen layer(s) only? BTW, did you know that you have two instances of LED1 and LED2 on the board? On the right hand side: ![enter image description here][1] and on the upper left: ![enter image description here][2] [1]: /editor/20171003/59d368905a2f2.png [2]: /editor/20171003/59d368cf5630a.png

> If an image is imported into a copper layer, there needs to be at least a Net attribute available to allow the copper representing the image to be assigned to a net so that it could be assigned to a different net from the surrounding copper. Then it will have a clearance set all the way round it. Another possibly useful attribute would be to be able to assign it a status of Solid, Cutout or NPTH in the same way that can be applied to a Solid Region. At the current state it's not possible. Will it anytime soon or ever be implemented? > Maybe what should happen is that an imported image is somehow converted to being represented in the PCB editor as a Solid Region? I guess so, it would totally work in my case. > At present, could you live with the image being on the silkscreen layer(s) only? Absolutely. > BTW, did you know that you have two instances of LED1 and LED2 on the board? Yes that's intended so I could use one or the other.

I messed up with the blockquote and I can't edit my post. My replies are the last sentences of each paragraph.

I'll try to make sure the conclusions of this thread are understood as the feature(s) being requested. May be a while before they get into a released version of EasyEDA. In the meanwhile, just import your image into the Silkscreen layer instead of copper. * Putting an image into copper has a downside anyway because it may adversely affect the return current paths in a power or ground plane.

The return paths warning was duly noted. I'll use the Silkscreen for the time being. Thanks for your time and patience.

I have created a dedicated Feature Request for the attributes we have discussed: https://easyeda.com/forum/topic/Assign_attributes_to_an_image_imported_into_a_copper_layer-O4YR93JDV

That sums everything up perfectly. Thanks for showing how to properly fill a feature request.