topic Workflow to get .PNG/.SVG into LT with editable lines / text in AutoCAD LT Forum

Workflow to get .PNG/.SVG into LT with editable lines / text

SandboxSolarLLC — Wed, 07 Aug 2024 04:49:37 GMT

Hey there,

I'm using a program to output electrical diagrams in .png / .svg.
From there, I'm using reaConverter to convert .svg --> .dxf.
Good news: The lines come out fine and are ready to edit. (IE: In the image, I could select red line > explode > modify)

Bad news: The text comes out as hatches and un-editable.

I need AutoCAD LT to somehow recognize these hatches as text.
I'm open to using a different workflow to go from .png/.svg --> in AutoCAD LT with editable lines + recognized text

Visual of hatch-that-I-wish-was-text... I wish I could have CAD recognize this as "10 IN BRANCH"

Any ideas on how to revise my workflow / leverage tools to get to where I want to go?

Re: Workflow to get .PNG/.SVG into LT with editable lines / text

paullimapa — Wed, 07 Aug 2024 05:49:14 GMT

Can your program output to pdf?

It's possible that with LT's PDFImport the text might be recognized if they're generated using Truetype fonts:

AutoCAD LT 2025 Help | PDF Import Settings Dialog Box | Autodesk

Re: Workflow to get .PNG/.SVG into LT with editable lines / text

pendean — Wed, 07 Aug 2024 12:09:49 GMT

You need a true raster-to-vector converter that also has built-in OCR ability @SandboxSolarLLC
I'm not aware of any free ones, but them again if this is a dire business need then it's time to purchase one or more.

Or subscribe to full AutoCAD: the RasterDesign add-on is free with subscription and it has a crude basic OCR tool that does a decent job too
https://help.autodesk.com/view/RSTR/2025/ENU/?guid=GUID-6791E749-FDE6-4E8E-9DBD-F118A80FCE4A

Re: Workflow to get .PNG/.SVG into LT with editable lines / text

SandboxSolarLLC — Sun, 11 Aug 2024 05:42:40 GMT

So, after some testing... here's the best workflow that I have.

In Figma, you can export as .PNG/.JPG/.SVG/.PDF. So, I am working with a text-filled frame on top of an image-filled frame. This is cumbersome (and something I'd love to eliminate) to have two overlaid frames per electrical diagram.

So, I export the text frame as a .PDF and then the image frame as an .SVG. (Gif to visualize superimposed frames: https://i.imgur.com/OWIw7FL.gif )

IMAGE LAYER

For the Figma image output file, the .SVG, I'm using a .bat file with reaConverter PRO's cons_rcp.exe, which is its command-line processing application. It converts .SVG into .DXF files, while preserving colors, which I need as electrical diagrams have red/green/white/black to differentiate grounding, L1, L2, etc.

TEXT LAYER

Once the .DXF is output via the command-line processing, I open that and then open (via IMPORT since no dialog box like PDFIMPORT) the Figma output .PDF, which is then automatically scanned and inserted as "PDF_Text" layer. As @paullimapa suggested, the .PDF export contains TrueType, so I think that's how this is accomplished.

So, at this point, what would be amazing would be a way to output ONE file (either .SVG or .PDF) in Figma...
And have some tool accurately separate the two layers (image and text).
Here's one almost-solution that I spent a night on; unfortunately, the geometry came out blurry.
(https://stackoverflow.com/questions/71054750/how-to-make-a-vector-pdf-searchable)
Here's another almost-solution, the Pro version of Scan2CAD, which has a nice OCR and the geometry was solid.
Too bad it's $80/month or I'd have gone with it.

That's where I'm at. I wish wish wish AutoCAD LT had more OCR options.
I'm open to suggestions on how to get a .SVG/.PNG/.JPG/.PDF --> .DXF with text as text and vectors as lines