Hey all,
I just saw this topic post get a hit: https://www.ukvac.com/forum/threads/sanwa-29e31s-clear-schematic-needed.74102/
I see time and time again people crying out for good quality scans and single sources of truth for arcade manuals, monitor manuals etc etc. I recently posted to the Facebook page about a YouTube video on how I personally tackle this (I'm sure there's plenty of room for improvement and of course I welcome feedback). But I'd like to throw it out there if people do have originals that they are happy to either archive themselves or loan out to me to archive for them, I think it would be a nice community project.
The video is here:
The script I use to do the heavy lifting is: https://github.com/theretrobristolian/Scripts/blob/main/Doc-Archiving/Doc_Archiving_Script.ps1
My process is:
1. disassemble the document (assuming stapled)
2. Scan at 600dpi to a multi page TIF
3. Use my script to:
1 - Source - in here put all your original scanned multipage TIF files exactly as they come out of the scanner.
2 - Extracted - In here will appear all the extracted singular pages from the multipage TIF files.
3 - Deskewed - In here will appear all the deskewed (straightened) singular pages.
4 - cropped - In here will appear all the deskewed pages cropped to A4.
5 - PDFs - In here will be your output assembled, deskewed, cropped A4 PDF documents.
4. use ABBYY FineReader to do OCR on the PDF - I use this as a) it has incredibly accurate detection anyway and b) it allows you to manually tweak and correct and compare the recognised text to the original document (as shown in my video).
Finally for everyone to enjoy and use I'd upload them to archive.org and the Arcade Otaku wiki pages. I've been testing my process with the Nanao MS9 monitors and have been drafting up a page here: https://wiki.arcadeotaku.com/w/User:Theretrobristolian/MS9-29 my aim is to have a consistent layout for pages, and try and capture all the useful information in a uniform way.
I suppose the question is a) is this a process people can get behind and support and b) what is the feedback on my current efforts?
As I say above, I'm sure someone more technical, more patient, with better source documents (my MS9 manual while original wasn't the highest quality to start with) could probably do a much better job, I'm not trying to say I have all the answer, but I do see this as a seemingly on going problem and it would be great to take a stab at the problem before good documents get lost permanently with places shutting down, people retiring etc etc.
Let me know your thoughts
I just saw this topic post get a hit: https://www.ukvac.com/forum/threads/sanwa-29e31s-clear-schematic-needed.74102/
I see time and time again people crying out for good quality scans and single sources of truth for arcade manuals, monitor manuals etc etc. I recently posted to the Facebook page about a YouTube video on how I personally tackle this (I'm sure there's plenty of room for improvement and of course I welcome feedback). But I'd like to throw it out there if people do have originals that they are happy to either archive themselves or loan out to me to archive for them, I think it would be a nice community project.
The video is here:
The script I use to do the heavy lifting is: https://github.com/theretrobristolian/Scripts/blob/main/Doc-Archiving/Doc_Archiving_Script.ps1
My process is:
1. disassemble the document (assuming stapled)
2. Scan at 600dpi to a multi page TIF
3. Use my script to:
1 - Source - in here put all your original scanned multipage TIF files exactly as they come out of the scanner.
2 - Extracted - In here will appear all the extracted singular pages from the multipage TIF files.
3 - Deskewed - In here will appear all the deskewed (straightened) singular pages.
4 - cropped - In here will appear all the deskewed pages cropped to A4.
5 - PDFs - In here will be your output assembled, deskewed, cropped A4 PDF documents.
4. use ABBYY FineReader to do OCR on the PDF - I use this as a) it has incredibly accurate detection anyway and b) it allows you to manually tweak and correct and compare the recognised text to the original document (as shown in my video).
Finally for everyone to enjoy and use I'd upload them to archive.org and the Arcade Otaku wiki pages. I've been testing my process with the Nanao MS9 monitors and have been drafting up a page here: https://wiki.arcadeotaku.com/w/User:Theretrobristolian/MS9-29 my aim is to have a consistent layout for pages, and try and capture all the useful information in a uniform way.
I suppose the question is a) is this a process people can get behind and support and b) what is the feedback on my current efforts?
As I say above, I'm sure someone more technical, more patient, with better source documents (my MS9 manual while original wasn't the highest quality to start with) could probably do a much better job, I'm not trying to say I have all the answer, but I do see this as a seemingly on going problem and it would be great to take a stab at the problem before good documents get lost permanently with places shutting down, people retiring etc etc.
Let me know your thoughts
