vflat scanner

[विश्वासो वासुकिजः (Vishvas Vasuki)]

From speaking to a leader of kannaDa gaNaka pariShat, I learned about their very satisfactory experience with using vFlat scanner. 

https://play.google.com/store/apps/details?id=com.voyagerx.scanner&hl=en_IN&gl=US

https://apps.apple.com/in/app/vflat-scan-pdf-scanner/id1540238220

It seems they've successfully used it  + crowdsourcing to rapidly scan large works despite the limit of 100 pages/ user-day .

Supposedly it handles 2-column text well, and exports pdf as well as text well. Besides camera, it seems to import pdf as well.

--

--
Vishvas /विश्वासः

[विश्वासो वासुकिजः (Vishvas Vasuki)]

As per https://t.me/vflat_india/595 - they use Google vision. " We want to provide unlimited OCR but since we use Google API and use premium model not the free on,"

[Pushpahasa NS]

Yes vflat is good i have seen people using this. Yesterday I recommended KSU people to use it. As per users review this is 99% accurate on most fonts 

--
You received this message because you are subscribed to the Google Groups "sanskrit-programmers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sanskrit-program...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sanskrit-programmers/CAFY6qgGbWN175dZpdM4CoWB6sTTDOmVHWfhAVpvpUCsjQRofEQ%40mail.gmail.com.

[विश्वासो वासुकिजः (Vishvas Vasuki)]

vlat's autoscan - . Just amazing.

To view this discussion on the web visit https://groups.google.com/d/msgid/sanskrit-programmers/CAD8GmKr1S5LHr-EpkgmiGyQ5K-G79cxHaWq2wOjHbhD%3Dz_jOBA%40mail.gmail.com.

[विश्वासो वासुकिजः (Vishvas Vasuki)]

Can someone suggest a mobile phone holder arm so that I can keep the phone steady while scanning books with vflat? (Should be non-made-in-China [Taiwan is ok] and should be available in India. The ones on amazon.in claiming to be made in India are most likely just Chinese imports and show Chinese models in images.)

[P S Ramachandran]

You received this message because you are subscribed to the Google Groups "samskrita" group.
To unsubscribe from this group and stop receiving emails from it, send an email to samskrita+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/samskrita/CAFY6qgG6Br2VJ7Xv3VJ%2BQq5fT%2BiX2%3DF4OAuv4JOMDiph-YNCwQ%40mail.gmail.com.

[विश्वासो वासुकिजः (Vishvas Vasuki)]

That just looks like an Indian reselling a Chinese product. Pure dandho. If you have firmer info, please let me know.

To view this discussion on the web visit https://groups.google.com/d/msgid/sanskrit-programmers/CAD1xEtdccMQHq36y11mZrbogjtswjSq4z0sYj-dBFNvzwDT9qQ%40mail.gmail.com.

[Vinod Appajanna]

Yes, Google's Vision API is by far the best I have seen (in terms of accuracy). I have used the Vision API's (DOCUMENT_TEXT_DETECTION) to convert more than 2000 pages.

Here is the python code https://cloud.google.com/vision/docs/pdf#vision_set_endpoint-python that I have used in my automation.

If you have a google cloud account, the initial 1000 pages are free I guess. After that they start charging. https://cloud.google.com/vision/pricing (refer: Document Text Detection feature).

If someone needs any help let me know.

Regards,

Vinod

To view this discussion on the web visit https://groups.google.com/d/msgid/sanskrit-programmers/CAFY6qgF%2BKCLjiocNrKCwO1Q5HmuVkwas9p0mc_RRP9syFjBQBA%40mail.gmail.com.

Disclaimer:- This footer text is to convey that this email is sent by one of the users of IITH. So, do not mark it as SPAM.

[विश्वासो वासुकिजः (Vishvas Vasuki)]

In case others need it -

Make	Model	Description	Unit Price (INR)	Qty	Line Total (INR)
IPEVO	Uplift	Multi-angle Arm for Smartphones Hold up your phone without holding up your phone.	INR 6,500.00	1	INR 6,500.00

 

Pricing terms         = Ex-Works BLR. This is PARTNER PURCHASE Price (not end user price)

Payment terms     = 100% in advance (Before delivery).

GST                      = Additional 18%

HSN Code            = 7616 9990

Delivery Terms     = Maximum of 2~3 weeks from date of Official PO.

Country of Origin  = TAIWAN

[विश्वासो वासुकिजः (Vishvas Vasuki)]

Automated scanning seems also available with camscanner - https://youtube.com/shorts/KFLCF_sU40A

[विश्वासो वासुकिजः (Vishvas Vasuki)]

On Mon, 30 Sept 2024 at 06:29, P S Ramachandran <ramu.v...@gmail.com> wrote:

Viswasji, 

Very valuable information Viswasji.

I use vflat free version only, through my android or apple phone, by using a normal phone stand, a little crude way only, and shift the book, which is kept against the wall, in an angle, and after the whole book is scanned and then transfer the images to desktop,

That's a good way to bypass their 10 pics/ month pdf restriction - thanks for the tip! Perhaps external apps would do as well ...

What's a "normal phone stand"?

 

and then check the quality of images, and rescan the misscanned pageś and add to the folder (noting the image file name, while checking, and delete the ’not properly scanned ones thereafter’ and once again check the whole folder’ and thereafter use small pdf or similar online resources  to pdf it.  I rarely textify the images, as the free version offers only 10 images per month. But scanning there is no limit, and pdfing there is limit; but once you transfer the images from phone to desktop, through cable, you can use online resources like smallpdf to pdf it and compress it, etc.   

Dhanyavadaḥ.

Rāmu

ऐकमत्य सूक्तम्   Rg. Veda. Book 10 - 191 (1-4).

संस॒मिद्यु॑वसे वृष॒न्नग्ने॒ विश्वा॑न्य॒र्य आ । इ॒ळस्प॒दे समि॑ध्यसे॒ स नो॒ वसू॒न्या भ॑र ॥

सं ग॑च्छध्वं॒ सं व॑दध्वं॒ सं वो॒ मनां॑सि जानताम् । दे॒वा भा॒गं यथा॒ पूर्वे॑ संजाना॒ना उ॒पास॑ते ॥

स॒मा॒नो मन्त्र॒: समि॑तिः समा॒नी स॑मा॒नं मन॑: स॒ह चि॒त्तमे॑षाम् । स॒मा॒नं मन्त्र॑म॒भि म॑न्त्रये वः समा॒नेन॑ वो ह॒विषा॑ जुहोमि ॥
स॒मा॒नी व॒ आकू॑तिः समा॒ना हृद॑यानि वः । स॒मा॒नम॑स्तु वो॒ मनो॒ यथा॑ व॒: सुस॒हास॑ति ॥

saṁsa̲midyu̍vasē vr̥ṣa̲nnagnē̲ viśvā̍nya̲rya ā | i̲ḷaspa̲dē sami̍dhyasē̲ sa nō̲ vasū̲nyā bha̍ra ||

saṁ ga̍cchadhvaṁ̲ saṁ va̍dadhvaṁ̲ saṁ vō̲ manāṁ̍si jānatām | dē̲vā bhā̲gaṁ yathā̲ pūrvē̍ saṁjānā̲nā u̲pāsa̍tē ||

sa̲mā̲nō mantra̲: sami̍tiḥ samā̲nī sa̍mā̲naṁ mana̍: sa̲ha ci̲ttamē̍ṣām | sa̲mā̲naṁ mantra̍ma̲bhi ma̍ntrayē vaḥ samā̲nēna̍ vō ha̲viṣā̍ juhōmi ||

sa̲mā̲nī va̲ ākū̍tiḥ samā̲nā hr̥da̍yāni vaḥ | sa̲mā̲nama̍stu vō̲ manō̲ yathā̍ va̲: susa̲hāsa̍ti ||

 

“Agni, showerer (of benefits), you who are the Lord, you alone among the gods pervade all living beings; and you give the sacrificer all the objects (phalāni) for which he prays

 “Meet together, talk together, let your minds apprehend alike; in like manner as the ancient gods concurring accepted their portion of the sacrifice.”

“Common be the prayer of these (assembled worshippers), common be the acquisition, common the purpose, associated be the desire. I repeat for you a common prayer, I offer for you with a common oblation.”

“Common, (worshippers), be your intention; common be (the wishes of) your hearts; common be your thoughts, so that there may be thorough union among you.”      

You received this message because you are subscribed to the Google Groups "samskrita" group.
To unsubscribe from this group and stop receiving emails from it, send an email to samskrita+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/samskrita/CAFY6qgFo82jy0WEN-q6%2BTWB5a5_qZ8x%2B0G_snGtm%2B7Hrv5z%3Dcw%40mail.gmail.com.

[विश्वासो वासुकिजः (Vishvas Vasuki)]

Actually vflat now has a limit on 2-page scans!

So camscanner has a non-automated 2 page scan mode - which is more usable.

[विश्वासो वासुकिजः (Vishvas Vasuki)]

Some tips from sheshadri -

Convert old books, printed forms, pages etc to editable, fillable format.
https://www.youtube.com/watch?v=4lApeCNK6Ic
1. Google OCR:
Upload file to Google Drive, right click and open with Google Docs (which has OCR).
➢ Use flatbed scanner to scan pages. Best, use Adobe Scan Mobile App, in Book mode to scan books.
➢ Ensure the pages are not bent at the edges while scanning.
➢ Full / big PDF files don't get OCR'd properly, the file limit may be 10Mb.
➢ Using Acrobat Pro OR PDFMate PDF Converter OR Universal Document Converter save the original multipage file as multiple
image files (File -> Export To -> Image). Then upload all the pages.
➢ With PDFMate, all the pages of the pdf file get saved as image files under a directory of the same file name. You can upload the
entire directory to your Google Drive in one go.
➢ Or using Acrobat Pro combine the image files into a single PDF file (Tools -> Combine files), then split the file into multiple small
files of 9Mb or less (Tools -> Organise Pages -> Split), then upload all the files to Google Drive.
➢ For multi column image files, using Paint, draw vertical lines to separate multi columns, save and combine/upload.
➢ OR using MS Window's Snipping Tool, snip each page, save it as Jpeg file and upload to Google Drive.
➢ OCR setting in Google drive:

Tick
Convert
Uploads:
Convert
uploaded
files to
Google
Docs
editor
format

The converted file will have the original image on top, followed by editable text. Once converted, save the file for editing using
Azhagi+ or BhashaIME etc.
File >> Download as >> MS word file.
2. Mobile apps:
- Smart Lens Android app directly gives output in editable format.
- You can also try Ocr Scanner, Google Lens, Pramukh OCR etc.,
3. Sanskrit OCR: This Sanskrit OCR can be used for all languages. The best feature is that you can copy & paste the image directly.
4. PIOCR: Piocr is too good, especially with multi column pdf/image files. Using Paint, draw vertical lines to separate multi columns,
save and upload the image for best output. Not a free package, but 5 free scans per day.
5. EasyScreenOCR: Good output for Tamizh input. Does not support other Indic languages.
6. vFlat: Good mobile paid app.
7. Splitting double side to single side scan doc. Use online2pdf.com or other such tools and then OCR.

[Seshadri Nadadhur]

camscanner is chinki? docscanner?

my DIY  "abnormal phone stand" for OCR.

--------------
Warm regards
Seshadri