Sunday, July 23, 2017

screenshot & OCR & copy paste google books & comparison converter image to pdf word onenote evernote



This post is in french & english... Sorry.
Désolé ce post est en français et anglais.

L’OCR, sigles de Optical Character Recognition est une technologie conçue pour analyser la structure de l’image du document et puis diviser le document en caractères, tableaux et images et aussi équations (sur une ligne)... Ensuite, le programme peut reconstituer la mise en page et les éléments en un texte modifiable.

Le convertisseur PDF est également capable de transformer un document PDF en Word, Excel ou d’autres formats éditables. Mais ce genre de logiciel ne permet que de changer le format de fichier sans analyser le contenu du texte ou sans extraire les caractères. Autrement dit, après la conversion de PDF, vous recevez quand même un fichier non modifiable contenant des images.

Ici nous intéressons à la rapidité de traitement lorsque l'on fait une copie d'écran d'un site qui protège le texte (produit du texte visualisable en image, ou comme google books ou les livres/articles en streaming).

If you don’t have a scanner, your phone or a camera can be used in place of one. There are a number of apps in the App Store (iOS mac)  android win GNU/linux that “scan” physical documents into clean PDFs using a phone’s camera.
Evernote's Scannable is just one of several apps that allows you to scan documents with your smart phone (see below).

Comparison of OCR softs

https://en.wikipedia.org/wiki/Comparison_of_optical_character_recognition_software

5 best commercial OCR softs

Abby FineReader

v14 windows and only v12 mac 12.1.6 supported macOS 10.12 Sierra (120€)
perpetual licenses
199€ (Convert PDFs and scans; Edit and comment PDFs)
or 299€ (and Compare documents  and Automate conversion 5000pages/month)

FineReader 14 offers new and enhanced features including wide range of PDF editing tools, document comparison feature and even more perceptive OCR for document conversion. Plus, faster performance across the board — from launching app to opening and converting.
PDF editing
Create PDFs from Word, Excel, and other formats
Retain text styles when exporting to Word
Single-line math formulas support
https://www.abbyy.com/en-eu/finereader/whats-new/comparison-charts/

10€ for this small utilities
https://www.abbyy.com/fr-fr/screenshot-reader/
and alternatives http://alternator.link/software/abbyy-screenshot-reader

ABBYY Real-Time Recognition SDK for mobile apps.
Extracting text directly from the smartphone’s preview screen is significantly faster and more convenient than taking a picture of the text followed by an OCR step

ABBYY Mobile Imaging SDK
The toolkit to analyze photos of documents captured with mobile devices to determine whether they are suitable for OCR or enhance their visual quality.

PDFELEMENT

59€
mac/PC
Autre point fort de l’outil, il est capable de rendre les formulaires numérisés modifiables et d’exporter les données numérisées au format CSV.

http://www.leptidigital.fr/technologie/logiciels-ocr-8915/

Omnipage

Omnipage 18 windows XP to 10
99€
educ: 69€

SDK win, linux, mac
http://www.nuance.fr/for-business/by-product/omnipage/for-developers/index.htm
http://www.nuance.fr/for-business/by-product/omnipage/for-developers/csdk-for-mac/index.htm

Readiris

Readiris 16
99€
mac/pc/iOS
OSX 10.9, 10.10,10.11 and 10.12
http://www.irislink.com/EN-US/c1466/Readiris-16-for-Macintosh---OCR-Software.aspx

Create indexed PDF files
This format is identical to the original PDF file except that it adds a layer of indexed text. This means you can easily search for the information you need in a document or a large volume of archived documents.
SDK
http://www.irislink.com/EN-US/c1096/iDRS-15.aspx

Skysoft PDF Editor 6 Professional for Mac or Win

commercial 99€
https://pdf.iskysoft.com/6std-vs-6pro.html

Renee PDF

par un français et gratuit mais que windows et pas open source
https://www.reneelab.fr/product-land-337.html

Ref

https://pdf.wondershare.fr/pdf-editing-tips/pdf-ocr-software.html

Prise de note, de scan, de photo et OCR limité

Greenshot

free soft and open source 
only windows
http://getgreenshot.org/
Greenshot is a light-weight screenshot software tool for Windows with the following key features:

Quickly create screenshots of a selected region, window or fullscreen; you can even capture complete (scrolling) web pages from Internet Explorer.
Easily annotate, highlight or obfuscate parts of the screenshot.
Export the screenshot in various ways: save to file, send to printer, copy to clipboard, attach to e-mail, send Office programs or upload to photo sites like Flickr or Picasa, and others.
...and a lot more options simplyfying creation of and work with screenshots every day.

Being easy to understand and configurable, Greenshot is an efficient tool for project managers, software developers, technical writers, testers and anyone else creating screenshots.

Screenshot to Speech

6€
http://macgenius.co/app/Screenshot-to-Speech/955098738
Poor OCR recognition

picatext

osx 10.12
4€
You need to get the text out of an image or from anywhere on your screen? With picatext this is as easy as selecting a file or a screen area. So now you can get the text you need in almost no time. No need to tediously typewrite everything again.
picatext is OCR made simple. We included just the functionality you need. Nothing that might distract you. That helps focusing on the important things and saves you a lot of time.
These are the awesome features of picatext:
- get text from images
- extract text from your screen
- supports the recognition of over 40 languages
- start screen selection by shortcut
- text gets copied to clipboard automatically
- set the font of the output text
http://www.macdrifter.com/2013/10/picatext-for-image-ocr-on-mac.html

Prizmo

50€
https://computers.tutsplus.com/tutorials/5-ways-to-ocr-documents-on-your-mac--mac-49683
https://creaceed.com/prizmo
https://itunes.apple.com/app/id546392952
Prizmo is a dedicated OCR app. It isn’t designed to help you crop or straighten your scanned documents, it’s sole purpose is to analyse the text of any scans and convert it into searchable text. It’s not just limited to plain text documents such as receipts and correspondence, Prizmo will even analyse old newspapers and magazines, book covers, pretty much anything with any shape, size and colour of text is something Prizmo will convert.
Prizmo includes the ability to capture scans directly from Image Capture, OS X’s built-in camera and scanner import app, so you can use it in conjunction with any existing scanner. You can also import existing files if you’ve been scanning them already.

Skim Clip

very good software.
It's the best for screenshot and built-in OCR.
http://www.domloveskim.com/software/skim-clip/#download
http://kimlovingnews.blogspot.fr/2012/08/skim-clip-11-headed-to-mac-app-store.html (the last post=dec 2012).
https://maczot.com/discuss/2012/11/23/skim-clip1353651425/

oldies dec 2012; shareware 10$. only MAC >10.6. Works on sierra 10.12.
The trial version is limited to 10 minutes of use at a time and has a 10 second delay at start up.

It uses tesseract (which support for a number of new image formats was added using the Leptonica library http://www.leptonica.com/). Leptonica is a pedagogically-oriented open source site containing software that is broadly useful for image processing and image analysis applications.
In addition to the freeform selection, we've added (v 1.1) a few more features such as playing a sound when clipping, an option to remove all headers from the exported output, and the ability to use a script to filter the OCR text before it is saved as a clipping. This last feature, filtering, is particularly interesting because it opens up many work flow possibilities as you may apply your understanding of UNIX and scripting tools like sed (https://en.wikipedia.org/wiki/Sed) and Perl to perform all kinds of creative transformations and processing on the scanned text.
Skim Clip is a unique new utility for OS X that combines Optical Character Recognition (OCR) with screen capture features to create a powerful and convenient new way of saving and organizing your information.
Using Skim Clip, you can capture the contents of any window, scan and save all its text including words in images, and store it into organized categories - all with a single keypress.
With the clipping converted into scanned computer text, you can instantly search thousands of clippings to find what you're looking for. Like a photographic memory with a search engine, Skim Clip makes it easy to keep, find, and reuse anything that's ever on your screen!

Features (v 1.0) include:
* Instantly clip with a single hot key from any application
* Fast and accurate OCR text conversion of any screen
* Scales to easily manage thousands of clippings
* Very quickly search OCR text, notes, and more
* Reuse your clipping by emailing, saving, and pasting them
* Present your clippings with slideshows and quicklook
* Create clipping categories and subcategories
* Export dozens of clippings in a single click
* Browse your clippings as thumbs or a listing
* Get visual feedback and categorize anywhere with the status menu
* Set your own hot key, export style, and more
* Automated custom scanned text processing
* Integrate Skim Clip into your workflow with AppleScript
* Double tap the hot key to instantly add a note for extra convenience
* Elegant interface to help keep your clips organized
* Keep focused on your work with Skim Clip always ready
* Optimized OCR for English with support for other languages

Version 1.1:
Clip anywhere on the screen by holding the clip hot key for freeform selection
Custom filtering of scanned text for advanced processing
Automate clipping with AppleScript
Improved image processing for better accuracy
Sound effect when clipping
Option to remove headers when exporting
Increased mail export compatibility
Plain text...

We've been invited to share the following comments from one Jim Miller, author of the well-regarded Rabbi Hawkins Mystery Series. Here's what Mr. Miller has to say about Skim Clip:
Skim Clip is amazing. I’m usually disappointed by Apps that claim to do everything, including saving the world, and then do nothing at all. That is not the case with Skim Clip. It does exactly what the developer says it will do. It is a screen capture and OCR combo that works.
Another plus for this program is the obvious time and effort that has gone into making the user manual simple and easy to understand. I was able to walk through the step-by-step instructions and capture a variety of screen shots. I do a lot of screen shots with the Mac’s commands, but they all end up as jpegs. With Skim Slip, I get the jpeg and if I choose, I can convert it into text and then insert it into Word, Pages and Scrivener very easily.
This is a timesaver I love. I used to cut and paste a captured image into a word processor or have it up on the screen so that I could then type the content into some other program just to make it useable. Not anymore. With Skim Clip, it’s all done for me.
Many thanks to Dominic and Kimberly,  (DomLovesKim Company)
Jim Miller
http://kimlovingnews.blogspot.fr/2012/10/from-author-using-skim-clip.html

Only for Intel, 64-bit processor, Mac OS X 10.6 or later, 64-bit processor
put the text OCR in the copy-buffer.

I try this soft on MAC 10.12.5 on chrome and others browsers :  it's the fastest.
Some minor pb with OCR when characters are tiny, just expand your window or try increasing your application's font size.
If you want another soft (but also oldies 2013), use PDF OCR X Community Edition (it also uses this Tesseract code) is a simple drag-and-drop utility for Mac OS X and Windows, that converts your PDFs and images into text documents or searchable PDF files. You must save the screenshot to a file (img) and drag-and-drop to it and copy the text. Then Skim Clip is faster and far better.

LEADTOOLS OCR App Review and SDK

iOS and android
https://www.leadtools.com/blog/document-imaging/leadtools-ocr-app-review/

OneNote

use Microsoft OneNote's built-in OCR functionality (free for Windows/Mac) ; it's in Pack Office.

menu
copy img
(right clics) copy picture to text

https://webapps.stackexchange.com/questions/57418/how-can-i-copy-paste-segments-from-google-books
In OneNote, users can enter typed text via keyboard, create tables, and insert pictures. However, unlike a word processor, users can write anywhere on a virtually unbounded document window by just clicking there. Also, users do not need to explicitly save their work – OneNote saves data automatically as the user works.
Microsoft also announced a number of new features in OneNote that use the service API:
OneNote Clipper: A browser bookmarklet that uses the OneNote service API and enables users to save a screenshot of a webpage to OneNote along with the link. The text in the screenshot is made searchable using Optical Character Recognition
https://en.wikipedia.org/wiki/Microsoft_OneNote

Evernote

As well as the keyboard entry of typed notes, Evernote supports image capture from cameras on supported devices, and the recording of voice notes. In some situations, text that appears in captured images can be recognized using OCR and annotated. Evernote also supports touch and tablet screens with handwriting recognition. Evernote web-clipping plugins are available for the most popular Internet browsers that allow marked sections of webpages to be captured and clipped to Evernote. If no section of a webpage has been highlighted, Evernote can clip the full page. 
Where suitable hardware is available, Evernote can automatically add geolocation tags to notes.

In 2010, the coding language for the suite was changed from C# for version 3.5 to C++ in version 4.0 to improve performance.
On supported operating systems, Evernote allows users to store and edit notes on their local machine, using a SQLite database in Windows.

The Evernote software can be downloaded and used as "stand-alone" software without using the online portion of an Evernote account (online registration is required for initial setup, however), but it will not be able to upload files to the Evernote server, or use the server to synchronize or share files between different Evernote installations. Also, no image or Image-PDF (Premium only) recognition and indexing will take place if the software is used entirely offline.

In June 2016, Evernote announced the limitation for users of its free Basic account to two devices per year and raise prices for its premium service tiers.Non-paying Evernote user is able to sync notes between two devices.

https://en.wikipedia.org/wiki/Evernote
https://help.evernote.com/hc/fr

Evernote does OCR images so they're searchable,  but the text is not extractable from the image.  OneNote is better for that specific use case - the text (or the best approximation of it) is available in a separate window to be copied and pasted elsewhere.  Voice to text is part of Evernote,  though you'd need to use one of the specialist pen inputs if you want to choose specific parts of the text to hear again.
in fact no OCR (may be with premium and for some specific img (buz card)).


Evernote with other apps:
Skitch is a free screenshot editing and sharing utility for OS X (iOS, Windows, and Android). The app permits the user to add shapes and text to an image, and then share it online. Images can also be exported to various image formats. Originally developed by Plasq, Skitch was acquired by Evernote on August 18, 2011. On December 17, 2015, Evernote announced that it will be ending support for Skitch for Windows, Windows Touch, iOS, and Android on January 22, 2016. Evernote said it will continue to offer Skitch for Mac and  iPad iPhone.
https://evernote.com/intl/fr/products/skitch

8 choices:
  1. Capture Viseur 
  2. Zone de capture précédente
  3. Capture viseur différé 
  4. Capture plein écran 
  5. Capture fenêtre... 
  6. Capture menu...
  7. Capture avec |'appareil photo... 
  8. Ouvrir une Image ou un PDF...

Scannable
If you're a die-hard Evernote user, Scannable may be the best option for you...assuming you use an iOS device, that is. Scannable on the iPhone and iPad allows you to instantly scan business cards, sketches, receipts, paper documents, and even multi-page documents with ease, and automatically file and organize the resulting images and files in your Evernote account. Your scans are automatically cropped to remove backgrounds (like the table behind the recipt, for example) and enhanced so the text is readable. If you scan a business card, the contact information from the card is automatically lifted and added to a contact card, so you can call the person, email them, or visit their website or social presence with a single tap. It's relatively new, and while it works best with Evernote, it also allows you to share your resulting scan with other apps on your iOS device, so you can upload it to Dropbox, email it, or save it to your camera roll. Best of all, it's completely and totally free—no unlocks, no premium version, although it definitely works best if you also use Evernote.
It is the business card wingman that knows the card is significant only because of the information it holds. While the paper is trivial, its words are full of potential. Once you connect to Evernote and LinkedIn, Scannable immediately digitizes every business card you scan. It takes that information and adds in details from your contact’s LinkedIn profile, including a photo, so you’ll always be able to put a face to the name. Instantly save the person to your contact list and feel free to ditch the paper entirely (we won’t tell anyone).
Scannable captures the paper in your life quickly and beautifully, transforming it into high-quality scans ready to save or share. Whether on the go or at the office, send paper on its way and move on.
Evernote's Scannable is just one of several apps that allows you to scan documents with your smart phone.
only for iPad & iPhone.

penultimate
Penultimate is the award-winning digital handwriting app for iPad that combines the natural experience of pen and paper with power of Evernote’s sync and search features. Lose the paper, keep the handwriting.
only for ipad
https://evernote.com/intl/en/products/penultimate

Evernote Web Clipper
Save webpages forever
From interests to research, save anything you see online—including text, links, and images—into your Evernote account with a single click.
plug-in for many browsers
https://evernote.com/intl/en/products/webclipper

evernote inc  and github
https://github.com/evernote

 Evernote and OneNote

comparison between 

Google OCR and drive

Google Drive includes a little-known free OCR tool that is a powerful, easy to use image to text converter.
Google Drive currently supports OCR for .jpg, .gif, .png, and PDF files up to 2MB in size

To add your documents to Google Drive to get them OCRed, first go in to your Google Drive account. On the My Drive page, click the My Drive button next to the New button on the left side of the page. Select Upload Files.
Use the Upload files option to upload a file to Google Drive. Find the file that you want to convert from PDF or image to text. Click the Open button. 
The document now appears in your Google Drive. Right-click on the document to bring up a drop-down menu.
Click the Open with option and click Google Docs. A sheet icon appears while the file is downloading. Google is now in the process of converting your PDF or image file to text with OCR.
The file will open in Google Docs with the PDF converted to text, but with little formatting applied. You can now edit and format the new text file as you like.
You will get:


Tutorial:

Free and Open soft

https://turtlapp.com/docs/
no OCR

Free Open source OCR

Comparison of the 3 open source OCR

Tesseract 

Tesseract has unicode (UTF-8) support, and can recognize more than 100 languages "out of the box".
Tesseract supports various output formats: plain-text, hocr(html), pdf, tsv, invisible-text-only pdf.
You should note that in many cases, in order to get better OCR results, you'll need to improve the quality of the image you are giving Tesseract.
Tesseract does various image processing operations internally (using the Leptonica library) before doing the actual OCR. It generally does a very good job of this, but there will inevitably be cases where it isn't good enough, which can result in a significant reduction in accuracy.
Installing Tesseract
You can either Install Tesseract via pre-built binary package or build it from source.
Supported Compilers are:
  • GCC 4.8 and above
  • Clang 3.4 and above
  • MSVC 2015, 2017
The Tesseract engine was originally developed as proprietary software at Hewlett Packard labs in Bristol, England and Greeley, Colorado between 1985 and 1994, with some more changes made in 1996 to port to Windows, and some migration from C to C++ in 1998. A lot of the code was written in C, and then some more was written in C++. Since then all the code has been converted to at least compile with a C++ compiler. Very little work was done in the following decade. It was then released as open source in 2005 by Hewlett Packard and the University of Nevada, Las Vegas (UNLV). Tesseract development has been "sponsored" by Google since 2006.
https://en.wikipedia.org/wiki/Tesseract_(software)
Tesseract is executed from the command-line interface.
Basic command line usage:
tesseract imagename outputbase [-l lang] [--oem ocrenginemode] [--psm pagesegmode] [configfiles...]
For more information about the various command line options use tesseract --help or man tesseract.
GUIs and Other Projects using Tesseract OCR
While Tesseract is not supplied with a GUI, there are many separate projects which provide a GUI for it. One notable example is OCRFeeder (end 2014).
https://github.com/tesseract-ocr/tesseract/wiki/User-Projects-%E2%80%93-3rdParty

PDF OCR X (uses this Tesseract code) is a simple drag-and-drop utility for Mac OS X and Windows, that converts your PDFs and images into text documents or searchable PDF files. It uses advanced OCR (optical character recognition) technology to extract the text of the PDF even if that text is contained in an image. This is particularly useful for dealing with PDFs that were created via a Scan-to-PDF function in a scanner or photo copier.
mac/win
0€=Limited to single-page images and PDFs (pro=29€ for unlimited pdf size)
http://solutions.weblite.ca/pdfocrx/

other projects with tesseract code

  • Online OCR services
  • Android/iOS
  • ocr-fileformat is a converter: Validate and transform between OCR file formats (hOCR, ALTO, PAGE, FineReader)
  • Audiveris is an open-source Optical Music Recognition software which processes the image of a music sheet to automatically provide symbolic music information in MusicXML standard.
  • Wolfram Mathematica 9.0 use tesseract for recognizing text
  • OCRmyPDF - Adds OCR text layer to scanned PDF files and images, allowing them to be searched. Processes pages in parallel on multi-core CPUs. Keeps exact resolution of original embedded images without recompressing JPEGs, when possible. Includes image several preprocessing options, detailed documentation, and support for many exotic PDFs.
  • Tally-Ho: Tally-Ho is a screen reader intended for sites like google books
    https://code.google.com/archive/p/tallyho/
    Tally-Ho is a screen reader intended for sites like google books (though it will also read anything copied to the clipboard). google books is a great site, especially if you're a Ph.D. student - however, for people who are visually impaired or dyslectic (or, like my gf, both...), it is quite inaccessible. Tally-Ho uses the Tesseract OCR engine (thanks google!) to convert the screen capture to text, so it can read graphic based sites such as google books. tested on windows xp sp2 w/at&t natural voices & vista w/microsoft
    End of dvpment 2008
  • IMPACT is a project funded by the European Commission. It aims to significantly improve access to historical text and to take away the barriers that stand in the way of the mass digitisation of the European cultural heritage. END = 2012 and other forks...
    http://www.impact-project.eu/
  • MathOCR Features mathematical formula recognition and logical layout analysis, can use OCR engines like Tesseract or Ocrad as back-end. https://github.com/chungkwong/MathOCR
    END=2015.

GOCR

END=2013

Cuneiform

https://en.wikipedia.org/wiki/CuneiForm_(software)
END=2011

web services OCR


No comments:

Post a Comment