This service allows you to get text from the image by optical character recognition (OCR).
For OCR, you need to upload the image first. Then choose the language in which the text is written. After that you'll get a result.
Important tips for the best result:
— upload only good quality images
— text orientation should be horizontal.
Wayback machine says that service started around 04.07.2014. But we do not know the exact day. For the details, please, go to Revisions History. We also have a lot of plans about the functionality, but we will be happy and your ideas that you can send us .
ToS page will be published later.
Tesseract is used for recognition process. The site is written on PHP using MVC. It has its own non-public API. We plan to create unique algorithms to improve recognition processing.
Several VPS servers are in charge of requests processing. During peak loads additional servers are connected.
- 0.2 The option to choose different languages for recognition was added.
- 0.2.2 Another server was added + minor edits.
- 0.3 The page "My queries" was added + possibility to view and download original file.
- 0.3.1 The new text of the page "About Service" was added.
- 0.3.2 11 languages for recognition were added: Dutch, Czech, Danish, Greek, Estonian, Finnish, Latvian, Lithuanian, Romanian, Slovakian, Slovenian.
- 0.4 Multi-language interface was created (Russian, English and Ukrainian), code refactoring; site design was improved.
- 0.4.1 Transition to a secure connection HTTPS was completed.
- 0.4.2 Another 10 recognition languages were added: Azerbauijani, Bulgarian, French, Hungarian, Norwegian, Portuguese, Spanish, Serbian, Swedish, Turkish.
- 0.4.3 Three new interface languages were added: Italian, Spanish and German
- 0.4.4 The statistic research completed. Now we know the most frequently uploaded file formats. BMP was added. PDF is in progress.
- 0.4.5 Polish was added.
- 0.4.6 French was added as interface language. AJAX processing and Cli requests were revised.
Recommend SMS Gateway.