19pm, as you prepare to go home, your boss sends you an email asking you to transcribe into Word the text of an article that he took a photo of with his smartphone… gulp. Long and laborious work, but it is not inevitable!
For several years, the tools of Optical character recognition (ROC in French, OCR in English) have made enormous progress. These tools work in stages to recover everything written on an image (or a PDF):
- Pre-analysis of the image to improve its quality: straightening, contrast, edge detection, etc.
- Segmentation: to identify lines of text in the image
- Character recognition
- Post-processing using linguistic and contextual methods to correct as many interpretation errors as possible
- Generating the output format.
Online OCR is as its name suggests a online ROC tool. This free tool works in 46 languages including French and as it is online, it does not require any installation.
To use it is very simple and requires only 3 steps. Click on "File" to locate the file (PDF, JPG, BMP, TIFF, GIF) you want to convert on your computer. Then choose the language of the text on the file and an output format: Word, Excel or text file. Finally, press "Convert" to start the work.
The processing will take a few seconds and the site will show you a preview of the work done.
Finally, it is possible to create a free account which will give you more features: conversion of PDFs of more than 15 pages, large images, ZIP archives…
In short, even if the tool is not perfect, it is very effective and deserves to be among the favorites of your favorite browser!
