How to extract text from an image

Snapping or clicking an image is the easiest way to capture text from paper documents conveniently in your phone or computer.

Imagine having a bunch of handwritten notes that you need to organize for a project, or a bunch of receipts that you want to digitize to better track your expenses.

While storing text as an image is convenient, you can’t readily modify, copy or edit the text in an image. You’d typically extract the text from the image to get a digital version that you can then easily edit on your computer or mobile device.

Copying or extracting text from an image is quite an easy process today, with tools that can even recognize handwriting, complex tabular data and check boxes. Such tools leverage machine learning algorithms and computer vision techniques to read/capture text from images.

In this article, you’ll learn how to easily extract text from image files in a few seconds.

Let’s look at 4 quick methods of converting an image into editable text using Adobe, Microsoft Word, Google Drive and Nanonets.

By first converting an image into a PDF file, you can copy text from it pretty easily in some cases.

1. Pick an appropriate image to PDF converter from Adobe Acrobat online – e.g. the JPG to PDF converter (supported image file types include JPG, PNG, BMP, and more).
2. Click “Select a file” to upload your image, or drag and drop it onto the converter.
3. Click open the downloaded PDF file.

You can now copy the text from the PDF.

Convert a picture to text on Microsoft Word

Converting an image to text in Microsoft Word also involves an intermediary step of converting the file to a PDF format.

1. Add or drop the image into a Word document.
2. Click File >> Save As >> and select the PDF option – this will save the file as a PDF.
3. Now again, click File >> Open >> and select the PDF file that you just saved in the previous step to open it in a new Word file.

Microsoft Word will automatically detect the text in the PDF and display it as editable text on the new Word document created in step 3.

Google Drive allows you to open any image (or PDF) file on Google Doc, thus rendering the text in an editable Doc format.

1. Upload your image on Google Drive.
2. Right-click the file >> Open with >> Google Docs.

It may take a while but you’ll eventually get a Google Doc with both the original image file and the extracted text in an editable format.

OCR software, such as Nanonets, use advanced Optical Character Recognition capabilities to extract text from pictures/images and documents.

This goes beyond the basic OCR that comes as part of the methods covered above. It can extract text from documents and images pretty accurately – even ones with complex data formatting. Such OCR software can not only maintain the original formatting of the text in the image, but also extract just the structured data that you need.

Here’s how you can convert image to text using Nanonets:

1. Upload or automatically ingest images from emails, cloud storage services, support tickets, and just about any data source.
2. Extract text or data accurately with advanced AI-powered OCR extractors that don’t rely on predefined templates.
3. Export clean structured data as XLS, CSV, or XML etc. or push data into your CRM, WMS, or database directly.

Why convert images to text?

Extracting text from images is a pretty common requirement – both for personal and business use cases. Here are a few reasons why converting an image document to text might be beneficial:

• Textual data in digital format is more convenient to store, edit, organize, search or even copy.
• Copying text from images is a much more efficient alternative to manual data entry – especially when dealing with images with lots of complex tabular text or handwritten data.

Additionally when using a software (such as OCR) for image to text extraction, you can process multiple images simultaneously or in batches thus saving a lot of time and effort.

How to ensure accurate text conversion from an image

Here are a few things to keep in mind while selecting the most appropriate image to text extraction method for you and minimising any potential rework:

• The image or picture needs to be clear with legible text – blurred or dark images with tiny non-standard text fonts might affect accuracy
• Try to maintain a standard orientation for the images – skewed images might against affect the accuracy of the text extraction
• The file size of images shouldn’t be Too large or too small – e.g. Google Drive ideally recommends image files smaller than 2MB
• If maintaining the original text formatting from the image is crucial, then select an appropriate method for you – not every image to text conversion method can guarantee this!
• Always review the extracted text – or a sample at least – for accuracy. While simple text extraction is pretty straightforward, errors can occur with images of more complex documents (invoices, bank statements, contracts etc.).