Scan text automatically

We've added a public Text Scanner model to our Neural Networks page to help you detect and read text in your images automatically. In this guide, we'll take a quick video tour of how to use the feature, followed by step-by-step instructions for setting up your own text scanner.

Getting started

Before we can start effortlessly pulling text from images and documents, we'll need to get three quick setup steps out of the way:

  1. Turn on the public Text Scanner model in the Neural Networks page.
  1. Create a new bounding box class that contains the Text subtype in the Classes page of your dataset. You can optionally add an Attributes subtype to distinguish between different types of text.

📘

Class types

The text scanner model can only be mapped to bounding box classes.

  1. Add a model stage to your workflow under the Settings page of your dataset. Select Text Scanner model from the dropdown list, and map your new text class. If the model stage is the first stage in your workflow, select Auto-Start to send all new images through the model automatically.

That's it! Now you're ready to kick back and let the model detect and read text so you and your team don't have to.

Text scanner will detect and read text on any kind of image, whether it's a document, photo, or video. As it is extensively pre-trained, it will be able to read characters that humans may find difficult to interpret.

Languages Supported: All

Alphabets supported: Latin, Cyrillic, Japanese, Chinese (Simplified)

Orientations supported: Rotated, curved, skewed text.

Fonts supported: All, including handwritten text

To add more automation to your labelling workflow check out this guide on how to train a Neural Network in V7, and this guide on how to use Neural Networks to automatically label your data.