---
schema_version: "1"
id: a1ed838b407cf647f641b2aadba8e651bc832083cd74528cfea8bcd601ad2195
title: Intelligent OCR (2020)
collection: mulesoft
version: current
locale: en-us
portal_url: https://docs.salesforce.com/mulesoft/current/en-us/rpa-builder/latest/toolbox-text-recognition-intelligent-ocr-2020.md
canonical_url: https://docs.mulesoft.com/rpa-builder/latest/toolbox-text-recognition-intelligent-ocr-2020
source_site: docs.mulesoft.com
content_type: documentation
content_hash: b20988d2d2996756c2284bc4ba9d0308318d19d1e0e981ce1d98a177a8cf96aa
description: MuleSoft Documentation Site
release: "6.7"
latest: true
---
---
product: RPA Builder
version: 6.7
is-latest-version: true
---

# Intelligent OCR (2020)

> For the full documentation index, see: https://docs.mulesoft.com/llms.txt

The _Intelligent OCR (2020)_ action step represents a good alternative to the previous OCR (Legacy
OCR). The new OCR contains the languages Arabic, Greek, Hebrew, and Thai.

## Properties

**Color Depth** (bpc) In documents with many images and colors, a setting with low color depth leads
to faster search results. You can select a value between 0 and 8. Value 0 means that the color depth
remains unchanged compared to the original.

**Enhance Contrast** Adjusts the image contrast so that the legibility of the text is improved.

**Image Type**

- **Auto** (default) The document type is determined automatically

- **Document** Can improve the OCR operation of a document (PDF)

- **Snippet** Can improve the OCR operation of selected text sections

**Recognition Strategy**

- **Fast** The OCR operation is carried out faster but this can lead to poorer scan quality

- **Advanced** (default) The OCR operation can take longer but the results are optimal

## Outbound Variables

The outbound variables are identical to those in the Legacy OCR.

## Wizard

**Scaling Mode**

**Auto Scale by Engine** (default) Scaling to a higher resolution takes place automatically if this
improves the OCR operation. This can lead to slightly longer search operations and also requires
more main memory.

**Scale by 1** The scaling of the recorded image is set to 1.

**Manual scale** In the _Manual scale_ mode, you can determine the scaling factor yourself. You can
enter an enlargement value of between 1 and 4 in each case. It is **not** recommended to use a high
value (4) in combination with a large image excerpt.

**Dictionary**

The following languages are available here

<table><colgroup><col> <col></colgroup><thead><tr><th>Arabic</th><th>Italian</th></tr></thead><tbody><tr><td><p>Chinese (simplified)</p></td><td><p>Japanese</p></td></tr><tr><td><p>Chinese (traditional)</p></td><td><p>Korean</p></td></tr><tr><td><p>Czech</p></td><td><p>Norwegian</p></td></tr><tr><td><p>Danish</p></td><td><p>Polish</p></td></tr><tr><td><p>English</p></td><td><p>Portuguese</p></td></tr><tr><td><p>Finnish</p></td><td><p>Russian</p></td></tr><tr><td><p>French</p></td><td><p>Spanish</p></td></tr><tr><td><p>German</p></td><td><p>Swedish</p></td></tr><tr><td><p>Greek</p></td><td><p>Turkish</p></td></tr><tr><td><p>Hebrew</p></td><td><p>Thai</p></td></tr><tr><td><p>Hungarian</p></td><td></td></tr></tbody></table>

In contrast to the situation with action step _Legacy OCR,_ here you can combine the Asian languages
with each other, and also with the other languages.

You can select a maximum of two languages at any time.

**Timing**

**Retries \[0..5\]** (Default: 3) With this menu option, it is possible to repeat the search
recognition up to 5 times. If the text recognition is successful, no further attempt is carried out.
This option only appears if you have selected text, Regex, or fuzzy comparison in search mode.