Add OCR step to Business Process

In Work.AI, the no-code OCR step lets you accomplish the following:

• Process documents asynchronously without submitting redundant tasks to Workers and receive responses instantly in Control Tower.

• Recognize a document and get an output compatible with a labelling Manual Task.

Create Business Process with no-code OCR step

To build a BP with a no-code OCR step, follow the steps below:

1. Create a BP as described in the guide.

2. Navigate to the Workflow tab of the newly created BP. In the Reusable components panel on the right, go to the Bot tab and click Models to open the list of available no-code steps.

   

3. Add the OCR step to the Workflow Designer canvas in one of the following ways:

   • Copy an existing step. In the Reusable components panel, hover over the OCR step and click the displayed Copy button. As a result, you create an independent OCR step instance: the changes you make in the copy are not synchronized with the source.

     

   • Reuse an existing step. Drag an existing OCR step from the Reusable components panel to the canvas. In this case, any changes you make to the step are applied across all BPs where the same step is included.

4. Double-click the step on the canvas and, if you want your BP to be schema-based, navigate to the Input / Output tab of the OCR step to set up the input and output contracts. For details, read the guide.

   You do not need to set the step contracts if you set the contracts for the entire BP.

5. Go back to the Design tab and specify required parameters. For detailed parameter descriptions, see Input parameters.

   

   If you select Google Vision or Azure AI Document Intelligence as your OCR engine, check the default configuration includes the following parameters:

   • OCR Type: Enriched
   • OCR Google parameters json:
     • exportFormat: xmlForCorrectedImage,xmlWithoutRecognitionVariants
     • writeRecognitionVariants: true
     • skipPreprocessing: false
     • storePreprocessedDoc: true
     • timeout: 60
   caution

   If any of the settings are not configured as above for Google Vision or Azure, OCR throws an exception.

   

6. Go back to the Design tab and click Save again. The BP is now ready for OCRing incoming documents.

tip

For an end-to-end tutorial on creating a BP with no-code OCR, Model, and Manual Task steps, see the guide.

Set up OCR REST configuration properties

warning

The Azure, Google Vision, and Paddle OCR providers are available only from the OCR step. ABBYY is available both from REST API and the OCR step.

Property	Default value	Description
execution.bridge.task.queue	exec.<CT_HOST>.result.ocr	Name of the OCR step queue
execution.bridge.prefetch.count	100	Prefetch count of the OCR step queue
execution.bridge.thread.count	20	Count of threads for consuming messages of the OCR step queue

Configure OCR step parameters

Input parameters

Parameter	Code	Default value	Description
Document URL	document_url	original_document_url	Specify the URL of the document to be processed by OCR. Typically, the data comes from the BP input.
When the step fails	go_forward_on_failure	Complete the step and move to the next one with an error (yes)	Specify the action to be done if the failure result is received. If you choose Complete the step and move to the next one with an error (set the code value to yes) and the failure answer is received, the step is completed, and an error message is sent to the next step.
OCR type	ocr_type	Enriched	Select Standard or Enriched. In the latter case, you enable preparing the OCR output for a labelling Manual Task.
S3 bucket name	s3_bucket_name	doc-upload	Specify the S3 bucket to store the OCR output.
S3 folder name	s3_folder_name	empty	Specify the S3 folder to store the OCR output.
OCR engine	ocr_provider	ABBYY	Select an OCR provider: ABBYY Google Vision Azure AI Document Intelligence Paddle OCR
OCR parameters	ocr_parameters	See below	See the parameters in the table below.

OCR JSON sample

Below is a JSON sample with default values you can find in the OCR parameters field:

{
    "profile": "dataExtraction",
    "exportFormat": "xmlForCorrectedImage,xmlWithoutRecognitionVariants",
    "language": "English",
    "correctOrientation": true,
    "writeRecognitionVariants": true,
    "discardColorImage": false,
    "skipPreprocessing": false,
    "removeGarbageSize": 0,
    "lowResolutionMode": false,
    "enhanceLocalContrast": false,
    "changeDPI": 0,
    "storePreprocessedDoc": true,
    "textTypes": [
        "normal",
        "handwritten"
    ],
    "timeout": 240,
    "detectLanguage": true,
    "detectTextOnImage": true,
    "detectHandwritten": true,
    "detectCheckmarks": true,
    "nativePdfRecognition": false
}

OCR parameters

The table below lists all available OCR parameters and their descriptions:

Parameter	Values	Description
profile		A scenario for fine-tuning the OCR engine to do a specific task.
	documentConversion	Converts documents into editable formats.
	textExtraction	Extracts text from documents.
	barcodeRecognition	Extracts barcodes.
	dataExtraction	The dataExtraction scenario is used to extract all possible data from a document and store it in a structured way. The result represents a document structure storing all document objects: printed and handwritten text, tables, barcodes, checkmarks, and images with their location and attributes. This format is optimal for further processing, storing data in a database, or integrating with another application.
exportFormat	xmlForCorrectedImage	Represents the OCR API export format. You can define multiple export formats using a comma as a separator: xml,xmlForCorrectedImage. For correct processing of a labelling MT, you cannot use more than three export formats.
	txt	It is the default export format.
	xml	Contains characters or words along with their location in the original document (coordinates or frames).
	xmlWithoutRecognitionVariants	The same as xml but does not contain character-related data.
	xmlForCorrectedImage	The same as xml, except the location is taken from a processed or adjusted document.
	pdfSearchable
	html
language	English German so on	Defines the default OCRed document language. You can also define multiple languages using a comma as a separator: English,German,Polish.
correctOrientation	boolean	Detects the page orientation. If it differs from normal, the page is rotated automatically.
writeRecognitionVariants	boolean	For the xml and xmlForCorrectedImage formats, provides variants of characters or words that OCR considers a possible recognition result.
discardColorImage	boolean	Leaves only the black-and-white plane in a prepared image.
skipPreprocessing	boolean	Disables all preprocessing steps for a document. The default value is false. If you set to true, the convertTo and changeDPI parameters are disabled.
useDefaultPattern	boolean	If true, it requires applying the default pattern from the OCR application bundle.
removeGarbageSize	integer	Removes garbage, such as excess dots smaller than a specific size. The setting is optional, and the valid values are > 0 and -1.
lowResolutionMode	boolean	Improves the recognition of images with low resolution.
enhanceLocalContrast	boolean	Specifies whether the local contrast of an image should be increased.
convertTo	none	Automatically detects the file type and converts it to TIFF before processing. Accepts the PDF, PNG, JPG, JPEG formats.
changeDPI	integer valid range from 50 to 3200 null or 0 for disable	Specifies the resolution value, for example, changeDPI=300. The recommended resolution for a source image is 300 DPI for typical texts (10 pt or larger) and 400-600 DPI for texts in smaller fonts (9 pt or smaller).
storePreprocessedDoc	boolean	Specifies whether preprocessed documents should be saved in the data storage.
textTypes	normal handwritten typewriter matrix ocrA ocrB e13b	Specifies the type of the text in the document. You can define several text types using a comma as a separator: normal,matrix.
timeout	integer	Sets the maximum allowed timeout in seconds for a single task execution by an OCR Worker. The timeout countdown starts when an OCR Worker begins executing a task. When the task execution exceeds the given timeout, the task is forcibly killed. If you do not set the value, the default value (0) is used, and your BP does not have time to process a task.
detectLanguage	boolean	Specifies whether the OCR engine should automatically detect the recognition language of a text. Auto detection is intended to be used during the recognition of documents the language of which is not known to you.
detectTextOnImage	boolean	Specifies whether the OCR engine should automatically detect text in images.
detectHandwritten	boolean	Specifies whether the OCR engine should automatically detect handwritten text.
detectCheckmarks	boolean	Specifies whether the OCR engine should automatically detect checkmarks.
nativePdfRecognition	boolean	If set to true, all native PDFs are processed without OCR, which allows to save OCR license usage.

If native_pdf_processing is set to true, the skipPreprocessing and convertTo parameters are disabled. Such parameters as changeDPI and storePreprocessedDoc take effect only after you save pages of an input PDF.

troubleshooting

In case of issues with input data, see Issues related to input data.

Output parameters

• OCR result: document processing results in the JSON format

  • Code: ocr_result

  • Example:

    {
        "originalDocument": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/1.pdf",
        "documents": {
            "xmlForCorrectedImage": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/bp-id/task-id/1- xmlForCorrectedImage.xml",
            "xmlWithoutRecognitionVariants": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/bp-id/task-id/1-xmlWithoutRecognitionVariants.xml"
        },
        "pageDetails": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/bp-id/task-id/preprocessed/links.json"
    }

• Task ID: the ID of an OCR task

  • Code: task_id
  • Example: 066416ed-20e8-4e96-9df1-7f7e133f1211

• Meta info JSON: extended info for the labelling Manual Task

  • Code: meta_info_json

  • Example:

    {
        "pages": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/bp-id/task-id/preprocessed/links.json",
        "ocrXmlUrl": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/bp-id/task-id/1-xmlWithoutRecognitionVariants.xml",
        "imageUrl": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/1.pdf"
    }

• Process time: record process time in milliseconds

  • Code: _sys_ocr_process_time
  • Example: 24053

See an output JSON sample for https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/bp-id/task-id/preprocessed/links.json below:

[
    {
        "imgUrl": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/bp-id/task-id/preprocessed/1.png",
        "charsUrl": "https://HOSTNAME-s3-ui.cloud.workfusion.com/buckets/doc-upload/browse/bp-id/task-id/preprocessed/1.json",
        "height": 3300,
        "width": 2550,
        "charsCount": 2216
    }
]

To set your Java Native Worker to work with the OCR step data, use the respective dependency from the WorkFusion's repository:

<groupId>com.workfusion.spa</groupId>
<artifactId>bridge-step-contracts</artifactId>

Using the library facilitates the implementation and reduces the time to production for such solutions.