Version: 10.2.4

Configure OCR Bridge step and OCR providers

OCR Bridge is introduced to improve:

Performance. The new OCR Bridge step reduces throughput due to processing documents asynchronously without submitting redundant tasks to Control Tower workers and receiving responses instantly in Control Tower.
Usability. Only one step is needed to recognize a document whose output is compatible with the ToD Manual Task.

Architecture

Control Tower handles the OCR Bridge steps in a particular way. It redirects input data to the OCR Bridge queue without involving the CT worker. After the document is processed, CT consumes the response from the result queue.

An OCR REST application can work with both REST and Bridge queue interfaces. It validates incoming tasks, checks and consumes licenses, stores tasks in DB with the correct status.

OCR REST configuration properties

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

OCR bridge step

Input parameters

Parameter	Code	Default value	Description
Document URL column name	`document_url_column_name`	``original_document_url`\| Document URL column name for processing by OCR.
OCR Type	`ocr_type`	`enriched`	OCR Type values: Standard and Enriched (prepare output for TOD Manual Task).
S3 bucket name	`s3_bucket_name`	`doc-upload`	An S3 bucket to store results.
OCR Provider	`ocr_provider`	Abbyy	Type of the OCR provider: Abbyy Google Vision Azure Form Recognition
OCR parameters JSON	`ocr_parameters`	See later	See parameters in the table below.

OCR JSON. Default values

{
    "exportFormat":"xmlForCorrectedImage,xmlWithoutRecognitionVariants",
    "language": "English",
    "correctSkew": true,
    "correctOrientation": true,
    "writeRecognitionVariants": true,
    "invertImage": false,
    "discardColorImage": false,
    "skipPreprocessing": false,
    "useOnlyCustomRegions": false,
    "useDefaultPattern": false,
    "profile": "documentConversion",
    "removeGarbageSize": 0,
    "lowResolutionMode": false,
    "enhanceLocalContrast": false,
    "allowedRegionTypes": null,
    "changeDPI": 0,
    "version": "V_2",
    "storePreprocessedDoc": true,
    "textTypes": ["normal"],
    "timeout": 60
}

OCR parameters

Parameter	Values	Description
`exportFormat`	txt `xml` `xmlWithoutRecognitionVariants` xmlForCorrectedImage pdfSearchable html	Step's export format.
`language`	English German , and so on	Predefined language. You can also define multiple languages using a comma as a separator: English,German,Polish
`correctSkew`	boolean	The page skew is detected and automatically corrected.
`correctOrientation`	boolean	The page orientation is detected and if differs from normal, it is rotated automatically.
`writeRecognitionVariants`	boolean	Causes the `xml`/`xmlForCorrectedImage` formats to include all variants of character or word, OCR considered as a possible recognition.
`invertImage`	boolean	Inverts image.
`discardColorImage`	boolean	Leaves only black-and-white plane in the prepared image.
`skipPreprocessing`	boolean	Skips preprocessing stage. It increases performance up to 30%.
`useOnlyCustomRegions`	boolean	Skips original analyzing stage and extracts information from custom regions only.
`useDefaultPattern`	boolean	If `true`, it requires to apply the default pattern from the OCR application bundle.
`profile`	documentConversion	Converts documents into editable formats.
	`textExtraction`	Extracts text from documents.
	`barcodeRecognition`	Extracts barcodes.
`removeGarbageSize`	integer	Removes garbage: excess dots that are smaller than a certain size from the image. Optinal, valid value > 0 and -1 for automatically detect garbage size.
`lowResolutionMode`	boolean	Improves recognition of images with low resolution.
`enhanceLocalContrast`	boolean	Specifies whether the local contrast of the image should be increased.
`allowedRegionTypes`	`BT_Table` `BT_Text` `BT_Barcode` `BT_VectorPicture` `BT_Separator` `BT_SeparatorGroup` `BT_Checkmark` `BT_CheckmarkGroup`	Deprecated. Specifies allowed region types for identified blocks classification. NOTE: narrowing down type of regions can break page layout. Do not use it if you’re not sure you need it.
`changeDPI`	integer [50 : 3200] valid range null or 0 for disable	Specifies the new value for the resolution.
`version`	`V_2`	ABBYY FRE 12
	`V_1`	ABBYY FRE 11
`storePreprocessedDoc`	boolean	Specifies whether preprocessed documents should be saved into data storage.
`textTypes`	normal typewriter matrix ocrA ocrB e13b	Specifies the type of the text in the document. This parameter can also contain several text types separated with commas, for example `normal,matrix`.
`timeout`	integer	Sets a timeout in seconds for task execution.

Output parameters

OCR Result. Document processing result in the JSON format.

Code: ocr_result

Example:

{
    "originalDocument": "https://minio-host/doc-upload/1.pdf",
    "documents": {
        "xmlForCorrectedImage": "https://minio-host/doc-upload/bp-id/task-id/1- xmlForCorrectedImage.xml",
        "xmlWithoutRecognitionVariants": "https://minio-host/doc-upload/bp-id/task-id/1-xmlWithoutRecognitionVariants.xml"
    },
    "pageDetails": "https://minio-host/doc-upload/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 ToD Manual Task.

Code: meta_info_json

Example:

{
    "pages": "https://minio-host/doc-upload/bp-id/task-id/preprocessed/links.json",
    "ocrXmlUrl": "https://minio-host/doc-upload/bp-id/task-id/1-xmlWithoutRecognitionVariants.xml",
    "imageUrl": "https://minio-host/doc-upload/1.pdf"
}

Process time: record's process time in milliseconds.
- Code: _sys_ocr_process_time
- Example: 24053

Page details json example https://minio-host/doc-upload/bp-id/task-id/preprocessed/links.json

[
    {
        "imgUrl": "https://minio-host/doc-upload/bp-id/task-id/preprocessed/1.png",
        "charsUrl": "https://minio-host/doc-upload/bp-id/task-id/preprocessed/1.json",
        "height": 3300,
        "width": 2550,
        "charsCount": 2216
    }
]

Build BP with OCR and ToD Manual Task

To build a Business Process with the OCR and ToD Manual Task, do the following:

Create a Business Process.
Create an OCR Bridge step.
Configure OCR Bridge.
Add IE Cognitive Bot Task, Start, End, and link all steps together

The Business Process is now ready to process incoming documents by OCR and Manual Task.

Performance

Documents number	OCR Settings + BP		OCR Bridge		Diff
	AVG	Best Time	AVG	Best Time	AVG	Best Time
10	02:00.8	01:37.3	00:51.1	00:46.0	58%	53%
50	04:34.0	02:56.6	02:03.3	01:53.1	55%	36%
100	06:52.3	06:01.9	05:05.2	04:51.9	26%	19%
300	13:07.7	11:35.9	10:10.2	10:01.6	23%	14%

OCR Bridge demonstrates up to 23% performance boost compared to OCR Setting + OCR BP approach. See Bridge Perfomance.xlsx.

Architecture​

OCR REST configuration properties​

OCR bridge step​

Input parameters​

OCR JSON. Default values​

OCR parameters​

Output parameters​

Build BP with OCR and ToD Manual Task​

Performance​