Table of Contents
Comparison Cloud OCR SDK vs. FineReader Engine SDK
The following tables list the major differences between ABBYY's Cloud OCR Service with API and a “traditional” ABBYY OCR SDK ABBYY FineReader Engine.
This comparison is rather technical and compares the “service approach” with the “deep integration approach”.
Usage
| Cloud OCR SDK | FineReader Engine | |
|---|---|---|
| Target Audience | Developers who need to add an OCR service to their applications, ISVs developing in the Cloud Enterprise companies Libraries | Developers ISVs Integrators Scan Service Providers |
| Integration | “High level” integration via web service; REST web-service calls | “Low level” integration via (COM) API |
| File Management | Uploading document images into the “cloud” | Local processing of files by the custom-made application or service |
| Processing | “Asynchronous” (out of the system) upload, process, receive the results, re-ocr if needed | “Synchronous” (in the system) open image, processing, “on the fly” result check, re-process right away, export |
| Scalability | Cloud based, ABBYY's responsibility | Own IT infrastructure, Custom upscale implementation |
| Hardware | Microsoft data centers within the European Union and the United States of America | Own hardware or data center |
| Security | HTTPs, Azure infrastructure, Azure security | Custom implementation, own hardware |
| Maintenance | No own maintenance by the developer. Regular updates conducted by ABBYY after pre-announcement | Regular updates and maintenance releases by the developer, custom set-up and administration |
Features
| Cloud OCR SDK | FineReader Engine 12 | |
|---|---|---|
| Image & PDF import | yes | yes |
| Office formats import | no | yes |
| OCR | yes | yes |
| OBR (Barcode recognition) | yes | yes |
| ICR | yes | yes (Windows & Linux platforms) |
| OMR (Optical Mark Recognition) | yes | yes (Windows & Linux platforms) |
| Zonal/Field Level Recognition | yes | yes |
| Business Card Recognition | yes | yes |
| MRZ Recognition | yes | yes |
| Receipt Recognition | yes | no ready-made functionality, only support of 'Receipt' text type. It allows processing of receipts, which can be subsequentially parsed to extract specif data. |
| AI-based Document Classification | no | yes |
| OCR Languages | all available recognition languages without additional costs | all standard languages + several language add-ons (Arabic, CJK, Hebrew, Thai…) |
| User patterns and custom dictionary integration | no | yes |
| Historic Font OCR (Fraktur/Gothic) | yes | yes, when licensed for the project |
| Processing profiles fine-tuned for a specific scenario | - documentConversion - documentArchiving - textExtraction - barcodeRecognition | Processing profiles available in 2 modes: accuracy and speed: - DocumentConversion - DocumentArchiving - BookArchiving - TextExtraction - BarcodeRecognition - FieldLevelRecognition - HighCompressedImageOnlyPdf - BusinessCardsProcessing - MachineReadableZone - EngineeringDrawingsProcessing - Default |
| Recognition Modes for accuracy and speed fine-tuning | 3 recognition modes: Normal, Balanced and Fast | |
| Document Separation | no | custom implementation required |
| GUI Elements | none, processing only | yes, available as Visual Components |
| Export | TXT, XML, ALTO Office Formats PDF, PDF/A | TXT, XML, ALTO Office Formats PDF, PDF/A Internal Engine Format some formats offer granular export settings |
| Core Technology | Version 12 | Windows, Linux, OS X: Version 12 |
Development
| Cloud OCR SDK | FineReader Engine | |
|---|---|---|
| Trial | No trial needed - Online subscription to the OCR service for free | Trial Software License Agreement can be signed online |
| Development Tools | All development tools that can interact with a web-service | Native OS tools that can integrate programming libraries |
| API | Compact API, suited for web-service implementation | Rich (COM) API with many objects, API gives low level access to a lot of objects settings and parameters Object Diagram of FineReader Engine 11 |
Operation / Production
| Cloud OCR SDK | FineReader Engine | |
|---|---|---|
| Setup | No own setup necessary, ready to integrate | Own application rollout |
| Maintenance | Regular updates provided by ABBYY - customer informed | Regular implementation of new releases by own Admin / IT staff |
| Hardware | Microsoft Azure Data Processing Centers in the European Union and the USA | own hardware |
| Scalability | Built-in scalability, controlled & maintained by ABBYY | Custom code - provided by the developer |
Business Models
| Cloud OCR SDK | FineReader Engine | |
|---|---|---|
| Payments | Payments for the OCR Service are based on number of processed pages | Developer & Runtime Licensing |
| Payment for Software Maintenance | No extra payment - the OCR service is maintained by ABBYY on regular base | Extra charge for Software Maintenance allows access to the latest software version and technical support. |
| Project Licensing | Low to mid-size page volumes can be purchased online. Contact ABBYY directly for high volume pages Cloud OCR Business & Pricing | Contact ABBYY to receive a pricing proposal |