Table of Contents
Comparison of Mobile SDKs: Mobile Imaging SDK II - Real-Time Recognition SDK
Feature Comparison
| Mobile Imaging SDK II | Real-Time Recognition SDK | |
|---|---|---|
| Key Features | - Automatic capture of best document image suitable for OCR directly from smartphone's camera preview screen - Automatic boundaries detection & cropping - Image quality evaluation 1) - Image compression & export - Image capture performed directly on the mobile device, no internet connection required. | - Text extraction directly from the smartphone's camera preview screen - Recognition of photos from gallery - Data capture from fields with help of regular expressions - Ready-to-use data capture methods for specific documents and fields: • IDs, passports, driver’s licenses • Credit and debit cards • MRZ - Text (word by word) translation - Image capture performed directly on the mobile device, no internet connection required. |
| Usage/Scenario | - Capture hight-quality image of documents on-the-go to submit it into other server side workflows for further processing. - Essential component in mobile apps for customer on-boarding or self-service processes. | Perfect for mobile apps: - Customer Registration, On-boarding or Customer Self-Service - Mobile Payment - Packaging OCR in Transport and Logistics - Machine Vision - Text to Speech - Translation on the go |
| Processing | ||
| Input | - Series of video frames - Image from gallery | - Series of video frames - Image from gallery |
| Image Preprocessing | - In video mode: document boundaries detection, cropping, perspective correction - For photos loaded from gallery: boundaries detection, cropping, rotation | - The input is a live video stream, there is no explicit image preprocessing API - For photos loaded from gallery: there is no explicit image preprocessing API |
| Layout Analysis | - The overall image quality is analyzed - Text elements are detected and examined to evaluate and demonstrate image quality and its suitability for OCR. | - Video mode: Text recognition areas are detected flexibly on any background and at any angle, this allows “finding” text in natural scenes and dynamic detection of text lines - Photo mode: Standard document analysis, which is looking for different elements (text, picture blocks) in a document. |
| Recognition Technologies | - none | - Text recognition in video or photo mode for Latin, Cyrillic or Greek alphabets and 4 Asian languages: Chinese (Simplified and Traditional), Japanese and Korean |
| Languages & Dictionary Support | - none | - OCR Support for 60+ languages - OCR dictionaries for 24 languages 2) - Possibility to use custom dictionaries |
| Translation | - none | - Built-in word by word translation - 18 Language combinations available for translation - No internet connection needed |
| Export | - Images - Multi-page PDFs (image only) | - No text export - API-access to the recognized and translated text |
| Further Details | ||
| Processing Power | - Low CPU requirements, no OCR applied | - CPU intense - Quad-core mobile processor recommended (available on all latest phones) |
| Related Articles | - ABBYY Mobile Imaging SDK II | - Real-Time Recognition SDK |
| Operating Systems | - Android - iOS | - Android - iOS |
| Cross-platform Support | - none, planned for the future | - Apache Cordova plugin available - other planned for the future |
| Product video | ||
| Market Availability | 2018 | 2016 |
Stand: 10/2018
Summary
With ABBYY's mobile tool-kits you can extend your existing or new apps with powerful image, text and data capture capabilities:
- Mobile Imaging SDK II: Enables developers to create apps for automatic capture of high-quality document image best suited for further processing.
- Real-Time Recognition SDK: Enables developers to add “instant” text capture functionalities to their mobile apps. Real-Time Recognition SDK immediately captures text by simply pointing the smartphone's camera on areas containing text or using images from the photo library for use within the mobile app.
Related Articles
Back to: Comparisons
- No tags, yet