Multi-Language OCR Segmentation and Parallel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional OCR engines face challenges in processing multi-language content efficiently, leading to increased processing time and reduced accuracy, especially when dealing with documents containing text in multiple languages such as English and Russian.
Innovation Solution
A method and system that extract text portions from images, segment keywords into layers based on associated languages, and generate OCR outputs for each layer using a multi-language processor, resulting in the creation of an electronic document that can be transmitted over a communication network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single OCR engine trained for one language is used to process multi-language documents, then the processing speed is maintained, but the recognition accuracy deteriorates for languages other than the training language
Solution Approach 1:
The patent segments the multi-language document into multiple language-specific portions using language detection and segmentation modules. Each portion is then processed by a dedicated OCR engine trained for that specific language, thereby maintaining high recognition accuracy for each language while preserving overall processing efficiency through parallel processing of segmented portions.
2Measurement precision
If multiple OCR engines trained for different languages are used to process a multi-language document, then the recognition accuracy is improved, but the processing time increases
Solution Approach 1:
The system segments the document into language-specific portions and processes each segment with the appropriate OCR engine simultaneously using parallel processing. This approach maintains high recognition accuracy for each language while minimizing overall processing time by avoiding sequential processing of all language portions.
Solution Approach 2:
The patent implements preliminary language detection and segmentation before OCR processing. By identifying and separating different language portions in advance, the system prepares the document for optimized parallel processing, reducing the total processing time while maintaining accuracy for each language.
3Measurement precision
If manual translation is performed by subject matter experts for multi-language documents, then the accuracy is improved, but the productivity and scalability deteriorate when dealing with large volumes of documents
Solution Approach 1:
The patent implements an automated system that performs language detection, segmentation, and OCR processing without requiring manual intervention by subject matter experts. The system autonomously handles multi-language documents by routing each language portion to the appropriate trained OCR engine, thereby maintaining high accuracy while achieving scalable processing of large document volumes.
Data Source
AI summary
A method and system are provided for optical character recognition (OCR) of multi-language content. The method includes extracting a text portion from an image received from a user-computing device. The text portion comprises a plurality of keywords associated with a plurality of languages. The method further includes segmenting the plurality of keywords into a plurality of layers. Each layer of the plurality of layers comprises one or more keywords which are associated with a language. The method further comprise generating an OCR output of each of the plurality of layers based on the language associated with the one or more keywords in each of the plurality of layers. The method further comprises generating an electronic document of the received image based on the generated OCR output of each of the plurality of layers. The method further includes transmitting the generated electronic document to the user-computing device.