Surya - an open-source OCR toolkit that supports recognition of 90+ languages and layout analysis.

What is Surya?

Surya is a powerful open-source OCR (Optical Character Recognition) tool.Optical character recognitionSurya is a toolkit specifically designed for document recognition, supporting over 90 languages. It accurately identifies text within documents, analyzes the reading order of the text, detects layout elements such as tables, images, and headings, and recognizes and parses table content. Surya is renowned for its efficient table recognition capabilities, outperforming many existing open-source models, such as Table Transformer. Completely free and intended for commercial use, Surya supports cross-platform operation, including Windows, Mac, and Linux systems, making it suitable for offline environments handling sensitive information.

Surya's main functions

• Multilingual OCR recognitionSurya supports optical character recognition in over 90 languages, processing documents in multiple languages including Chinese, Japanese, Korean, and Arabic.
• Table recognitionSurya enhances table recognition capabilities, enabling it to identify rows, columns, and cells in a document, including rotated and complexly laid-out tables.
• Complex layout recognitionSurya can recognize complex layouts in documents, such as headings and images, and process various elements within the document.
• Text detection and reading orderSurya can perform line-level text inspection to determine the reading order of the text and ensure that the output text content is in the correct order.

Surya's technical principles

• Deep learning modelsSurya uses a deep learning model to identify text and layout elements in documents. The model is trained on a large amount of data to recognize and understand the structure and content of documents.
• Semantic segmentationIn terms of text detection, Surya uses deep learning-based semantic segmentation technology to separate text regions from non-text regions in a document.
• Object detectionFor layout analysis, Surya uses object detection technology to identify different elements in the document, such as tables, images, and headings.
• Sequence ModelIn reading order detection, Surya uses sequence models to analyze the relative positions and directions between text lines to determine the correct reading order.
• Optimized AlgorithmSurya optimizes the algorithm to improve processing speed and accuracy.

Surya's project address

• GitHub repository:https://github.com/VikParuchuri/surya

Surya's application scenarios

• Document digitizationConverting paper documents into electronic formats facilitates storage, retrieval, and editing. This is extremely useful for records management, library digitization projects, and personal document organization.
• Data extractionAutomatically extract data from structured documents such as tables, invoices, and reports for use in data analysis, financial auditing, or database population.
• Multilingual processingSupporting over 90 languages, Surya is suitable for processing documents in multilingual environments, such as document processing for multinational corporations and digitizing multilingual books.
• Automated officeIn office automation, Surya automatically identifies and processes documents such as emails, letters, and contracts, improving work efficiency.
• academic researchResearchers can quickly extract text content from large amounts of scientific literature, ancient books, or historical documents to facilitate research and analysis.