This topic describes the features, advantages, and application scenarios of the general-purpose text recognition products from Alibaba Cloud OCR. It also provides quick links to the product APIs.
Product introduction
Alibaba Cloud OCR general-purpose recognition products can recognize and restore text from various common document images or scanned documents while preserving the original document format. To better restore text and document structure, the document recognition feature adds layout analysis and document image editing capabilities to its general full-text recognition features. These features include text localization, line analysis, and text recognition. This allows for the structured extraction of document elements from document images and improves the user experience.
Demo URL: https://duguang.aliyun.com/experience?type=universal
Activate the service to receive a free quota: https://ocr.console.aliyun.com/overview
Purchase: https://common-buy.aliyun.com/?commodityCode=ocr_general_dp_cn
Product features
General-purpose text recognition
Alibaba Cloud OCR general-purpose text recognition is suitable for unstructured text recognition in various industry scenarios. It returns text content and coordinate information.
High-accuracy full-text recognition (Recommended)
Alibaba Cloud OCR high-accuracy full-text recognition supports accurate recognition of multiple layouts, complex document backgrounds, and various lighting conditions. The document recognition rate exceeds 99.7%. For documents with seals or handprints, it can recognize text after the seals are removed. It also supports advanced features such as low-confidence-level filtering and pattern detection.
General-purpose handwriting recognition
The Alibaba Cloud OCR general-purpose handwriting recognition model supports handwriting recognition in complex scenarios, including Chinese, English, and numbers. It can also recognize printed text, which makes it suitable for various handwritten notes and whiteboard content.
Table recognition
Alibaba Cloud OCR table recognition can effectively recognize lined, striped, and unlined tables.
Intelligent table parsing: Performs general-purpose table parsing to extract table styles, content, and key-value pairs from text and tables. It supports PDF documents up to 100 MB and 100 pages, and up to 30 images for image document formats. Try it now for free
E-commerce image text recognition
Alibaba Cloud OCR E-commerce image text recognition is a core product designed for fast and accurate character recognition in online images. This includes E-commerce product promotions, online community posts, and user-generated content (UGC). This feature is highly valuable for scenarios such as illegal ad detection, information review management, and network security administration.
Structured document recognition
Alibaba Cloud OCR structured document recognition can recognize document information in a structured manner. It provides layout information output from two perspectives: a tiled element view and a hierarchical tree view. It can extract and sequentially output text elements, such as single characters, text blocks, and lines, and their corresponding layout formats, such as titles, paragraphs, and tables. Currently, this feature only supports single-page documents.
Intelligent document parsing: Extracts logical hierarchical structures, text content, table content, key-value fields, and style information from documents. It analyzes the content, layout, and logical information of a document to output the extracted results as structured data. It supports PDF documents up to 100 MB and 100 pages, and up to 30 images for image document formats. Try it now for free
Advantages
High accuracy
The models are trained on massive image samples to achieve industry-leading accuracy. For example, the accuracy for ID card recognition exceeds 99%.
Low latency
The service relies on Alibaba's self-built EAS online service clusters and continuously optimized inference technology. This provides a low-latency service with elastic scaling.
Advanced technology
The service is built on Alibaba Cloud Platform for AI. It uses Alibaba's deeply optimized deep learning framework, PAI-TensorFlow, to train advanced text detection and recognition models.
Stable service
The service offers elastic scaling based on call volume, which ensures good extensibility. Continuous algorithm iterations and optimizations do not affect service stability for users.
Scenarios
Image content moderation
You can use various general-purpose APIs to recognize content for moderation in different scenarios. This helps promptly detect non-compliant behavior, significantly reduces manual labor costs, and is widely used in E-commerce content administration.
Contract and document recognition
You can use general-purpose text recognition to recognize text from images of contracts, documents, and novels. This feature is useful for scenarios such as contract proofreading, document retrieval, and PDF content extraction. It is widely used in industries such as judicial case file management, corporate legal contract review, and automated processes in finance and insurance.
API quick links
Alibaba Cloud Marketplace API links (Legacy) |
Official website API links (New) |