Intelligent Data Capture
AUTOMATED DOCUMENT AND DATA CAPTURE
The application suite dedicated to the acquisition and digitalization of any type of information, data, multimedia content, paper or digital documents. Specific acquisition processes can be defined for capturing information and documents from any source or channels such as management systems, databases, information flows, network folders, ftp servers or any other general data capture device (scanners, multifunctional devices, healthcare devices, etc). The information captured can be forwarded directly to the document management system or to any external system.
A SUITE OF APPLICATIONS DEDICATED TO DATA CAPTURE
SCAN
Automated bulk scanning and digitalization of paper documents, page split and save file using predefined rules for name composition. Driver support ISIS, TWAIN, KOFAX and special scanner.
CLASSIFICATION
Documents are automatically categorized by analyzing several factors based on appearance and content.
IMAGE PROCESSING
Orientation adjustment, removing black borders, improving readability, removing dots, boxes, zone anchor lines.
ANOMALY DETECTION
Centralized AI-based anomaly analysis capable of detecting any irregular case determined by the rules set during training.
OPTICAL RECOGNITION
Typed text (OCR), handwritten text (ICR), cursive handwritten text (CHR), optical marker boxes (OMR), and 1D and 2D barcodes (BCR).
INFORMATION VALIDATION
Manual validation of extracted or recognized information that did not pass the configured auto verification steps directly on the specific area of interest on the image (enlarged up to 200%)
DOCUMENT CLASSIFICATION SUPPORTED BY AI ANALYSIS
Documents are automatically classified based on graphic appearance and content.
The classification process is based on the analysis of several factors to determine the category of the document.
APPEARANCE
The AI uses machine learning to learn the graphical model of each document category being stored, so it will be able to automatically recognize and catalog any other document belonging to the same type.
SYMBOLS
The AI is trained to identify recurring symbols or shapes on the image of the sample documents, the real-time analysis process will be able to detect their presence in certain areas and use this information in the automatic classification process.
CONTENT
AI can recognize and extract the content of images and documents and detect specific information or types of content in certain areas, the identification of a content, along with the location, will contribute to the automatic classification process.
CONFIGURE CUSTOMIZED ACQUISITION PROCESSES
The built-in workflow engine allows you to define custom acquisition processes that can capture information and documents from any external system, data source or device, analyze content, detect and identify any anomalies before forwarding the information into target repositories.
- Optical Acquisition
- Bulk import
- Data recognition and extraction
- Content analysis and classification
- Indexing and metadata creation
- Automatic anomaly detection
- Assisted and manual validation
- Information certification
- File conversion
- Information Export
OUR SOLUTIONS ARE BASED ON THE MOST INNOVATIVE TECHNOLOGIES
MICROSERVICES ARCHITECTURE
Provides scalability and resiliency by distributing workloads across services
ARTIFICIAL INTELLIGENCE
Powerful integrated AI engine supports collaboration between users and groups on specific work tasks
EVENT-BASED COMMUNICATION
With the exchange of asynchronous events, each application component and its execution is independent of the others.
MACHINE LEARNING
Automated content classification is supported by natural language processing (NLP) algorithms
CLOUD DISTRIBUTE PERFORMANCE
Distribute services such as processing servers, storage resources, repositories, and application services onto the network
INTEGRATION AND OPEN STANDARDS
Ability to integrate at any level thanks to public APIs and open standards such as XML, Json and Open Data
