Anantha Lakshmi Kannan![Automate Document Data Extraction for Apps with Syncfusion Smart Extractors [Webinar Show Notes]](https://www.syncfusion.com/blogs/wp-content/uploads/2026/09/Automate-Document-Data-Extraction-for-Apps-with-Syncfusion-Smart-Extractors-Webinar-Show-Notes.jpg)
Table of Contents
Extracting structured information from PDFs and scanned documents is often a challenging task for business applications.
In this webinar, presenter Anantha Lakshmi Kannan demonstrated how the Syncfusion® Smart Data Extractor, Smart Table Extractor, and Smart Form Recognizer help .NET applications extract structured information from PDFs and images.
Using an ASP.NET Core application, the session showcased document structure extraction, table extraction, form recognition, structured JSON output, confidence scores, and PDF-based results.
If you missed the webinar, the recording is available on our YouTube channel and is embedded below.
What you’ll learn
- Extract structured content from PDFs and images.
- Convert table data into usable formats.
- Recognize form fields and create fillable PDFs.
- Work with JSON-based extraction results.
- Process documents locally without external service dependencies.
Demo highlights
Data extraction
The Smart Data Extractor identifies document elements such as text blocks, images, tables, form fields, headers, footers, and other layout regions. In the demo, it processed a PDF invoice and returned structured JSON with confidence information.
Table extraction
The Smart Table Extractor identifies and extracts structured table data, including rows, columns, and cells. In the demo, it extracted table data from an invoice image. The application then used the Syncfusion Excel Library to convert the JSON result into an Excel file.
Form recognition
The Smart Form Recognizer detects form controls and their locations in static forms. In the demo, the recognized result was used to create a fillable PDF that users could download from the application.
Timestamps
[00:00] Introduction
[01:35] The document data extraction challenge
[03:03] Demo overview and application setup
[05:01] Smart Data Extractor
[09:12] Smart Data Extractor API walkthrough
[12:47] Smart Table Extractor
[18:34] Smart Form Recognizer
[22:19] Smart Form Recognizer API walkthrough
[24:44] Real-world applications and key considerations
[27:07] Key takeaways
Q&A
Q: The documents processed must be text PDFs, or is the system also able to recognize image PDFs?
A: Yes. The extractors support both text-based PDFs and scanned image PDFs.
Q: Can you extract to an Excel sheet or SQL Input statements?
A: Currently, the extracted structured data can be exported in JSON and Markdown formats. Direct export to Excel or SQL statements is not supported at this time.
Q: Can you extract from a handwritten document?
A: The solution uses internally trained ONNX models to extract information from PDFs and images. Recognition accuracy for handwritten content depends on the model’s ability to identify and classify elements such as text, tables, and images within the document.
Q: Can the Table Extractor work on a data row that is spread over multiple printed rows?
A: Yes. The Smart Table Extractor can detect and extract table structures, including rows that span multiple printed lines, while preserving the row and column relationships.
Q: Is data/table extraction processed locally, or is there an external server call in the background?
A: Data and table extraction can be processed entirely locally without any external server communication. You can deploy the library directly within your application or on-premises infrastructure, enabling fully offline document processing.
Q: The assemblies (libraries) package references that you are using are they able to recognize special characters (localization)?
A: Yes. The libraries support recognizing and extracting special characters from PDF documents.
Q: Is this resource-intensive?
A: The extraction process is optimized for performance. Typical memory consumption is under 250 MB, though actual resource usage may vary depending on document complexity, size, and extraction settings.
Q: Can you extract from an HTML-based email?
A: No. Currently, data extraction is supported only for PDF and image inputs. HTML email content is not supported.
Q: Are there any samples up on GitHub?
A: Yes. Sample projects are available in the Syncfusion GitHub repository.
Q: Do you guys use Tesseract under the hood?
A: Yes. For extracting text from scanned PDFs and images, the OCR processing utilizes Google’s Tesseract Optical Character Recognition (OCR) engine as part of the document processing workflow.
Conclusion
Smart Data Extractor, Smart Table Extractor, and Smart Form Recognizer simplify extracting structured information from documents. Whether you’re building invoice-processing workflows, document automation systems, or form-driven applications, these tools transform PDFs and images into usable data that you can integrate directly into .NET applications.
Explore the related resources below to learn more and start building your own document-processing solutions.
No spam, just valuable updates.