With the new Enhanced Form Support feature, DocuWare IDP takes automated form processing to the next level. The enhanced OCR model now enables native extraction of checkbox values across the most common checkbox patterns—for the first time without workarounds or complex model configurations.
Enhanced Form Support allows users to capture checkboxes in forms in a structured and reliable way, whether they represent single consents, mutually exclusive options, or multiple selections. This is complemented by improved OCR support for fillable forms with input fields arranged in boxes or segments (one character per field), as commonly found in government and administrative documents.
Challenges in Form Processing
Forms are among the most commonly used document types in organizations—from consent declarations and application forms to government documents. At the same time, they are some of the most challenging documents to automate:
- Checkboxes appear in many different variations
- Selection options can be single, mutually exclusive, or multiple
- Layouts vary depending on version, source, or language
- Text fields are often split into individual boxes
- Manual post-processing is error-prone and time-consuming
- Captures a single yes/no choice — for example, consent or confirmation.
- Values: checked / not checked
- Captures exactly one choice from several options (like radio buttons).
- Example: choose Credit Card or PayPal
- Captures several choices at once from a list of options.
- Example: select Email and/or Phone
- Reliable Checkbox Extraction: Native support for all common checkbox patterns
- Reduced Model Complexity: No workarounds needed for common checkbox patterns
- Higher Data Quality: Validations and annotation feedback reduce errors
- Optimized for Forms: Improved OCR for segmented and structured input fields
- Consent and Compliance Declarations
(e.g., data protection, newsletters, data sharing) - Single and Multiple Selections
(payment methods, contact preferences, interests, languages, allergies) - Government and Administrative Forms
with checkboxes and boxed or segmented input fields (one character per box), such as applications, registrations, or ID documents
Traditional OCR approaches have quickly reached their limits in this area. Checkboxes often had to be interpreted indirectly or modeled through complex workarounds—resulting in higher effort and limited reliability. Enhanced Form Support closes this gap.
Seamless Integration: How Enhanced Form Support Helps Your IDP Workflows
New Form Capabilities work best when your Custom Extraction uses the latest OCR version. This applies to all Custom Extractions created after December 19, 2025. If there is a need for older Custom IDP Workflows, please contact Professional Service to manually upgrade the OCR version.
At the core of the enhancement are three new checkbox field types that can be defined directly in extraction models:
-
Single Checkbox
-
Multiple Checkboxes – One Selection
-
Multiple Checkboxes – Multiple Selections
Annotation is deliberately simple and consistent: only the checkbox itself is annotated—regardless of whether it is checked or not. Intelligent validations ensure that multiple-selection fields are correctly defined. Visual feedback in the annotation interface highlights incomplete checkbox groups, reducing errors during training.
The results are available directly in the interface and via the API. They are structured according to the field type: Boolean (checked/not checked), Enum (single choice), or Enum List (multiple choices).
Key Benefits
Automated Form Processing with High Accuracy
Companies and organizations that regularly process form-based documents benefit particularly from Enhanced Form Support. Typical use cases include:
The feature supports both fixed form layouts and forms with the same semantic content but varying structure. Even fillable forms with character-by-character input fields are processed much more reliably thanks to improved OCR logic.
The result: significantly less manual rework, consistently structured data, and faster processing times—even for complex form landscapes.
Transparency: Current Limitations and Outlook
As with any OCR-based processing, there are currently known limitations—for example, checkboxes without a clear border or highly segmented text inputs. To mitigate this, aggregation and normalization logic has already been implemented to merge and standardize characters.
These areas are actively being developed and will be further improved in upcoming OCR versions, with the clear goal of continuously increasing the automation rate.
Rethinking Form Processes – with DocuWare IDP
With Enhanced Form Support, DocuWare IDP significantly improves automated form processing. The combination of improved OCR, native checkbox support, and structured output enables more reliable automation for common form patterns.
Companies can reduce manual effort, improve data quality, and lay the foundation for fully digital, rule-based processes – from document intake to the business application.
Discover how DocuWare IDP can transform your form-based IDP workflows – and get started now with intelligent automation.