Contextual Form Verification Interface

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users face laborious and time-consuming tasks when verifying automatically recognized data from source forms due to imperfect optical character recognition and the need to locate information across large, complex forms.

Innovation Solution

A user interface is developed that identifies regions of a form image containing values and labels, performs automatic recognition, and presents the relevant region to the user for verification, allowing acceptance of recognized values and labels as correct.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If optical character recognition is used to automatically extract data from forms, then transcription time is reduced, but verification accuracy deteriorates due to imperfect recognition

Engineering Contradiction:
Improvedata entry speedVSAvoidrecognition accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system presents the recognized value to the user for verification, allowing the user to confirm or correct the automatically extracted data. This feedback loop compensates for recognition errors while maintaining high productivity.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The user interface acts as an intermediary between the optical character recognition system and the final data entry, providing a verification step that resolves discrepancies without completely abandoning automated extraction.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the user is required to locate and verify each value in a complex form, then verification accuracy is maintained, but time consumption increases

Engineering Contradiction:
Improveverification accuracyVSAvoidverification time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system extracts and highlights only the specific region containing the value to be verified, removing the need for the user to search through the entire complex form. This maintains verification accuracy while dramatically reducing the time required.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of presenting the entire form, the interface focuses on the local region containing the relevant value and its label, providing targeted verification support without overwhelming the user with unnecessary information.

Inventive Principle:
Principle #3Local quality

3Loss of information

If the entire form is presented to the user for verification, then complete context is provided, but user attention is分散 and verification efficiency decreases

Engineering Contradiction:
Improvecontext completenessVSAvoidverification efficiency
Core Design Contradiction:
Loss of informationVSProductivity

Solution Approach 1:

The system extracts and displays only the necessary portion of the form containing the value and its label, eliminating distracting elements while preserving the contextual information needed for accurate verification.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS11182544B2User interface for contextual document recognition
Publication Date: 2021.11.23 HRB INNOVATIONS
  • US11182544B2 patent drawing
  • US11182544B2 patent drawing
  • US11182544B2 patent drawing

AI summary

Media and method for presenting a user interface for verifying automatically recognized information from forms. Embodiments of the invention may import an image of a form and break the form down into a series of bounding boxes, each including the value of a form item and a description and/or label for the form item. When making use of a particular value from the form image, embodiments of the invention may present the bounding box containing that value to the user, together with enough context for the user to verify that the correct item has been identified and that the automatically recognized value is correct.