Document Validation via Motion Tracking and Image Categorization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for scanning and validating documents using portable devices suffer from low image quality and inconsistency, making it difficult to achieve reliable full-page document recognition and fine-grained OCR capabilities, especially with varying image sensor types and capture conditions.

Innovation Solution

The system employs video motion tracking and image categorization to improve document capture quality by combining full-document and Region of Interest processing, using a smartphone app to guide users in capturing continuous video or still images, and performing authentication based on feature recognition and pattern matching.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If full-page document scanning is performed using portable device cameras, then document capture coverage is improved, but image quality and recognition reliability deteriorate

Engineering Contradiction:
Improvedocument capture coverageVSAvoidimage quality
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent divides the document scanning process into two distinct phases: full-page capture mode for obtaining complete document coverage, and zoom-in mode for capturing high-resolution details of specific regions. This segmentation allows the system to optimize for different requirements simultaneously - coverage in one mode and quality in another.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent adds a temporal dimension to the scanning process by capturing continuous video frames rather than single static images. This allows the system to select optimal frames from the video sequence and combine information across multiple frames to achieve both full-page coverage and high-quality detail recognition.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If zoom-in capture is used for detailed OCR, then pattern recognition accuracy is improved, but document continuity verification becomes more difficult

Engineering Contradiction:
Improvepattern recognition accuracyVSAvoidvideo continuity monitoring
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements feedback mechanisms that monitor video continuity and provide real-time guidance to users. The system analyzes successive video frames to detect whether the camera has moved away from the document or lost tracking, and provides feedback prompts to the user to maintain proper capture conditions throughout the scanning process.

Inventive Principle:
Principle #23Feedback

3Productivity

If portable device cameras are used for scanning, then accessibility and speed are improved, but image consistency across different devices deteriorates

Engineering Contradiction:
Improvescanning speedVSAvoidimage consistency
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent dynamically adjusts capture parameters such as frame selection criteria, zoom level, and exposure settings based on the specific characteristics of each device's camera and the current capture conditions. This allows the system to optimize performance for each individual device while maintaining consistent output quality across different hardware platforms.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS9619701B2Using motion tracking and image categorization for document indexing and validation
Publication Date: 2017.04.11 CONDUENT BUSINESS SERVICES LLC
  • US9619701B2 patent drawing
  • US9619701B2 patent drawing
  • US9619701B2 patent drawing

AI summary

Systems and methods include an application operating on a device. The application causes the graphic user interface of the device to display an initial instruction to obtain a full-view image that positions all of an item within a field of view of a camera on the device. The application automatically recognizes identified features of the full-view image, by using a processor in communication with the camera. After displaying the initial instruction, the application causes the graphic user interface to display a subsequent instruction to obtain a zoom-in image that positions only a portion of the item within the field of view of the camera. Also the application automatically recognizes patterns from the zoom-in image, using the processor. Furthermore the application performs an authentication process using the identified features and the patterns to determine whether the item is valid, using the processor.