Dynamic Document Edge Tracking for Mobile Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile check capture technologies face challenges with larger documents, such as invoices, due to issues like poor image quality, incorrect orientation, and inadequate resolution, leading to errors in Optical Character Recognition (OCR) processing, which can result in time-consuming and undesirable situations for both consumers and banks.
Innovation Solution
A method and system for dynamic tracking of document edges within a camera's field of view, superimposing a guiding rectangle on the display to align the document, and automatically capturing an optimal image when edges align with the rectangle, ensuring proper distance and orientation for higher resolution imaging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional manual check capture methods are used, then the process is simple to operate, but image quality is poor and OCR processing fails frequently
Solution Approach 1:
The system performs self-alignment and self-capture by automatically detecting document edges, calculating the optimal framing rectangle, and triggering the camera when alignment is achieved, eliminating the need for manual positioning and button pressing by the consumer
Solution Approach 2:
The patent replaces manual mechanical alignment operations with automated computer vision-based edge detection and rectangle calculation algorithms, substituting human manual positioning with algorithmic guidance and automatic capture triggering
2Area of stationary object
If the camera is positioned closer to capture larger documents, then more of the document fits in the frame, but image resolution and quality deteriorate
Solution Approach 1:
The patent introduces a visual guidance dimension by superimposing a calculated rectangle on the camera display, providing the consumer with spatial orientation information that enables them to achieve proper framing and distance without trial and error, thereby optimizing both coverage and resolution
3Reliability
If manual alignment is required, then the consumer has control over the capture process, but alignment errors occur frequently leading to capture failures
Solution Approach 1:
The system provides visual feedback by displaying the calculated rectangle on the camera viewfinder, allowing the consumer to see whether the document is properly framed and make real-time adjustments, thereby ensuring reliable capture without requiring complex automated mechanisms
4Adaptability or versatility
If the screen orientation is preset in landscape mode, then the device setup is simple, but it cannot accommodate larger documents that require different orientations
Solution Approach 1:
The patent makes the screen orientation dynamic by allowing the camera interface to adapt to different orientations based on how the consumer holds the device, rather than being fixed in landscape mode, enabling flexibility for various document sizes and formats without requiring manual configuration
Data Source
AI summary
Video frames of a document are captured. The edges of the document are identified from the frames as they appear within a field of view of a camera and a document rectangle for the document is tracked within the field of view. An area of the document determined from the frames is compared against the available area of the camera preview mode and a larger superimposed guiding rectangle is displayed as a guide within the camera preview mode. When a frame for the document is detected as having document edges that substantially fit within the guiding rectangle, the guiding rectangle is removed and an optimal image of the document is captured for subsequent downstream document image processing.