Mobile Image Validation Using Native ML for Check Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile operations software programs for remote check deposit face challenges such as slower processing times, crashes, and ineffective validation due to reliance on third-party libraries, particularly in capturing and validating check images under varying environmental conditions and hardware configurations, which often fail to meet specific criteria like brightness, contrast, and rectangular format.
Innovation Solution
A mobile application leveraging native image processing software (e.g., Apple VisionKit®) on client devices uses machine-learning models to perform client-side image processing and validation, capturing frames at preconfigured intervals, and obtaining information like brightness, contrast, and text recognition to validate document dimensions and security features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If third-party libraries are used for image processing in mobile applications, then image processing functionality is provided, but processing speed decreases and system stability deteriorates
Solution Approach 1:
The patent extracts the image processing functionality from third-party libraries and implements it using native iOS frameworks (AVFoundation, Core Image, Core Graphics). This removal of the problematic third-party dependency eliminates the crashes and performance issues while maintaining the required image capture, processing, and validation capabilities through Apple's optimized native tools.
Solution Approach 2:
The patent replaces the mechanical dependency on external third-party libraries with a native system architecture that leverages iOS built-in frameworks. This substitution eliminates the integration overhead and compatibility issues associated with third-party libraries, resulting in faster processing and improved system stability through direct access to native hardware and software optimizations.
2Productivity
If third-party libraries are used for image processing, then image processing capabilities are provided, but processing time increases
Solution Approach 1:
By extracting image processing operations from third-party libraries and implementing them natively using AVFoundation and Core Image, the patent eliminates the overhead of library initialization, method dispatch, and cross-platform compatibility layers. This direct native implementation significantly reduces processing time for image capture, validation, and manipulation operations.
Solution Approach 2:
The patent changes the implementation parameters from third-party library functions to native iOS framework functions, which are optimized for the specific hardware architecture. This parameter change enables more efficient memory management, faster processor access, and optimized algorithm execution, thereby reducing overall processing time.
3Ease of operation
If remote deposit capture is offered to improve convenience, then user convenience increases, but verification of user identity and check authenticity becomes challenging
Solution Approach 1:
The patent performs preliminary validation actions directly on the client device before submission, including checking document dimensions, aspect ratios, brightness, contrast, and detecting potential tampering. This preliminary verification ensures that only properly captured and authentic documents are submitted for remote processing, maintaining high verification accuracy while preserving user convenience.
Solution Approach 2:
The patent implements immediate feedback mechanisms that provide users with real-time validation results, including alerts for improper document capture, brightness issues, or potential fraud indicators. This feedback loop allows users to correct issues before submission, ensuring both convenience and reliable verification without requiring additional manual review steps.
Data Source
AI summary
Presented herein are systems and methods for the employment of machine learning models for image processing. A mobile application for client-side image processing and validation, which interacts with and leverages native image processing software of the client device, where the image processing software and the mobile application include any number of machine-learning models for identifying a document and attributes of the document for recognition and validation. This mobile application uses the image processing software from a client operating system to control the camera. The image processing software generates various types of information about a video frame and the document, and the mobile application invokes APIs or software libraries of the image processing software to access the information and validate the frame and document.


