Matching Text Entries with Image Objects Using Ambiguity Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques face challenges in accurately matching line item textual entries with individual image objects from digital images containing multiple objects, particularly due to ambiguous matches and varying spatial characteristics of the images.
Innovation Solution
A computer-implemented method that receives a document with line item textual entries and splits images into individual objects using deep learning, extracts attributes, and applies an assignment model to resolve ambiguous matches by defining priority constraints and iteratively assigning pairs based on confidence scores and other constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional matching techniques are used to match line item textual entries with image objects, then the matching process is simple, but the accuracy is low due to ambiguous matches and varying spatial characteristics
Solution Approach 1:
The patent segments the matching process into distinct stages: initial matching to identify candidate pairs, ambiguity detection to identify problematic matches, and resolution processing to resolve ambiguities. This segmentation allows the system to handle complex matching tasks systematically while maintaining manageable complexity at each stage.
Solution Approach 2:
The patent introduces an intermediary resolution process that acts as a mediator between initial matching and final results. This intermediary stage identifies ambiguous matches and applies resolution rules to determine the correct associations, thereby improving accuracy without requiring complete redesign of the matching system.
2Measurement precision
If ambiguous matches are resolved using simple methods, then the processing is fast, but the accuracy of one-to-one matching is compromised
Solution Approach 1:
The patent performs preliminary identification of ambiguous matches before final resolution. By detecting and flagging ambiguous cases in advance, the system can apply targeted resolution strategies only where needed, rather than processing all matches uniformly, thus balancing accuracy and efficiency.
Solution Approach 2:
The patent changes the processing parameters dynamically based on match ambiguity. For unambiguous matches, simple fast assignment is used; for ambiguous matches, more complex resolution rules are applied. This parameter adaptation allows the system to optimize the trade-off between speed and accuracy for different types of matches.
3Reliability
If all possible matches are considered to ensure accuracy, then the matching is thorough, but the computational complexity increases
Solution Approach 1:
The patent applies partial action by focusing computational resources on resolving only the ambiguous matches rather than re-evaluating all possible matches. This selective approach maintains reliability for critical ambiguous cases while avoiding unnecessary computational overhead for already-clear matches.
Solution Approach 2:
The patent applies different levels of analysis to different matches based on their local characteristics. Unambiguous matches receive minimal processing, while ambiguous matches undergo thorough resolution analysis. This local quality adaptation ensures high reliability where needed while reducing overall computational complexity.
Data Source
AI summary
A computer implemented method includes receiving a document with line item textual entries and an attachment containing images of different objects characterizing different transactions. The images of the different objects are split into individual image objects. Attributes from the individual image objects are extracted. The line item textual entries are matched with the individual image objects to form matched image objects. The matched image objects include ambiguous matches with multiple individual image objects assigned to a single line item textual entry or a single individual image object assigned to multiple line item textual entries. An assignment model is applied to resolve the ambiguous matches. The assignment model defines priority constraints, assigns pairs of line item textual entries and individual image objects that meet highest priority constraints, removes highest priority constraints when ambiguous matches remain, and repeats these operations until no ambiguous matches remain. One-to-one matches of line item textual entries and individual image objects are returned.