Image Recognition for Transaction Data Entry

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current business systems require users to manually input detailed information for transactions, which is time-consuming and prone to errors, especially when information is not entered immediately after the event.

Innovation Solution

An image recognition architecture that captures images within business transactions, identifies the information, and displays it for user confirmation, allowing users to automatically or manually enter the information into the appropriate fields within the transaction interface.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If users manually input detailed information for transactions, then the information can be accurately recorded, but the process becomes time-consuming and error-prone

Engineering Contradiction:

Improveaccuracy of information recordingVSAvoidtime required for data entry

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical data entry process with an automated image recognition system. The system captures images of documents or scenes, automatically extracts relevant information through computer vision algorithms, and populates transaction fields without manual typing, thereby reducing time while maintaining accuracy through automated verification

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the transaction system to automatically capture and process information without requiring user intervention for data entry. The image recognition system independently identifies, extracts, and validates information, making the process self-sufficient while improving both speed and accuracy

Inventive Principle:

Principle #25Self-service

2Measurement precision

If users manually enter information immediately after events, then accuracy is maintained, but the process is still time-consuming and complex

Engineering Contradiction:

Improveaccuracy of transaction informationVSAvoidcomplexity of data entry process

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary image recognition system between the user and the transaction database. This intermediary automatically captures images, extracts information, validates data, and populates fields, thereby simplifying the user interface to basic actions like capturing an image while maintaining high accuracy through automated processing and validation layers

Inventive Principle:

Principle #24Intermediary (Mediator)

3Ease of operation

If users delay information entry, then the workflow becomes more flexible, but memory retention decreases leading to errors

Engineering Contradiction:

Improveflexibility of transaction workflowVSAvoidaccuracy of recalled information

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent implements preliminary action by capturing images at the moment of the event rather than requiring later recall. The system takes pictures of documents, receipts, or scenes immediately when they occur, preserving visual information in its original form. This allows users to delay actual data processing while the visual record remains available, eliminating memory issues while maintaining workflow flexibility

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11416948B2Image tagging for capturing information in a transaction

Publication Date: 2022.08.16 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

An image is captured within an application transaction. Identifying information corresponding to the captured image is obtained and displayed for user confirmation. The identifying information is entered into an appropriate place within the transaction.