Character Recognition Feedback via Likelihood Maps

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Character recognition apparatuses face challenges in user convenience due to the lack of clear feedback during the recognition process and degraded visibility when displaying multiple results, making it difficult for users to determine if the recognition is properly executed.

Innovation Solution

The apparatus includes a processor with multiple display processors that provide feedback information, such as likelihood maps and frame borders, to indicate the recognition process status and results, enhancing user understanding and convenience by displaying character candidate areas and recognition results effectively.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If character recognition processing is executed, then character recognition accuracy is improved, but processing time increases and user feedback is delayed

Engineering Contradiction:
Improvecharacter recognition accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary processing by displaying feedback information during the recognition process. The processor generates and displays intermediate results or status indicators before the final recognition is complete, allowing users to understand the process status without waiting for the full processing time to elapse.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements continuous feedback by displaying feedback information that shows the recognition process status. This feedback mechanism allows users to see progress and intermediate results during processing, reducing the perceived waiting time while the actual recognition accuracy is maintained through complete processing.

Inventive Principle:
Principle #23Feedback

2Device complexity

If only the taken image is displayed during recognition, then display simplicity is maintained, but user awareness of recognition status is lost

Engineering Contradiction:
Improvedisplay simplicityVSAvoidrecognition status information
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The display is segmented into multiple regions: the original taken image area and a separate feedback information display area. This segmentation allows the main image to remain simple and recognizable while providing a dedicated space for status information, preventing information loss without significantly increasing overall display complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Feedback information acts as an intermediary element between the taken image and the final recognition result. It provides intermediate status indicators that bridge the gap between the simple image display and the need for recognition status awareness, allowing users to understand processing state without complicating the main image view.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Quantity of substance

If multiple character recognition results are displayed, then recognition completeness is improved, but visibility and user comprehension are degraded

Engineering Contradiction:
Improvenumber of recognition resultsVSAvoidvisibility and comprehension
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

Different regions of the display provide different types of information with different qualities. The feedback information area presents processed, structured data in a simplified format, while the original image area maintains its visual integrity. This local differentiation allows multiple results to be presented without degrading overall visibility, as each region serves its specific purpose optimally.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3379452B1Information processing apparatus, method and computer program product
Publication Date: 2024.10.23 KK TOSHIBA
  • EP3379452B1 patent drawingFigure 1
  • EP3379452B1 patent drawingFigure 2
  • EP3379452B1 patent drawingFigure 3

AI summary

According to one example, an information processing device (10) includes a memory (12) and a processor (13). The processor is configured to acquire an image, determine a likelihood of presence of an objected within the image, display information on a position of the object within the image based on the likelihood, identify the object within the image, and generate a recognition result based on the identified object.