Graphic Code Recognition Using Deep Learning Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing graphic code recognition methods struggle to efficiently recognize multiple graphic codes in a single image, often requiring users to manually select and recognize specific codes, leading to reduced efficiency.

Innovation Solution

A graphic code recognition method that uses target detection based on deep learning to identify positions and types of graphic codes in an image, allowing for simultaneous recognition of multiple codes by determining the target code based on user intent through position information and decoding using corresponding decoders.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional long press recognition is used to identify graphic codes in an image, then the system can recognize graphic codes, but when multiple graphic codes are present, the terminal can only recognize one code, requiring manual recapture which reduces recognition efficiency

Engineering Contradiction:
Improvegraphic code recognition efficiencyVSAvoidtime for manual recapture
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent applies segmentation by dividing the image into multiple candidate regions, each potentially containing a graphic code. The system identifies multiple candidate graphic codes in different regions simultaneously, allowing the user to select from multiple options rather than requiring manual recapture of the entire image.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary action by automatically detecting and identifying multiple candidate graphic codes in the image before user selection. This pre-processing step of finding multiple candidates eliminates the need for manual recapture, as the system has already prepared multiple options for the user to choose from.

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If the terminal recognizes only one graphic code in an image with multiple codes, then the recognition process is simple, but the user may not get the expected code, requiring manual intervention

Engineering Contradiction:
Improveautomatic code selectionVSAvoidaccuracy of recognizing expected code
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent applies dynamics by making the target graphic code selection dynamic based on user interaction. The system initially presents multiple candidate codes, and the target code is determined based on user selection (such as long press position). This dynamic adaptation ensures both ease of operation and reliability, as the system can automatically select the correct code when the user provides selection input.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses feedback by monitoring user interaction (such as long press position) to determine which candidate graphic code is the target. The user's selection feedback allows the system to accurately identify the expected code among multiple candidates, improving reliability while maintaining ease of operation through automatic selection.

Inventive Principle:
Principle #23Feedback

3Productivity

If the system performs target detection on the entire image to identify multiple graphic codes, then multiple codes can be recognized, but the detection complexity increases

Engineering Contradiction:
Improvenumber of codes recognized simultaneouslyVSAvoiddetection system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent reduces detection complexity by segmenting the image into multiple regions and performing target detection in each region separately. This regional approach allows the system to identify multiple graphic codes simultaneously while keeping the detection process manageable by dividing it into smaller, independent tasks.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary region division before conducting target detection. By pre-segmenting the image into candidate regions, the system simplifies the subsequent detection process and can efficiently identify multiple graphic codes without overwhelming complexity in the detection algorithm itself.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3767519B1Graphic code recognition method and apparatus, and terminal, and storage medium
Publication Date: 2024.12.04 TENCENT TECHNOLOGY (SHENZHEN) CO LTD
  • EP3767519B1 patent drawingFigure 1~2
  • EP3767519B1 patent drawingFigure 3
  • EP3767519B1 patent drawingFigure 4~5

AI summary

A method for graphic code recognition includes: displaying a target image, the target image including at least two graphic codes; in response to receiving a graphic code recognition operation on the target image, obtaining graphic code position information of the at least two graphic codes in the target image; determining a target graphic code indicated by the graphic code recognition operation according to the graphic code position information; and displaying a target graphic code recognition result corresponding to the target graphic code.