Lane-Aware Road Sign Recognition for Branch Road Misclassification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional vehicle systems struggle to accurately distinguish road signs for branch roads from the main travel lane, often misidentifying signs due to their placement above the branch road, leading to erroneous detection.

Innovation Solution

An image recognition apparatus that includes an image information acquirer, a road sign recognizer, and a traffic lane recognizer, with a memory to store traffic lane information in a time series, and a determiner to identify the most suitable traffic lane for a road sign based on this data, ensuring accurate recognition even when lanes become unrecognizable.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a road sign is placed above a branch road to indicate a different speed limit, then the road sign can be detected by the camera system, but the system erroneously detects it as a road sign for the main traveling road

Engineering Contradiction:
Improveroad sign detection accuracyVSAvoidroad sign classification reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent segments the road space into multiple traffic lanes (main traveling road and branch road) and associates detected road signs with specific lanes based on spatial relationships. By dividing the recognition task into lane-specific segments, the system can distinguish whether a road sign belongs to the main road or a branch road, thereby resolving the contradiction between detecting all road signs and accurately classifying them.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces traffic lane information as an intermediary element that mediates between the detected road sign and its classification. The lane recognition results serve as a bridge to determine which traffic lane the road sign corresponds to, enabling accurate attribution of road signs to either the main traveling road or branch road, thus improving classification reliability.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the vehicle travels for a long time on the motorway, then the branch road becomes difficult to detect from camera images, but road signs above the branch road remain detectable

Engineering Contradiction:
Improvebranch road detection precisionVSAvoidtraffic lane information loss
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent performs preliminary recognition and storage of traffic lane information when the branch road is still visible and recognizable in camera images. By capturing and storing lane configuration data in advance, the system preserves traffic lane information even when the branch road later becomes undetectable due to prolonged travel, preventing information loss while maintaining detection precision during the recognition window.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12354377B2Image recognition apparatus, driver assistance system and image recognition method
Publication Date: 2025.07.08 DENSO CORP
  • US12354377B2 patent drawing
  • US12354377B2 patent drawing
  • US12354377B2 patent drawing

AI summary

An image recognition apparatus comprises an image information acquirer that acquires image information in a given cycle by capturing an image of a traveling path along which an own vehicle travels, a road sign recognizer that recognizes a road sign based on the image information and a traffic lane recognizer that recognizes traffic lanes located in the traveling path based on the image information. A memory is provided in the image recognition apparatus to store traffic lane information about a traffic lane recognized by the traffic lane recognizer in a time series. A determiner is also provided to determine if the road sign recognized by the road sign recognizer is for an own traffic lane by choosing a most suitable traffic lane among the traffic lanes recognized for a given period based on the traffic lane information stored in the time series.