Vehicle Light Source Pair Identification with Reliability Feedback

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing vehicle external environment recognition systems face challenges in accurately identifying paired light sources, such as tail lamps or brake lamps, especially at night, leading to reduced precision and potential erroneous recognition of preceding vehicles, which can cause unstable vehicle behavior.

Innovation Solution

A vehicle external environment recognition apparatus that includes a light source extraction processor, a light source pair identification processor, a degree of reliability derivation processor, and a light source pair re-identification processor, which extracts and identifies light sources, determines their reliability, and re-identifies pairs in the vicinity if initial pairs have low reliability, to enhance vehicle identification precision.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If light source pairs are identified based on positional relation in low-luminance conditions, then vehicle identification can be performed at night, but erroneous recognition increases and precision decreases

Engineering Contradiction:

Improvelow-luminance detection capabilityVSAvoidvehicle identification precision

Core Design Contradiction:

Illumination intensityVSMeasurement precision

Solution Approach 1:

The system derives a degree of vehicle reliability for identified light source pairs and uses this feedback to determine whether re-identification is necessary. When reliability is low, the system performs re-identification to correct potential errors, thereby maintaining precision even in low-luminance conditions.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary reliability assessment of light source pairs before final vehicle identification. By evaluating the degree of vehicle reliability in advance, the system can prepare for potential re-identification needs, ensuring accurate vehicle recognition even when initial detection confidence is low.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If re-identification of light source pairs is performed when reliability is low, then vehicle identification precision is improved, but processing time and computational load increase

Engineering Contradiction:

Improvevehicle identification precisionVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs re-identification only partially - specifically when the degree of vehicle reliability falls below a predetermined threshold. This selective approach ensures precision is improved only when necessary, avoiding unnecessary processing time and computational load for high-confidence detections.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system dynamically adjusts its identification process based on reliability assessment. When reliability is high, it proceeds directly to vehicle identification; when reliability is low, it dynamically triggers re-identification. This dynamic approach optimizes processing time while maintaining precision.

Inventive Principle:

Principle #15Dynamics

3Measurement precision

If multiple processing steps including reliability derivation and re-identification are implemented, then identification precision is improved, but device complexity increases

Engineering Contradiction:

Improvelight source pair identification precisionVSAvoidprocessing system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The degree of vehicle reliability derivation processor serves multiple functions: it assesses identification confidence, determines the need for re-identification, and provides feedback for quality control. This multi-functional component improves precision without proportionally increasing system complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The reliability derivation acts as an intermediary between light source extraction and final vehicle identification. This intermediate assessment layer coordinates the decision-making process for re-identification, managing system complexity while enabling precision improvement through structured multi-step processing.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11670093B2Vehicle external environment recognition apparatus

Publication Date: 2023.06.06 SUBARU CORP

AI summary

A vehicle external environment recognition apparatus includes a light source extraction processor, a light source pair identification processor, a degree of reliability derivation processor, a light source pair re-identification processor, and a vehicle identification processor. The light source pair identification processor identifies a first pair of light sources based on positional relation of extracted light sources. The degree of reliability derivation processor derives a degree of vehicle reliability of the first pair of light sources. The degree of vehicle reliability indicates how reliably the first pair of light sources is regarded as belonging to the identical vehicle. When the degree of vehicle reliability of the first pair of light sources is lower than a re-identification threshold, the light source pair re-identification processor identifies a second pair of light sources in the vicinity of the first pair of light sources having the degree of vehicle reliability lower than the re-identification threshold.