Ground Feature Position Identification Using Multi-Viewpoint Image Analysis

Solution Overview

Problem

Existing systems struggle to accurately identify the position of ground features like pedestrian bridges in real-time, leading to potential unnecessary traffic assistance alerts.

Innovation Solution

The proposed system utilizes an assistance device that acquires images from multiple viewpoints, analyzes these images to determine the presence or absence of ground features, and calculates their positions based on differences in image content and viewpoint positions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the system uses existing map information to identify ground features, then the process is simple, but the position identification accuracy is insufficient and cannot reflect real-time changes

Engineering Contradiction:

Improveground feature position identification accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent creates a virtual map by capturing images from multiple viewpoints and synthesizing a three-dimensional representation of the environment. This virtual map copies the actual ground feature positions and allows for accurate identification without requiring complex real-time processing of all captured images simultaneously.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system performs preliminary actions by capturing images from multiple viewpoints in advance and pre-processing them to create a three-dimensional virtual map. This preliminary map construction enables accurate ground feature position identification when needed, without requiring complex real-time analysis.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If the system provides traffic assistance alerts based on incomplete ground feature position information, then the response time is fast, but unnecessary alerts are generated reducing reliability

Engineering Contradiction:

Improvetraffic assistance alert accuracyVSAvoidtime for position identification

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by capturing images from multiple viewpoints in advance and pre-processing them to create a three-dimensional virtual map. This preliminary map construction enables accurate ground feature position identification when needed, without requiring complex real-time analysis.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent creates a virtual map by capturing images from multiple viewpoints and synthesizing a three-dimensional representation of the environment. This virtual map copies the actual ground feature positions and allows for accurate identification without requiring complex real-time processing of all captured images simultaneously.

Inventive Principle:

Principle #26Copying

3Measurement precision

If the system uses a single viewpoint image to identify ground features, then the processing is simple, but the position determination is inaccurate

Engineering Contradiction:

Improveground feature position accuracyVSAvoidimage processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the image processing task by capturing images from multiple discrete viewpoints and processing each image independently to extract feature information. These segmented results are then combined to create an accurate three-dimensional virtual map, achieving high precision without overwhelming complexity in any single processing step.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system transitions from two-dimensional single-viewpoint images to three-dimensional virtual map representation. By adding the vertical dimension and integrating multiple viewpoints, the system achieves accurate ground feature position identification that cannot be obtained from single-viewpoint images alone.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS20250046102A1Ground feature position information output device, assistance control device, ground feature position information output method, and computer-readable storage medium

Publication Date: 2025.02.06 HONDA MOTOR CO LTD

AI summary

A ground feature position information output device includes: an image information acquisition unit which acquires a plurality of images from a plurality of viewpoints different from each other on a roadway and first information indicating respective viewpoint positions and viewing directions of the plurality of images; a ground feature position identification unit which, based on a difference in states of figures of a ground feature in each of the plurality of images and the respective viewpoint positions and viewing directions of the plurality of images identified from the first information, identifies a position of the ground feature; and a ground feature position information output unit which outputs information indicating the position of the ground feature identified by the ground feature position identification unit.