Object Recognition via Oriented 3D Point Cloud Projection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for identifying objects from three-dimensional point cloud data require significant computational resources and may not accurately identify objects based on their shape or orientation.

Innovation Solution

An image generator that projects three-dimensional point cloud data onto a reference plane oriented depending on the detectable object to generate a two-dimensional image, reducing computational load while enhancing identification accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If three-dimensional point cloud data is projected onto the horizontal plane only, then the projection process is simple, but object identification accuracy deteriorates depending on shape or orientation

Engineering Contradiction:

Improveprojection process simplicityVSAvoidobject identification accuracy

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent applies dynamics by making the reference plane orientation adjustable and adaptable. Instead of using a fixed horizontal plane, the system dynamically determines the optimal reference plane orientation based on object characteristics, allowing the projection approach to adapt to different object shapes and orientations while maintaining identification accuracy

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the orientation parameter of the reference plane to resolve the contradiction. By varying the reference plane orientation from fixed horizontal to object-dependent angles, the system achieves both accurate object identification and maintains computational efficiency through selective projection

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If object is identified directly using three-dimensional point cloud data, then identification accuracy is maintained, but computational load increases hugely

Engineering Contradiction:

Improveobject identification accuracyVSAvoidcomputational load

Core Design Contradiction:

Measurement precisionVSPower

Solution Approach 1:

The patent extracts essential object features by projecting three-dimensional point cloud data onto a two-dimensional reference plane. This extraction process retains the critical shape and orientation information needed for identification while removing redundant spatial dimensions, significantly reducing computational load

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent creates a two-dimensional projected copy of the three-dimensional point cloud data onto a reference plane. This copy preserves the essential geometric features needed for object identification while requiring far less computational resources to process than the original three-dimensional data

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20250356451A1Image generating device, object recognition device, and image generating method

Publication Date: 2025.11.20 KAWASAKI JUKOGYO KK

AI summary

This image generating device generates an image for identifying a predetermined search object. The image generating device is provided with an input unit and a processing device. The input unit inputs three-dimensional point group data generated by a sensor searching the surroundings. The processing device projects the three-dimensional point group data input in the input unit on a reference plane having a direction set according to the search object and thereby generates a two-dimensional image.