Image Processing Device Feature Point Extraction for Robot Positioning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing image processing devices face increased processing load and time in recognizing the position and orientation of objects due to the need to perform conversion processes on all points of captured image data.

Innovation Solution

The image processing device employs a storage section for a three-dimensional shape model with feature amounts and positional information, an extraction process to extract feature amounts and positional information from two-dimensional images, and a recognition process to match feature points with the three-dimensional model, reducing the processing load by focusing only on feature points.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conversion process is performed on all points of captured image data, then position and orientation recognition accuracy is improved, but processing load increases

Engineering Contradiction:
Improveposition and orientation recognition accuracyVSAvoidprocessing load
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only feature points from the captured image data instead of processing all points. The feature point extraction unit identifies and extracts characteristic points from the image, and the position/orientation recognition unit then processes only these extracted feature points to determine the object's position and orientation, significantly reducing processing load while maintaining recognition accuracy

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the image processing task into distinct stages: feature point extraction and position/orientation recognition. By dividing the processing into these separate functional units, the system can focus computational resources on the most critical aspects of the task, improving efficiency without sacrificing accuracy

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If conversion process is performed on all points of captured image data, then position and orientation recognition accuracy is improved, but processing time increases

Engineering Contradiction:
Improveposition and orientation recognition accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only feature points from the captured image data instead of processing all points. The feature point extraction unit identifies and extracts characteristic points from the image, and the position/orientation recognition unit then processes only these extracted feature points to determine the object's position and orientation, significantly reducing processing load while maintaining recognition accuracy

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by processing only the necessary feature points rather than all image data points. This selective approach processes just enough information to achieve accurate position and orientation recognition without the excessive processing time that would result from analyzing every pixel in the image

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS11972589B2Image processing device, work robot, substrate inspection device, and specimen inspection device
Publication Date: 2024.04.30 FUJI CORP
  • US11972589B2 patent drawing
  • US11972589B2 patent drawing
  • US11972589B2 patent drawing

AI summary

The image processing device comprises: a storage section storing a three-dimensional shape model in which feature amounts and three-dimensional positional information, for multiple feature points of a target object, are associated; an extraction process section configured to extract the feature amounts and two-dimensional positional information of the feature points from a two-dimensional image of the target object captured with a camera; and a recognition process section configured to identify three-dimensional positional information of the feature points of the two-dimensional image and recognize the position and orientation of the target object by matching the feature points of the two-dimensional image with the feature points of the three-dimensional model using the feature amounts.