Monocular Camera 3D Detection Using Intersecting Light Beams

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Industrial robots face challenges in accurately detecting the three-dimensional position and posture of objects in real time, especially when using monocular cameras, as they often require complex processing and take significant time, limiting their productivity and flexibility in tasks like fast picking and handling.

Innovation Solution

A detection apparatus equipped with a monocular camera and a light beam irradiation unit that calculates the distance and three-dimensional position of an object by projecting light beams at intersecting angles, using image processing to create profile data and interpolate or extrapolate the object's position, enabling real-time processing and accurate object detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a compound eye camera is used to obtain three-dimensional position information, then measurement precision is improved, but processing time increases significantly (about 0.4 seconds)

Engineering Contradiction:
Improvethree-dimensional position detection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the three-dimensional detection problem into two separate two-dimensional image acquisitions taken at different time points. By capturing images at different moments and analyzing the temporal change in object position, the system achieves three-dimensional detection without requiring complex compound eye camera processing, thus reducing processing time while maintaining measurement precision.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces time as an intermediary variable to enable three-dimensional detection. By using the temporal change in two-dimensional image positions as a mediator, the system can infer depth information without directly capturing it through complex optical systems, thereby achieving fast three-dimensional detection with a simple monocular camera.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If a monocular camera is used to reduce processing time, then productivity is improved, but the ability to accurately detect three-dimensional position is degraded

Engineering Contradiction:
Improveoperational speedVSAvoidthree-dimensional position detection accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent applies dynamics by capturing images at different time points rather than using a static single-image approach. By analyzing the dynamic change in object position between two time points, the monocular camera can infer three-dimensional information, achieving both high productivity through simple hardware and accurate measurement through temporal analysis.

Inventive Principle:
Principle #15Dynamics

3Reliability

If sequential images are used to monitor work quality and detect pickup failures, then reliability is improved, but processing complexity increases

Engineering Contradiction:
Improvework quality monitoring capabilityVSAvoidprocessing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent maintains continuous useful action by using sequential images to monitor work quality, pickup failures, and work completion. By continuously analyzing the temporal sequence of images, the system ensures reliable detection of work quality issues without requiring additional sensors or complex processing systems, achieving both high reliability and simple processing.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP2940422B1Detection apparatus, detection method and manipulator
Publication Date: 2020.07.22 KIKAI SEKKEI NAKAHATA
  • EP2940422B1 patent drawingFigure 1
  • EP2940422B1 patent drawingFigure 2
  • EP2940422B1 patent drawingFigure 3A

AI summary

A detection unit comprising a manipulator, a imaging unit, a light beam irradiation unit and a computing unit, the imaging unit and the light beam irradiation unit are provided on a manipulator. The detection apparatus includes a horizontal imaging plane, which includes an optical axis of the imaging unit, and a vertical imaging plane, which includes the optical axis of the imaging unit and 1) a projection of a light beam from the light beam irradiation unit on the horizontal imaging plane and 2) a projection of a light beam from the light beam irradiation unit on the vertical imaging plane respectively form an intersecting angle with the optical axis of the imaging unit.