Camera Calibration Using Normal Vectors and Virtual Planes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing camera calibration methods face accuracy issues when obstacles are present in the real space, requiring large calibration equipment and complex 3D coordinate measurements, especially in wide and complicated target areas.

Innovation Solution

A camera calibration device that estimates camera parameters using freely-arranged reference points, acquiring normal vectors from images to estimate rotation and translation matrices without the need for special calibration equipment or 3D coordinate measurements, by projecting vectors on a virtual plane and optimizing based on perpendicularity and re-projection errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If special calibration equipment with known arrangement of reference points is installed, then camera parameters can be estimated using projection equations, but installation place is restricted by obstacles and reference points cannot be evenly acquired from shot image

Engineering Contradiction:
Improvecamera calibration accuracyVSAvoidadaptability to environments with obstacles
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent uses markers (copies of reference points) placed on existing objects in the environment instead of requiring special calibration equipment. These markers can be freely arranged on any objects within the camera's field of view, eliminating the need for dedicated calibration equipment and allowing calibration in environments with obstacles.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent separates the reference points from their traditional calibration equipment context by placing markers on various existing objects in the environment. This segmentation allows reference points to be distributed throughout the entire field of view rather than being confined to a single calibration device, enabling even distribution of reference points across the image.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If piece of calibration equipment is used, then reference points arrangement is known, but when target real space is wide, practically-impossible large calibration equipment is required to acquire reference points from everywhere on the screen

Engineering Contradiction:
Improvecamera calibration accuracyVSAvoidsize of calibration equipment
Core Design Contradiction:
Measurement precisionVSLength of stationary object

Solution Approach 1:

The patent makes the calibration system universal by allowing markers to be placed on any objects within the environment rather than requiring a specific calibration device. This multi-functionality enables calibration in both small and large spaces using the same marker-based approach, eliminating the need for oversized calibration equipment in wide target spaces.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent transitions from using a single calibration device in one location to distributing multiple markers throughout the three-dimensional space being calibrated. By placing markers on various objects at different positions and orientations, the system achieves comprehensive coverage of wide target spaces without requiring excessively large calibration equipment.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Ease of operation

If freely-arranged reference points are used, then easiness of camera calibration is improved, but 3D coordinates have to be measured one by one in the real space which is very complicated in wide and complicated target space

Engineering Contradiction:
Improveease of camera calibrationVSAvoidcomplexity of 3D coordinate measurement
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent uses markers as simplified copies of reference points that can be easily identified in images. These markers have known geometric characteristics that allow automatic detection and 3D coordinate calculation without manual measurement, maintaining ease of operation while eliminating the complexity of measuring 3D coordinates for freely-arranged reference points.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent changes the parameters of the reference points by using markers with specific, known geometric properties (such as circular shapes or specific patterns). This allows the system to automatically determine 3D coordinates from 2D image data through geometric constraints, eliminating the need for manual 3D measurement while preserving the flexibility of freely-arranged reference points.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2808645B1Camera calibration device, camera calibration method, and camera calibration program
Publication Date: 2019.02.20 NEC CORP
  • EP2808645B1 patent drawingFigure 1
  • EP2808645B1 patent drawingFigure 2(a)~2(b)
  • EP2808645B1 patent drawingFigure 3

AI summary

There is provided a camera calibration device capable of highly precisely and easily estimating camera parameters without installing a piece of special calibration equipment or measuring 3D coordinates of reference points employed for calibration. A normal vector acquisition means 81 acquires normal vectors perpendicular to a reference horizontal plane from an image of a camera to be calibrated. A rotation matrix estimation means 82 projects the acquired normal vectors on a projection virtual plane perpendicular to the reference horizontal plane and evaluates that the projected normal vectors are perpendicular to the reference horizontal plane thereby to estimate a rotation matrix employed for calibrating the camera.