Camera Cluster Calibration for Real-Time 3D Imaging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current image processing methods for capturing three-dimensional spaces, such as those using fisheye lenses or video stitching programs, face limitations in resolution and real-time capability, particularly failing to provide live streaming due to time-consuming stitching processes and the need for normalized, flat field calibration.

Innovation Solution

An image processing method utilizing a camera cluster with multiple camera devices, where each device is calibrated individually and collectively, allowing for flexible positioning and orientation, enabling the generation of seamless, shadow-free images with a 360° field of view by applying calibration parameters for image distortion, brightness, and color correction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If video stitching programs are used to create full spherical images, then the field of view coverage is improved, but the processing time increases significantly making real-time display impossible

Engineering Contradiction:
Improvefield of view coverageVSAvoidprocessing time
Core Design Contradiction:
Area of stationary objectVSLoss of time

Solution Approach 1:

The patent performs image distortion correction, brightness correction, and color correction in advance during a calibration phase. The corrected images are stored with pre-calculated transformation parameters. During live streaming, only simple coordinate mapping and blending are needed, enabling real-time display while maintaining full spherical coverage.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent divides the full spherical image into multiple overlapping images from different cameras positioned at specific locations. Each camera captures a specific region with known distortion characteristics. The images are processed and blended in real-time using pre-calculated parameters, avoiding the need to process the entire spherical image at once.

Inventive Principle:
Principle #1Segmentation

2Area of stationary object

If fisheye lenses are used to capture wide field of view, then the coverage area is improved, but the image resolution is severely limited by the camera chip

Engineering Contradiction:
Improvefield of view coverageVSAvoidimage resolution
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

Instead of using a single fisheye lens that spreads pixels across the entire field of view resulting in low resolution, the patent uses multiple standard cameras positioned at different locations. Each camera captures a region with sufficient resolution, and the overlapping regions are blended to create a complete spherical view with uniformly high resolution across all areas.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from a single-camera 2D sensor limitation to a multi-camera 3D spatial arrangement. By positioning cameras at different spatial locations and combining their views, the system achieves both wide field of view coverage and high resolution simultaneously, overcoming the fundamental limitation of single-sensor fisheye systems.

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

3Area of stationary object

If multiple camera devices are used to image three-dimensional space, then the field of view coverage is improved, but the system complexity increases

Engineering Contradiction:
Improvefield of view coverageVSAvoidcamera system complexity
Core Design Contradiction:
Area of stationary objectVSDevice complexity

Solution Approach 1:

The patent uses identical standard camera devices for multiple purposes: each camera serves as both a primary imaging device and contributes to the overall spherical coverage. The same calibration and correction procedures apply to all cameras, and the same blending algorithm combines all inputs. This universal approach simplifies the system compared to using specialized cameras for different functions.

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

Solution Approach 2:

The patent positions cameras and defines blending regions such that all cameras have equivalent roles and contributions in the final spherical image. No single camera is privileged or requires special processing; all follow the same calibration, correction, and blending protocol. This equipotential design simplifies the control system and makes the multi-camera system as manageable as a single-camera system.

Inventive Principle:
Principle #12Equipotentiality

Data Source

PatentEP3427474B1Image processing method, image processing means and image processing device for generating images of a portion of a three-dimensional space
Publication Date: 2023.02.22 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
  • EP3427474B1 patent drawingFigure 1
  • EP3427474B1 patent drawingFigure 2A
  • EP3427474B1 patent drawingFigure 2B

AI summary

The invention relates to an image processing method for generating images of a portion of a three-dimensional space (100) with a camera cluster (10) having at least two camera devices (1, 2, 3, 4, 5) and/or with a camera cluster (10) having at least one camera device (1, 2, 3, 4, 5) which is displaceable along a spatially known path with a known movement variable. The invention is characterized in that a) the camera cluster (10) is positioned in the three-dimensional space (100), and that b) a calibration of the parameters defining the recording characteristics of the camera device (1, 2, 3, 4, 5) is carried out by the recording of calibration markings (21) arranged in the three-dimensional space (100) and by an assignment of the calibration markings (21) in the three-dimensional space (100) to at least one recording generated by the camera cluster (10), wherein the positions of the calibration markings (21) are known relative to each other in a coordinate system, and that c) values for a correction of image distortion, brightness and/or color for at least a focus area (16) with predetermined spatial distances (D) to the camera cluster (10) are calculated by means of the at least one recording generated by the camera cluster (10). The invention also relates to image processing means and to an image processing device.