Dynamic Camera Grouping for Image Harmonization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing multi-camera image processing systems face challenges in harmonizing images from multiple cameras, particularly in varying brightness and color temperature conditions, leading to visible seam lines and inefficient resource utilization.

Innovation Solution

A system on a chip (SoC) with an image signal processor (ISP) dynamically groups cameras based on measured view brightness and color temperature, assigning the same exposure/gain setting to cameras in each group to balance image harmonization and dynamic range, eliminating the need for external ISPs and complex post-processing methods.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If cameras are assigned individual exposure/gain settings to preserve dynamic range, then dynamic range is improved, but image harmonization deteriorates causing visible seam lines

Engineering Contradiction:
Improvedynamic rangeVSAvoidimage harmonization
Core Design Contradiction:
Illumination intensityVSManufacturing precision

Solution Approach 1:

The patent segments cameras into multiple groups based on their observed brightness and color temperature characteristics. Each group is assigned a common exposure/gain setting, while different groups can have different settings. This segmentation allows the system to preserve dynamic range across diverse camera views while maintaining image harmonization within each group, thereby resolving the contradiction between individual camera optimization and overall image consistency.

Inventive Principle:
Principle #1Segmentation

2Manufacturing precision

If external ISPs and complex post-processing methods are used to harmonize images, then image harmonization is improved, but device complexity and cost increase

Engineering Contradiction:
Improveimage harmonizationVSAvoidsystem complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent merges the statistics engine and image signal processing functions into a single integrated device. The statistics engine collects brightness and color temperature data from multiple cameras, and the image signal processor uses this statistics to dynamically assign exposure/gain settings. This consolidation eliminates the need for external ISPs and complex post-processing systems, achieving image harmonization through a unified, cost-effective architecture.

Inventive Principle:
Principle #5Merging (Combining)

3Manufacturing precision

If advanced multi-camera harmonization systems are implemented, then image quality is improved, but power consumption and cost increase

Engineering Contradiction:
Improveimage qualityVSAvoidpower consumption
Core Design Contradiction:
Manufacturing precisionVSUse of energy by moving object

Solution Approach 1:

The patent implements a self-service system where the integrated device automatically collects statistics from camera images and dynamically adjusts exposure/gain settings without requiring external processing systems. The statistics engine continuously monitors brightness and color temperature, and the image signal processor autonomously assigns appropriate settings to camera groups. This self-contained approach achieves high image quality while minimizing power consumption and eliminating the need for additional external hardware.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3729803B1Multi camera image processing
Publication Date: 2024.06.26 TEXAS INSTRUMENTS INC
  • EP3729803B1 patent drawingFigure 1
  • EP3729803B1 patent drawingFigure 2
  • EP3729803B1 patent drawingFigure 3

AI summary

In described examples, a system on a chip (SoC) (102) implements dynamic grouping of multiple cameras (105) in an image processing system (100). The SoC (102) includes an image signal processor (ISP) (104) configured to receive a signal corresponding to an image from two or more cameras (105), dynamically group the cameras (105) into groups based on a measured view brightness and color temperature observed by each of the cameras (105), and assign cameras (105) with a group of the groups to a same exposure/gain setting. The SoC (102) further includes a statistics engine (106) configured to receive the image signals and statistics related to the image signals from the ISP (104) and determine a measurement of image brightness and color temperature of each image based on the image signals for forwarding to the ISP (104).