Surround-View Camera Image Enhancement for Low-Light Recognition

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing vehicle surround-view systems struggle with image recognition and display in low-light conditions without additional lighting, leading to performance deterioration and increased costs due to the need for multiple lamps.

Innovation Solution

A method using a convolutional neural network (CNN) trained with image pairs of varying brightness to enhance image data from multiple vehicle cameras, adjusting brightness and color distribution to create optimized output images, which are then combined for improved display and recognition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If additional lamps are installed to illuminate side regions, then image quality in low-light conditions is improved, but device complexity and cost increase

Engineering Contradiction:

Improveimage brightnessVSAvoidsystem complexity

Core Design Contradiction:

Illumination intensityVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical/optical system of additional physical lamps with an algorithmic image processing system. A neural network-based image processing unit analyzes and enhances images from existing cameras, adjusting brightness and contrast through computational methods rather than physical illumination, thereby avoiding the complexity and cost of additional lighting hardware

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the parameters of the image data itself (brightness, contrast, color distribution) through algorithmic processing rather than changing the physical lighting conditions. The image processing unit modifies image parameters post-capture to achieve the effect of improved illumination without requiring additional light sources

Inventive Principle:

Principle #35Parameter changes

2Illumination intensity

If additional lamps are installed to illuminate side regions, then image quality in low-light conditions is improved, but manufacturing cost increases

Engineering Contradiction:

Improveimage brightnessVSAvoidmanufacturing cost

Core Design Contradiction:

Illumination intensityVSEase of manufacture

Solution Approach 1:

The patent substitutes physical lighting hardware with software-based image enhancement, eliminating the need to manufacture, install, and maintain additional lamps. The solution uses computational algorithms running on existing processing units, significantly reducing manufacturing costs while achieving the same functional outcome of improved image brightness in low-light conditions

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Illumination intensity

If algorithmic methods like gamma correction are used to lighten images, then image brightness is improved, but color information is lost

Engineering Contradiction:

Improveimage brightnessVSAvoidcolor information

Core Design Contradiction:

Illumination intensityVSLoss of information

Solution Approach 1:

The patent applies sophisticated parameter changes through neural network-based image processing that simultaneously adjusts multiple image parameters (brightness, contrast, color distribution) in a coordinated manner. Unlike simple gamma correction that only adjusts brightness, the patent's method modifies color distribution and other parameters together to preserve color information while improving brightness

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent combines multiple image processing functions (brightness adjustment, contrast enhancement, color correction) into a composite processing system. The neural network integrates these different processing operations to achieve overall image enhancement that preserves color information while improving brightness, rather than applying single-function corrections

Inventive Principle:

Principle #40Composite materials

Data Source

PatentUS12506840B2Converting input image data from a plurality of vehicle cameras of a surround-view system into optimised output image data

Publication Date: 2025.12.23 CONTINENTAL AUTONOMOUS MOBILITY GERMANY GMBH

AI summary

The present disclosure relates to a machine learning method, to a method and to a device for converting input image data from a plurality of vehicle cameras of a panoramic-view system into optimized or enhanced output image data. The method for converting input image data from a plurality of vehicle cameras of a panoramic-view system into optimized or enhanced output image data includes input image data acquired by the vehicle cameras and having a current brightness or color distribution provided to a trained artificial neural network. the trained artificial neural network is configured to convert the input image data having the current brightness or color distribution into optimized or enhanced output image data having different output brightness or color distribution; and the trained artificial neural network is configured to output the output image data.