Adaptive Image Fusion for High Dynamic Range Mobile Photography

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Night photography in mobile devices faces challenges due to high dynamic range scenes, leading to overexposure and underexposure issues, and existing solutions for merging images with different exposures often result in mediocre imaging quality.

Innovation Solution

An image processing apparatus and method that selects different image processing modes based on the stability of the device and the dynamic range of the scene, using varying exposure values and fusion modes to capture and merge images effectively, including first, second, third, and fourth exposure fusion modes for stable and unstable conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple frames with different exposures are merged to capture high dynamic range scenes, then the dynamic range coverage is improved, but the image quality deteriorates due to overexposure and underexposure artifacts

Engineering Contradiction:
Improvedynamic range coverageVSAvoidimage quality
Core Design Contradiction:
Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The patent applies dynamics by making the exposure value selection adaptive rather than fixed. The system dynamically adjusts exposure values based on real-time scene analysis, identifying both bright and dark regions and selecting exposure values that optimize capture for each region. This dynamic adaptation allows the system to handle high dynamic range scenes without the static limitations of traditional multi-exposure merging, thereby improving image quality while maintaining broad dynamic range coverage.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the exposure value parameter based on scene characteristics. Instead of using fixed exposure values (e.g., -2, 0, +2 EV), the system analyzes the scene to identify bright and dark regions, then selects exposure values from a range that are optimized for those specific regions. This parameter adaptation allows the system to capture both bright and dark areas with appropriate exposure, reducing artifacts and improving overall image quality in high dynamic range scenes.

Inventive Principle:
Principle #35Parameter changes

2Device complexity

If a single exposure value is used for photographing, then the device complexity is reduced, but the image quality deteriorates in high dynamic range scenes due to overexposure or underexposure

Engineering Contradiction:
Improveprocessing complexityVSAvoidimage quality
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent implements self-service by enabling the system to automatically analyze scene characteristics and select optimal exposure values without requiring complex multi-frame merging processing. The scene analysis unit identifies bright and dark regions, and the exposure value selection unit automatically chooses appropriate values based on this analysis. This self-service approach simplifies the overall processing complexity while maintaining high image quality in high dynamic range scenes, eliminating the need for cumbersome multi-exposure workflows.

Inventive Principle:
Principle #25Self-service

3Loss of information

If multiple exposure frames are captured and merged, then the brightness information coverage is improved, but the noise level increases in low brightness areas

Engineering Contradiction:
Improvebrightness information coverageVSAvoidnoise level
Core Design Contradiction:
Loss of informationVSObject-generated harmful factors

Solution Approach 1:

The patent applies local quality by treating different regions of the scene differently based on their brightness characteristics. The scene analysis unit identifies specific bright and dark regions, and the exposure value selection unit chooses exposure values optimized for each region's characteristics. This localized approach ensures that dark regions are captured with sufficient brightness information while avoiding the noise amplification that occurs in traditional multi-exposure merging, thereby improving brightness information coverage without increasing noise levels.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3599760B1Image processing method and apparatus
Publication Date: 2023.02.08 HUAWEI TECH CO LTD
  • EP3599760B1 patent drawingFigure 1
  • EP3599760B1 patent drawingFigure 2
  • EP3599760B1 patent drawingFigure 3

AI summary

Embodiments of the present invention disclose an image processing apparatus, where the apparatus includes: an obtaining module, configured to obtain status information of a terminal device; an image photographing module, configured to obtain photographing scene information of the terminal device; a mode selection module, configured to determine an image processing mode based on the status information and the photographing scene information, where the image photographing module is further configured to obtain a to-be-displayed image; and a processing module, configured to process the to-be-displayed image based on the image processing mode, to obtain a processed image. The embodiments of the present invention further disclose an image processing method. According to the image processing method and the image processing apparatus in the embodiments of the present invention, different image processing modes may be selected based on the status information of the terminal device and a determining result of the photographing scene information, and different exposure values and different image fusion modes are used to obtain a better merged image.