Image Processing Device Using Weight Map for Region-Specific Brightness Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods fail to independently adjust the brightness of specific content within an image without affecting adjacent content, leading to difficulties in differentiating between interested and surrounding regions, and global or local gain processing either amplifies the entire image or influences nearby content.
Innovation Solution
An image processing method that sets multiple regions of interest, generates a weight map, applies different gain values to duplicate images, and synthesizes them using the weight map to control brightness while minimizing influence from other content, incorporating a weight smooth function for natural transitions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If global gain processing is applied to the input image, then the overall brightness of the entire image is improved, but the ability to independently adjust brightness of specific content is lost
Solution Approach 1:
The patent divides the image into multiple regions of interest (ROIs) based on object detection, creating separate processing zones for different objects. This segmentation allows independent gain application to each ROI while maintaining overall image coherence, resolving the contradiction between global brightness improvement and independent adjustment capability.
Solution Approach 2:
The patent applies different gain values to different regions of interest based on their specific brightness requirements. Each detected object receives localized brightness adjustment through its own gain parameter, enabling independent control of specific content brightness while preserving the ability to adjust overall image brightness when needed.
2Illumination intensity
If local gain processing is applied to adjust brightness of interested content, then the brightness of specific content is improved, but adjacent content is obviously influenced and differentiation becomes difficult
Solution Approach 1:
The patent introduces a mask map as an intermediary element that precisely defines the boundaries of each region of interest. The mask map acts as a mediator between the gain processing and the image data, ensuring that brightness adjustments are applied only to the intended ROI and preventing spill-over effects to adjacent content, thus maintaining clear differentiation between objects.
3Adaptability or versatility
If multiple gain values are applied to different regions, then flexible exposure control is achieved, but image artifacts and unnatural transitions may occur at region boundaries
Solution Approach 1:
The patent employs dynamic gradient transition processing at the boundaries of regions of interest. The transition zones between adjacent ROIs use interpolated gain values that smoothly blend between different gain levels, creating dynamic gradient effects that adapt to the local image content. This dynamic approach prevents abrupt transitions and eliminates visible artifacts while maintaining the flexibility of multi-ROI exposure control.
Data Source
AI summary
An image processing method of increasing image quality is applied to an image processing device and includes setting a plurality of regions of interest within an input image, generating a weight map in accordance with the plurality of regions of interest identified from the input image, applying a plurality of gain values to the input image for respectively generating a plurality of duplicated images, and utilizing the weight map to synthesize the plurality of duplicated images for acquiring an output image.