HDR Image Capture via Multi-ROI Exposure Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional high-dynamic range (HDR) imaging techniques often result in images with limited brightness range, where objects in shaded or brightly lit areas may be over or underexposed, leading to unsatisfactory capture of scenes with multiple objects of interest, as they rely on a single region of interest (ROI) for exposure determination and fixed exposure values.
Innovation Solution
A system that identifies multiple regions of interest (ROIs) in a preview image and determines specific exposure values for each, allowing for the capture of multiple images with varying exposure levels, which are then blended to create an HDR image, ensuring no ROI is too dark or too light.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If a single region of interest (ROI) is used for exposure determination in conventional HDR imaging, then the exposure values are fixed and simple to determine, but objects in shaded or brightly lit areas may be over or underexposed, resulting in limited brightness range
Solution Approach 1:
The image is divided into multiple regions of interest (ROIs), each with its own exposure determination. The system identifies multiple ROIs based on object detection and segmentation algorithms, allowing different exposure values to be assigned to different regions, thereby capturing both shaded and brightly lit areas effectively without overwhelming complexity
Solution Approach 2:
Different exposure values are assigned to different regions of interest based on their local characteristics. Each ROI receives a customized exposure value determined by its specific lighting conditions and object importance, enabling localized optimization of brightness range while maintaining overall image quality
2Illumination intensity
If multiple images with different exposure values are captured for HDR imaging, then the brightness range is improved, but the number of images to be captured and processed increases, affecting processing time and complexity
Solution Approach 1:
The system dynamically determines the number of images to capture based on the number of identified ROIs and their brightness characteristics. Instead of using a fixed number of exposure values, the system adapts the capture sequence length to match the scene complexity, reducing unnecessary captures and processing time while maintaining HDR quality
Solution Approach 2:
The system changes exposure values dynamically for each ROI based on local brightness measurements and object characteristics. By adjusting exposure parameters adaptively rather than using fixed intervals, the system captures optimal images fewer times, reducing processing time while achieving the desired brightness range
3Illumination intensity
If conventional HDR imaging uses fixed exposure values, then the capture process is simple, but the exposure values are not optimal for multiple objects with varying brightness levels
Solution Approach 1:
The system uses feedback from brightness measurements of each ROI to determine optimal exposure values. By continuously monitoring the brightness characteristics of identified objects and adjusting exposure values accordingly, the system achieves optimal exposure for multiple objects with varying brightness levels while maintaining a manageable determination process through iterative refinement
Data Source
AI summary
Aspects of the present disclosure relate to systems and methods for generating High-Dynamic Range (HDR) images. An example device may include one or more processors and a memory. The memory may include instructions that, when executed by the one or more processors, cause the device to identify a number of regions of interest (ROIs) in a preview image, wherein the number of ROIs is more than one, determine a number of images to be captured for an HDR image to be the number of identified ROIs, and for each ROI associated with a respective image of the number of images to be captured, determine an exposure value to be used in capturing the respective image of the number of images.