Adaptive WDR Image Fusion for Motion Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital image processing apparatuses face challenges in capturing images with wide dynamic range (WDR) due to motion artifacts, especially when capturing scenes with large luminance differences, leading to deteriorated image quality.
Innovation Solution
An image processing apparatus and method that determine whether to obtain a first WDR image maximizing dynamic range or a second WDR image with corrected motion artifacts by comparing short exposure image frames with a long exposure image frame, selecting the frame with the least motion change, and fusing them to output a high-quality WDR image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If WDR processing is performed by fusing short exposure and long exposure images, then the dynamic range is improved and both bright and dark portions are clearly viewable, but motion artifacts are generated when moving objects are present in the scene
Solution Approach 1:
The patent applies dynamics by making the exposure time selection adaptive rather than fixed. The system dynamically determines whether to use short exposure or long exposure images based on motion detection results, allowing the WDR processing to adapt to different scene conditions. This resolves the contradiction by enabling the system to switch between exposure modes depending on whether motion is present, thus maintaining both wide dynamic range and image quality.
Solution Approach 2:
The patent changes the exposure time parameter based on motion detection. When motion is detected in specific regions, the system uses short exposure images for those regions; when no motion is detected, it uses long exposure images. This parameter change strategy allows the system to optimize both dynamic range and motion artifact reduction by adjusting the exposure time parameter according to scene conditions.
2Illumination intensity
If short exposure time is used, then bright portions are clearly viewed, but dark portions lose sufficient information
Solution Approach 1:
The patent applies local quality by performing WDR processing on different regions of the image with different exposure times. Motion detection is performed on specific regions, and the system selectively applies short exposure or long exposure images to different regions based on motion presence. This allows bright portions to be captured with short exposure while dark portions are captured with long exposure, optimizing both regions independently.
Solution Approach 2:
The patent segments the image processing into region-based operations. The image is divided into multiple regions, and motion detection is performed independently for each region. Based on the motion detection results of each region, the system selectively fuses short or long exposure images for that specific region, rather than applying a uniform exposure time to the entire image.
3Illumination intensity
If long exposure time is used, then dark portions are clearly viewed, but bright portions become saturated
Solution Approach 1:
The patent applies local quality by using different exposure times for different regions based on motion detection. For regions with motion, short exposure images are used to prevent saturation in bright portions. For regions without motion, long exposure images are used to capture dark portions. This local differentiation resolves the contradiction by optimizing exposure for each region's specific requirements.
4Device complexity
If conventional WDR processing is applied without motion detection, then processing complexity is low, but motion artifacts significantly deteriorate image quality
Solution Approach 1:
The patent applies preliminary action by performing motion detection before the WDR fusion process. The motion detection is conducted on intermediate images during the exposure sequence, and the results are used to determine which exposure images to fuse. This preliminary motion analysis allows the system to prepare the appropriate fusion strategy in advance, preventing motion artifacts while maintaining reasonable processing complexity.
Solution Approach 2:
The patent implements feedback by using motion detection results to control the WDR fusion process. The motion detection continuously monitors the scene and provides feedback that determines whether to use short or long exposure images for fusion. This feedback mechanism allows the system to adapt to motion conditions and adjust the fusion strategy accordingly, improving image quality while managing processing complexity through intelligent control.
Data Source
AI summary
An apparatus for processing images includes an image sensor to output a long exposure image frame based upon a first exposure time and at least two short exposure image frames having a single, second exposure time different from the first exposure time; a mode determination unit to determine the second exposure time based on the number of saturated pixels in the image frame output from the image sensor; a shutter controller to control the exposure time of each frame of the image sensor based on an output signal from the mode detection unit; and an image processing unit to fuse the long exposure image frame and the at least two short exposure image frames to output a wide dynamic range (WDR) image. Methods for processing WDR images are also disclosed.