Image Processing Apparatus High-Saturation Pixel Combination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques for combining images taken in rapid succession, such as during fireworks or astronomical photography, often result in image quality deterioration due to long shutter speeds and fail to effectively manage high saturation areas, leading to unwanted noise and loss of detail in bright regions.
Innovation Solution
An image processing apparatus and method that includes a saturation judgment section to identify high-saturation pixels and performs high-saturation comparative bright combination processing by replacing these pixels with values from pixels with a larger luminance component, thereby enhancing image vividness and reducing noise in bright areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If long shutter speed is used for photographing to capture bright moving objects, then the brightness of the image is improved, but image quality deteriorates due to dark current noise
Solution Approach 1:
The patent segments the image processing into multiple short-exposure images captured in time sequence, then combines them through comparative bright combination processing. This divides the single long-exposure shot into multiple short-exposure shots, capturing brightness while avoiding dark current noise accumulation.
Solution Approach 2:
The patent merges multiple short-exposure images through comparative bright combination processing, where pixel values from multiple images are compared and the brighter values are retained. This combines the brightness benefits of multiple exposures while eliminating dark current noise present in each individual short exposure.
2Reliability
If conventional comparative bright combination processing is applied to all pixels, then noise is reduced, but high saturation areas lose detail and become overly bright
Solution Approach 1:
The patent applies different processing strategies to different regions of the image based on saturation levels. High saturation areas (fireworks, headlights) are processed differently from low saturation areas (sky, background), preserving detail in bright regions while still achieving noise reduction in other areas.
Solution Approach 2:
The patent applies comparative bright combination processing selectively rather than universally. By identifying high saturation areas and applying different processing to these regions versus other regions, it achieves partial action that preserves detail where needed while still providing noise reduction elsewhere.
3Reliability
If short shutter speed is used to avoid dark current noise, then image quality is maintained, but the brightness and vividness of the image deteriorates
Solution Approach 1:
The patent combines multiple short-exposure images through comparative bright combination processing, merging the brightness information from each exposure. This achieves the brightness equivalent of a long exposure while maintaining the image quality benefits of short exposures by eliminating dark current noise through the combination process.
4Device complexity
If conventional image processing is used without saturation judgment, then processing is simpler, but low saturation elements like smoke and clouds are not removed, reducing image vividness
Solution Approach 1:
The patent performs saturation judgment as a preliminary step before the main image combination processing. By identifying high saturation areas in advance, the processing can be optimized to preserve these regions while applying noise reduction and combination processing to other areas, improving vividness without excessive complexity.
Data Source
AI summary
An image processing apparatus includes a saturation judgment section adapted to judge whether or not saturation of individual pixels making up image data of plural images falls within a high saturation area, the image data being acquired by picking up images of an object in time sequence; and an image combining section adapted to perform high-saturation comparative bright combination processing on the image data of plural images, where the high-saturation comparative bright combination processing involves replacing only pixels judged to fall within the high saturation area with a pixel value of a pixel which has a larger index including a luminance component.