Image Highlight Recovery Using Border Pixel Brightness Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques struggle to accurately predict brightness values in areas with blown-out highlights, as the actual brightness varies and conventional methods fail to uniformly correct saturated pixel values, leading to incorrect image representation.
Innovation Solution
An image processing device extracts unsaturated and saturated areas from image data and predicts pixel values in saturated areas using border pixels from adjacent unsaturated areas, employing prediction processes based on ratios of different color components and changes in base color components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If uniform correction is applied to saturated areas with pixel values at the upper limit, then the processing is simple, but the actual brightness variation in the photographing object is lost and incorrect image representation occurs
Solution Approach 1:
The patent applies different correction strategies to different regions within saturated areas based on local characteristics. By dividing saturated areas into multiple regions and applying region-specific correction methods, the system preserves local brightness variations while handling saturation, thereby achieving both processing feasibility and brightness accuracy.
Solution Approach 2:
The patent performs preliminary analysis of saturated areas to identify regions with actual brightness variation before applying correction. By pre-processing the image data to detect and classify saturated regions, the system can then apply appropriate correction methods that preserve brightness information, avoiding uniform correction that would lose detail.
2Device complexity
If only one image data from one imaging device is used, then the device complexity is low, but it is difficult to predict the photographing object's image in areas with blown-out highlights
Solution Approach 1:
The patent introduces intermediate processing steps that analyze the relationship between different pixel values and brightness characteristics within the single image data. By using intermediate calculations and prediction algorithms, the system extracts brightness information from saturated areas without requiring additional imaging devices, thereby maintaining device simplicity while improving prediction accuracy.
3Ease of operation
If pixel values in saturated areas are uniformly set to the upper limit, then the imaging device operation is simple, but the image quality deteriorates with blown-out highlights
Solution Approach 1:
The patent dynamically adjusts pixel values in saturated areas based on local brightness variation characteristics rather than uniformly setting them to the upper limit. By changing the parameter assignment strategy from fixed uniform values to variable values derived from local analysis, the system maintains operational simplicity while significantly improving image quality and preventing blown-out highlights.
Data Source
AI summary
An image processing device includes an extracting part and a prediction part. The extracting part extracts an unsaturated area, in which pixel values are not saturated, and a saturated area, in which pixel values are saturated, from image data that is captured by an imaging device and that shows an image including a plurality of pixels. The prediction part predicts a pixel value of a pixel of interest in the saturated area based on pixel values of a plurality of border pixels in a border area, the border area being located in the unsaturated area and bordering the saturated area.