High Resolution Image Obtaining Apparatus Region-Specific Enhancement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for obtaining high resolution images from low resolution images, such as sharpening, interpolation, reconstruction-based, and example-based algorithms, face limitations in handling complex motion and noise, particularly in digital image recording and processing where optical limits and pixel insufficiencies are prevalent.
Innovation Solution
A high resolution image obtaining apparatus and method that divides an input image into background and foreground regions, applying region-specific resolution enhancement algorithms, combining reconstruction-based and example-based methods using wavelet transforms to optimize image enhancement, with a database storing wavelet transform coefficients for effective high resolution image generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If reconstruction-based resolution enhancement algorithm is used, then high resolution image can be obtained, but it is unsuitable for general image having complex motion of object
Solution Approach 1:
The patent segments the image processing task by dividing the input image into multiple patches and processing each patch independently through the reconstruction-based algorithm. This segmentation allows the algorithm to handle local regions with complex motion more effectively, as each patch can be processed with appropriate motion compensation parameters.
Solution Approach 2:
The patent introduces dynamic motion compensation by estimating motion parameters for each patch and adjusting the reconstruction process accordingly. This dynamic adaptation allows the reconstruction-based algorithm to handle varying motion conditions across different regions of the image, improving its versatility for general images with complex motion.
2Measurement precision
If example-based resolution enhancement algorithm is used, then high resolution image can be obtained from single image frame, but it is weak against noise and undesired artifacts are generated
Solution Approach 1:
The patent merges the example-based algorithm with the reconstruction-based algorithm, combining the advantages of both approaches. The example-based method provides high resolution enhancement from single frames, while the reconstruction-based method with motion compensation reduces noise and artifacts, creating a more reliable hybrid system.
Solution Approach 2:
The patent incorporates feedback mechanisms where the processing results from one algorithm inform and adjust the parameters of the other algorithm. This feedback loop allows the system to adapt to noise conditions and artifact generation, improving overall reliability by using the strengths of each algorithm to compensate for the weaknesses of the other.
3Illumination intensity
If sharpening method is used, then visibility of image is improved, but resolution enhancement is limited
Solution Approach 1:
The patent applies sharpening as a preliminary action before the main resolution enhancement process. By pre-enhancing the visibility and edge definition through sharpening filters, the subsequent reconstruction and example-based algorithms can work more effectively on the already-enhanced image, achieving better overall resolution improvement.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is a high resolution image obtaining apparatus and method. The high resolution image obtaining apparatus may divide an input image frame into a background region and foreground region and apply an optimized resolution enhancement algorithm to each region, thereby effectively obtaining a high resolution image frame with respect to the input image frame.