Image Processing Device for Unknown Degradation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies require a sufficient number of input images to accurately learn the degradation process, making it impossible to effectively improve the image quality of a single input image with an unknown degradation process.
Innovation Solution
An image processing device that performs degradation processing on a second image using different degradation processes to generate multiple degraded images, compares the input image with these degraded images, and selects a parameter to improve image quality based on the comparison results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a converter learns the degradation process using a sufficient number of input images, then the accuracy of learning the degradation characteristics is improved, but it becomes impossible to effectively process a single input image with unknown degradation process
Solution Approach 1:
The system pre-generates multiple degraded images from a reference image using different degradation processes (compression, blur, noise, etc.) before comparing with the input image. This preliminary creation of degraded samples enables the system to handle single images without requiring extensive training data, as the degradation patterns are pre-established for comparison.
Solution Approach 2:
The system creates multiple copies of a reference image, each subjected to different degradation processes. These copied and degraded images serve as templates for comparison with the input image, allowing the system to identify the degradation type without needing the original training images.
2Adaptability or versatility
If multiple degraded images are generated using different degradation processes, then the ability to handle unknown degradation processes is improved, but the computational complexity increases
Solution Approach 1:
The degradation processes are segmented into distinct types (compression degradation, blur degradation, noise degradation, etc.), each handled by separate conversion units. This segmentation allows the system to manage complexity by dividing the problem into manageable, specialized components rather than treating all degradations uniformly.
Solution Approach 2:
A comparison unit acts as an intermediary between the input image and multiple degraded images, calculating similarity metrics to determine the best match. This intermediary component simplifies the overall system by providing a standardized method for comparing the input image against various degradation patterns without requiring complex analysis of each degradation type.
3Measurement precision
If the input image is compared with multiple degraded images, then the accuracy of selecting the appropriate parameter is improved, but the processing time increases
Solution Approach 1:
The system generates and compares with multiple degraded images (excessive action) to ensure accurate parameter selection, but uses a comparison unit that efficiently calculates similarity metrics to minimize the time penalty. The excessive generation of degraded samples is justified by the significant improvement in parameter selection accuracy.
Solution Approach 2:
The system changes parameters such as similarity calculation methods and degradation levels in the generated images to optimize the balance between accuracy and processing time. By adjusting these parameters, the system can achieve high accuracy in parameter selection while controlling the computational burden.
Data Source
AI summary
The present technology relates to an image processing device, an image processing method, and a recording medium capable of improving an image quality of an input image of which the degradation process is unknown. An image processing device of the present technology includes a degradation conversion unit that performs degradation processing including mutually different degradation processes on a second image different from an input first image to generate a plurality of degraded images; a comparison unit that compares the first image with each of the plurality of degraded images; and a selection unit that selects, based on a result of comparison by the comparison unit, a parameter to improve the image quality of the first image from among parameters associated with the degradation processes of the plurality of degraded images. The present technology can be applied to, for example, an image processing device that performs super-resolution processing.