Super-Resolution Image Processing via Block Matching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current super-resolution technologies face high calculation complexity and memory resource occupancy due to processing initial high-resolution images, which are derived from low-resolution images, leading to inefficient image processing.

Innovation Solution

The method involves performing image magnification on a low-resolution image to obtain an initial high-resolution image, selecting a central similar block from the initial high-resolution image, and a search block from the low-resolution image of the same size, calculating similarity using Euclidean distance, and applying weighted averaging to determine pixel values, thereby reducing calculation complexity and memory usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If image processing is performed on an initial high-resolution image obtained from low-resolution image, then image quality is improved, but calculation complexity increases

Engineering Contradiction:
Improveimage qualityVSAvoidcalculation complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent divides the image into multiple blocks and processes each block separately using block-matching algorithms. This segmentation approach reduces the overall calculation complexity by breaking down the large-scale image processing into smaller, manageable units that can be handled more efficiently.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by first identifying similar blocks and calculating their similarity metrics before performing the actual image processing. This preliminary block-matching and similarity calculation prepares the data in advance, reducing the computational burden during the main processing stage.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If image processing is performed on an initial high-resolution image, then image quality is improved, but memory resource occupancy increases

Engineering Contradiction:
Improveimage qualityVSAvoidmemory resource
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

By dividing the high-resolution image into smaller blocks, the patent reduces the memory required to hold and process the entire image at once. Each block can be processed independently with minimal memory overhead, significantly reducing total memory resource occupancy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the necessary block information (such as similarity metrics and displacement vectors) from the low-resolution image rather than processing the entire high-resolution image data. This extraction approach minimizes memory usage by keeping only the essential processing data in memory.

Inventive Principle:
Principle #2Taking out (Extraction)

3Productivity

If block size is increased to improve processing efficiency, then processing speed is improved, but processing precision decreases

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidprocessing precision
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent dynamically adjusts the block size based on the specific processing requirements and image characteristics. By making the block size adaptable rather than fixed, the system can optimize for both processing efficiency and precision in different scenarios, resolving the trade-off between speed and accuracy.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent applies different processing approaches to different regions of the image by using local block-matching. Each block can be processed with appropriate parameters tailored to its specific characteristics, maintaining high precision locally while achieving overall processing efficiency through the distributed block-based approach.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP2887306B1Image processing method and apparatus
Publication Date: 2018.01.31 HUAWEI TECH CO LTD
  • EP2887306B1 patent drawingFigure 1
  • EP2887306B1 patent drawingFigure 2
  • EP2887306B1 patent drawingFigure 3~4

AI summary

Embodiments of the present invention provide an image processing method and apparatus, where the image processing method includes: performing image magnification processing on a low-resolution image to obtain an initial high-resolution image; selecting a central similar block corresponding to a first pixel from the initial high-resolution image; selecting a search block corresponding to the first pixel from the low-resolution image, and determining each similar block in the search block, where the central similar block and the search block have a same block size; and determining a pixel value of the first pixel according to the central similar block and each similar block in the search block. By using technical solutions in the embodiments of the present invention, calculation complexity and an occupied memory resource can be effectively reduced.