Non-Intrusive Blockiness Measurement in Video Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring blockiness in digitally compressed images and videos are intrusive, requiring comparison to a reference signal, which is not feasible for in-service analysis, and existing non-intrusive metrics fail to accurately assess blockiness without comparing pixel values across blocks.
Innovation Solution
A non-intrusive method that calculates an intra-block measure by determining row and column standard deviations within each block, setting a penalty based on threshold values, and summing these measures to quantify blockiness across the entire image or video signal, allowing for in-service analysis without a reference signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If intrusive metrics requiring comparison to reference signal are used, then measurement precision is improved, but device complexity and ease of operation worsen due to requiring reference signal access
Solution Approach 1:
The patent segments the image into blocks and further divides each block into sub-blocks for analysis. By calculating standard deviations within sub-blocks and comparing them across block boundaries, the method achieves accurate blockiness measurement without requiring reference signals, thus resolving the contradiction between measurement precision and ease of operation.
Solution Approach 2:
The patent employs self-service by using only the decoded image signal itself for blockiness measurement. The method calculates intra-block variations and inter-block differences directly from the available signal, eliminating the need for external reference signals and making the system easy to operate in service conditions while maintaining measurement accuracy.
2Ease of operation
If non-intrusive metrics without reference signal are used, then ease of operation is improved for in-service analysis, but measurement precision worsens due to inability to compare with original signal
Solution Approach 1:
The patent transitions from comparing pixel values across different blocks to analyzing the statistical distribution of pixel values within blocks. By calculating standard deviations and examining the variance of these statistics, the method creates a new dimensional approach that achieves accurate blockiness measurement without reference signals, resolving the precision-ease contradiction.
Solution Approach 2:
The patent changes the measurement parameter from direct pixel value comparison to statistical parameters (standard deviation, variance) of pixel distributions. This parameter transformation enables accurate blockiness assessment using only the decoded signal, maintaining measurement precision while improving ease of operation for in-service analysis.
3Device complexity
If pixel values from one block are compared with neighbouring blocks, then blockiness detection is simplified, but measurement precision worsens due to inter-block comparison limitations
Solution Approach 1:
The patent segments each block into multiple sub-blocks and performs analysis at the sub-block level. By calculating standard deviations within sub-blocks and examining the distribution of these statistics across the entire block, the method achieves more precise blockiness measurement while maintaining relative simplicity, overcoming the limitations of direct inter-block pixel comparison.
Solution Approach 2:
The patent moves from direct spatial comparison of pixel values between blocks to a statistical dimension by calculating and comparing standard deviations. This dimensional transformation provides more accurate blockiness measurement while keeping the detection method relatively simple, resolving the contradiction between simplicity and precision.
Data Source
AI summary
The present invention relates to an apparatus and method for processing images, in particular for determining the degree of blockiness in coded images. The invention comprises apparatus and methods for determining a degree of blockiness in a digital image comprising blocks each block comprising a plurality of rows and a plurality of columns of pixels each pixel having a value by performing the steps of: for each block determining an intra block measure in dependence upon the values of a set of pixels within said block; and determining said measure in dependence upon a plurality of said intra block measures corresponding to each block. Method and apparatus for generating a quality measure in dependence upon the measure of blockiness are also provided.