Block-Based Video Frame Sequence Detection System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining if a sequence of video frames is static are not robust against noise and are computationally intensive, often failing to detect small movements or being overly sensitive to noise, leading to false positives or inefficiencies.
Innovation Solution
The method partitions images into blocks and calculates a difference measure between corresponding blocks, comparing it to a threshold to assess similarity, which reduces computational burden and improves sensitivity to small movements by focusing on key areas and adjusting the threshold based on image properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the threshold is made large to exclude noise effects, then reliability is improved, but measurement precision deteriorates as small movements cannot be detected
Solution Approach 1:
The video frame is divided into multiple blocks, and the difference measure is calculated for each block separately. This segmentation allows the system to identify small movements in specific blocks while using a threshold that is robust against noise for the overall assessment.
Solution Approach 2:
The patent applies different treatment to different parts of the frame by evaluating blocks locally. The difference measure is calculated for each block, allowing local variations to be detected while maintaining overall reliability through the block-based approach.
2Measurement precision
If pixel-by-pixel comparison is performed for all pixels, then measurement precision is improved, but productivity deteriorates due to computational intensity
Solution Approach 1:
The frame is segmented into blocks, and the difference measure is calculated for each block rather than every pixel. This reduces the computational load from O(N*M) for all pixels to O((N*M)/K) for blocks, where K is the number of blocks per frame.
Solution Approach 2:
The patent calculates the difference measure for blocks rather than all individual pixels. This partial action approach provides sufficient detection accuracy by evaluating representative blocks while significantly reducing computational requirements.
3Productivity
If block-based comparison is used, then productivity is improved, but measurement precision may deteriorate if blocks are too large to detect small movements
Solution Approach 1:
The frame is divided into multiple small blocks, ensuring that even small movements within a block will result in a detectable difference measure. The block size is optimized to balance computational efficiency with the ability to detect fine-grained changes.
Solution Approach 2:
The patent adjusts the block size parameter to optimize the trade-off between processing efficiency and detection sensitivity. By carefully selecting the block dimensions, the system achieves both computational efficiency and adequate sensitivity to small movements.
Data Source
AI summary
An apparatus and method of evaluating whether an image under test is similar to a reference image, the method comprising performing a test on pairs of blocks in turn, each pair of blocks comprising a block of the image under test and a corresponding block of the reference image, the test comprising calculating a difference measure indicative of the difference between the blocks and comparing the difference measure to a threshold, wherein the test is passed if the difference measure betters the threshold, and the image under test and the reference image are deemed to be similar if the test is passed by all of the pairs of blocks.