Video Motion Direction Estimation Using Edge Differential Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating the direction of movement in video images require processing multiple images, leading to high memory usage and latency, which is undesirable for applications requiring rapid estimation with minimal memory cost.
Innovation Solution
A method and system that determine a differential image from two video frames, select pixels corresponding to scene edges, classify local movements, and combine indicators to estimate a general movement direction using a processing module.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple images are processed to estimate movement direction, then measurement precision is improved, but memory requirements and latency increase
Solution Approach 1:
The patent extracts only the essential information needed for movement direction estimation from the video images, specifically using differential images that highlight changes between frames. By selecting only the differential information rather than storing complete image sequences, the system achieves accurate movement estimation with minimal memory requirements.
Solution Approach 2:
The patent segments the image processing into discrete steps: creating differential images, selecting edge pixels, determining local movement classifiers, and combining indicators. This segmentation allows the system to process information in manageable chunks without requiring storage of the entire image sequence, resolving the contradiction between precision and memory usage.
2Measurement precision
If multiple images are processed to estimate movement direction, then measurement precision is improved, but processing time and latency increase
Solution Approach 1:
The patent performs preliminary processing by pre-calculating differential images and preparing edge detection parameters before actual movement estimation is needed. This preliminary action reduces the computational burden during real-time estimation, allowing accurate movement direction determination without significant latency.
Solution Approach 2:
The patent uses partial action by processing only the necessary portions of images (edge pixels and differential regions) rather than analyzing every pixel in every frame. This selective processing maintains measurement precision while significantly reducing processing time and latency.
3Quantity of substance
If a simple differential image approach is used, then memory requirements are reduced, but measurement precision may deteriorate
Solution Approach 1:
The patent applies local quality by analyzing different regions of the differential image with different levels of detail. Edge pixels are selected based on their local characteristics, and movement classifiers are determined independently for each pixel region. This allows the system to use minimal memory while maintaining precision by focusing computational effort only where needed.
Solution Approach 2:
The patent introduces intermediary structures (differential images and movement classifiers) that mediate between the raw image data and the final movement direction estimation. These intermediaries compress the information while preserving essential characteristics, enabling accurate estimation without storing complete image sequences.
Data Source
AI summary
The present description concerns a system and method of determining at least one classifier of a general movement along a first direction in video images of a scene, comprising determining a differential image based on two video images, selecting pixels of the differential image corresponding to edges of objects, determining, for each selected pixel, at least one classifier of a local movement along the first direction at least at a first or second value, determining a first indicator of the local movement along the first direction, which depends on the sum of the local movement classifiers at the first value and a second indicator of the local movement along the first direction, which depends on the sum of the local movement classifiers at the second value, and determining the general movement classifier based on the comparison of the first and second local movement indicators.