Video Recognition Control for Zoom Magnification Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Monitoring systems face challenges in detecting moving objects consistently when zoom magnification changes, leading to incorrect detection or failure to detect objects that were previously detectable, due to mismatched maximum and minimum size settings for video image recognition processing.
Innovation Solution
A control apparatus that includes an acquisition unit for image data and a control unit to adjust recognition processing parameters based on changes in zoom magnification, ensuring consistent detection by rescaling frame images and updating maximum and minimum size settings for human body detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If zoom magnification is changed after recognition parameters are set, then the field of view and object size in the image change, but the detection accuracy deteriorates because the maximum and minimum size settings for object detection no longer match the actual object sizes
Solution Approach 1:
The patent applies dynamics by making the recognition parameters (maximum size and minimum size for object detection) variable rather than fixed. When zoom magnification changes are detected, the system automatically updates these parameters to match the new object sizes in the image, ensuring continuous detection accuracy across different zoom levels
Solution Approach 2:
The patent directly implements parameter changes by modifying the maximum and minimum size thresholds for object detection based on the current zoom magnification level. The system calculates appropriate parameter values that correspond to the zoomed-in or zoomed-out state, ensuring objects remain detectable regardless of their apparent size changes
2Productivity
If fixed maximum and minimum size settings are used for object detection, then processing is simpler and faster, but objects may be missed or falsely detected when zoom magnification changes
Solution Approach 1:
The patent applies preliminary action by detecting zoom magnification changes and proactively adjusting recognition parameters before detection errors occur. The system monitors zoom status and pre-adjusts the maximum and minimum size thresholds to prevent missed detections or false positives, maintaining both speed and reliability
Solution Approach 2:
The patent implements feedback by using zoom magnification information to continuously adjust recognition parameters. The system receives feedback about the current zoom state and automatically modifies detection thresholds accordingly, creating a closed-loop system that maintains detection reliability without sacrificing processing efficiency
Data Source
AI summary
A control apparatus and method that captures video image includes acquiring a parameter for recognition processing with respect to acquired image data and changing the acquired parameter according to a change in zoom magnification of an image capturing unit.