Motion Trigger Level Setting for Video Stream Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video monitoring systems face challenges in reliably detecting motion without triggering false alarms, particularly due to irrelevant changes in scenes like swaying trees or waves on puddles, leading to increased costs from unnecessary responses.
Innovation Solution
A method that divides a video stream into specific portions, evaluates statistical features of bitrate associated with encoding block pixels over time to determine a motion base level, and sets a motion trigger level for each portion, allowing for customized detection without requiring calibration in event-free environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a simple algorithm compares current image with reference image to detect motion, then motion detection is easy to implement, but false alarms increase due to irrelevant motion changes
Solution Approach 1:
The patent changes the parameter from simple pixel intensity comparison to statistical evaluation of bitrate changes over time. By analyzing the statistical features of bitrate variations and comparing them against learned thresholds, the system maintains ease of implementation while significantly reducing false alarms caused by irrelevant motion changes.
Solution Approach 2:
The system performs self-calibration by automatically learning the statistical features of bitrate changes during a calibration phase without requiring manual intervention or event-free environments. This self-service approach enables the system to adapt to specific scene characteristics and establish appropriate thresholds autonomously.
2Reliability
If filters such as small object filtering and swaying object filtering are implemented to reduce false alarms, then false alarm rate decreases, but device complexity increases
Solution Approach 1:
The patent extracts the essential characteristic of motion detection from complex filter implementations and focuses solely on analyzing bitrate changes. By taking out the core function of detecting motion through bitrate statistical evaluation, the system achieves reliable false alarm reduction without implementing complex filtering mechanisms.
Solution Approach 2:
The patent replaces complex mechanical filtering systems with a statistical analysis approach. Instead of using multiple filtering algorithms, the system substitutes them with statistical feature evaluation of bitrate data, achieving the same goal of reducing false alarms with simpler, more maintainable logic.
3Device complexity
If motion trigger level is set uniformly for the entire scene, then processing is simplified, but detection accuracy decreases due to different motion characteristics in different portions
Solution Approach 1:
The patent segments the scene into multiple portions, each with its own motion trigger level. This segmentation allows the system to analyze different regions with different motion characteristics independently, improving detection accuracy while maintaining manageable processing complexity through structured division of the scene.
Solution Approach 2:
The patent applies local quality by assigning different motion trigger levels to different portions of the scene based on their specific motion characteristics. Each portion receives customized threshold values that match its local motion patterns, enabling precise detection without requiring uniform processing across the entire scene.
Data Source
AI summary
A method for setting a motion trigger level is used in detection of motion in a video stream depicting a scene which includes receiving data pertaining to a video stream depicting the scene; dividing the scene into a plurality of specific portions; wherein each image frame of the video stream comprises multiple blocks of pixels, wherein each specific portion of the scene is associated with one or more block of pixels; and for each specific portion of the scene: evaluating, over time, statistical features of bitrate associated with an encoding of block of pixels pertaining to the specific portion of the scene; determining a motion base level based on the evaluated statistical features of bitrate associated with the encoding of block of pixels pertaining to the specific portion of the scene; and setting a motion trigger level based on the motion base level.