Background Model Regeneration for Moving Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing moving object detection technologies face inefficiencies in determining whether to regenerate or reuse background models, especially in scenarios with periodic changes in brightness, leading to unnecessary inference processing and increased calculation costs.
Innovation Solution
A moving object detection apparatus and method that generates a background model based on image features, detects moving objects, and determines whether to regenerate the model by calculating the difference in moving object region areas between images captured at different times, allowing for efficient decision-making on model reuse or regeneration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If background models are continually inferred and regenerated using luminance values before and after brightness changes, then the background model can adapt to environment changes, but calculation costs increase and processing becomes inefficient in cases of periodic brightness changes
Solution Approach 1:
The patent applies periodic action by storing multiple background models corresponding to different brightness periods and selectively reusing them based on current brightness conditions. Instead of continuously regenerating background models, the system periodically stores models during brightness changes and then reuses appropriate stored models, thereby reducing calculation costs while maintaining adaptability to periodic brightness variations.
Solution Approach 2:
The patent implements preliminary action by pre-storing background models during brightness change periods before they are needed. When brightness changes occur, the system has already prepared corresponding background models in advance, eliminating the need for real-time inference and regeneration during operation, thus reducing computational load while ensuring model availability.
2Reliability
If background models are regenerated frequently to adapt to environment changes, then detection reliability improves, but processing time increases
Solution Approach 1:
The system uses periodic action by establishing a determination period after brightness changes during which background model regeneration is suppressed. Instead of immediately regenerating models upon every brightness change, the system waits for a predetermined period to elapse, allowing time for stability assessment before deciding whether regeneration is necessary, thus reducing unnecessary processing time.
Solution Approach 2:
The patent implements feedback by calculating the amount of change in detected moving object regions and using this information to determine whether background model regeneration is truly necessary. The system monitors detection results and only regenerates models when actual changes warrant it, rather than regenerating based solely on brightness changes, thereby improving detection reliability while minimizing unnecessary processing time.
3Loss of energy
If background models are reused based on brightness information, then calculation costs are reduced, but detection precision may deteriorate when models are inappropriately reused
Solution Approach 1:
The patent uses feedback by calculating the amount of change in moving object regions and comparing it against thresholds to determine whether stored background models should be reused or regenerated. This feedback mechanism ensures that models are only reused when detection precision requirements are met, preventing inappropriate reuse that would compromise detection accuracy while still reducing calculation costs through selective reuse.
Solution Approach 2:
The system applies dynamics by making the background model selection adaptive rather than static. Instead of always regenerating or always reusing models based on fixed rules, the system dynamically adjusts its behavior based on real-time assessment of moving object region changes, allowing it to optimize between calculation cost and detection precision according to current conditions.
Data Source
AI summary
A moving object detection apparatus has a generation unit that generates a background model based on a feature of a background region of a captured image that is captured by image capturing unit, a detection unit that detects a moving object region from an image input using an input unit, based on the background model, and a determination unit that determines whether to cause the generation unit to newly generate a background model, based on an amount of change in the moving object region detected by the detection unit for a first image and a second image captured at different times and input using the input unit.