Dynamic Background Model for Moving Object Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing intelligent video surveillance systems face challenges in accurately detecting moving objects, particularly in dynamic backgrounds with rapid changes, as they often fail to distinguish between moving backgrounds and foregrounds, leading to increased human labor and a need for real-time monitoring.

Innovation Solution

A system and method that utilize pixel information and time information to model backgrounds, allowing for dynamic background handling by setting and updating background models based on usage time and visual properties, including the deletion of unused model elements and extension of peripheral areas, to differentiate between background and foreground pixels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If parametric background modeling is used to detect moving objects, then the detection process is simplified, but moving backgrounds (e.g., swaying trees or waves) cannot be appropriately removed

Engineering Contradiction:

Improvebackground modeling complexityVSAvoidmoving object detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent implements dynamic background modeling by periodically updating the background model based on detected foreground pixels. The system transitions from a static background model to a dynamic one that adapts to changing environments, allowing appropriate handling of moving backgrounds while maintaining detection accuracy.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary background modeling before actual moving object detection. By establishing an initial background model and then iteratively refining it, the system prepares the detection mechanism in advance, allowing it to distinguish between legitimate moving objects and dynamic background elements.

Inventive Principle:

Principle #10Preliminary action

2Device complexity

If a static background model is used, then the model structure is simple, but it cannot handle dynamic backgrounds or rapid changes in backgrounds

Engineering Contradiction:

Improvebackground model structureVSAvoidbackground adaptability

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent transforms the static background model into a dynamic one by implementing periodic updates. The background model automatically adapts to environmental changes by incorporating new information from detected foreground pixels, enabling it to handle dynamic backgrounds while maintaining a relatively simple structural framework.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The background modeling system performs self-updates by automatically detecting foreground pixels and using them to refine the background model. This self-service mechanism allows the system to adapt to changing backgrounds without requiring external intervention or complex manual reconfiguration.

Inventive Principle:

Principle #25Self-service

3Ease of manufacture

If manual monitoring is used in CCTV security systems, then system implementation is straightforward, but enormous human labor is required

Engineering Contradiction:

Improvesystem implementation easeVSAvoidmonitoring efficiency

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The patent replaces manual mechanical monitoring with an automated intelligent video surveillance system. The system uses background modeling and pixel comparison techniques to automatically detect moving objects, substituting human labor with computational processes while maintaining straightforward system implementation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The surveillance system performs self-monitoring by automatically analyzing video frames, comparing them against the background model, and identifying moving objects. This self-service capability eliminates the need for continuous human monitoring while improving productivity and reducing labor requirements.

Inventive Principle:

Principle #25Self-service

4Measurement precision

If Gaussian mixture model with multiple Gaussian distributions is used, then moving backgrounds can be modeled, but the model complexity increases significantly

Engineering Contradiction:

Improvebackground modeling accuracyVSAvoidmodeling complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the background modeling process into distinct components: initial background model creation, foreground detection, and periodic updates. By dividing the complex modeling task into manageable segments, the system achieves accurate modeling of moving backgrounds while keeping the overall complexity controllable through modular processing.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS10373015B2System and method of detecting moving objects

Publication Date: 2019.08.06 HANWHA AEROSPACE CO LTD

AI summary

Provided are a system and method of detecting moving objects. The system stores pixel information regarding each of pixels included in frames of the video in a storage, sets a background model comprising at least one background model element, the at least one background model element indicating at least one of a reference visual property and a reference usage time, determines whether the pixels are background pixels or foreground pixels by comparing the pixel information with the at least one background model element, and updating the background model based on a result of the comparing.