Multi-Camera Behavior Analysis Using Attribute-Specific Movement Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing behavior analysis technologies using surveillance cameras struggle with accuracy, particularly in distinguishing normal from suspicious behavior, as they rely on a common average value for suspicion degree calculations, which fails to account for individual attributes like age, leading to inappropriate evaluations.
Innovation Solution
An analyzing apparatus and method that detects individuals across multiple camera feeds, computes movement time between camera areas, determines attributes, and compares this time with attribute-specific reference values to generate output information, allowing for more accurate behavior analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a common average value is used for suspicion degree calculations, then the computation is simple, but the behavior analysis accuracy deteriorates because individual differences (age, mobility) are not considered
Solution Approach 1:
The patent applies local quality by transitioning from a uniform evaluation approach to an individualized one. Specifically, it determines attributes (such as age group) for each person and uses attribute-specific reference values instead of a common average value. This allows the suspicion degree calculation to be adapted to individual characteristics, thereby improving behavior analysis accuracy while maintaining computational feasibility through structured attribute-based classification.
2Measurement precision
If attribute-specific reference values are used, then behavior analysis accuracy is improved, but the system complexity increases due to additional attribute determination and reference value management
Solution Approach 1:
The patent applies segmentation by dividing the population into attribute-based groups (e.g., age groups) and assigning different reference values to each group. This segmentation approach manages system complexity by organizing reference values in a structured manner according to attributes, making the increased complexity tractable and systematic rather than chaotic or unmanageable.
Solution Approach 2:
The patent applies parameter changes by introducing attribute parameters (such as age group) that modify the reference values used in suspicion degree calculations. Instead of using a fixed common average value, the system dynamically selects reference values based on the determined attributes of each person, thereby improving accuracy while managing complexity through parameter-based adaptation.
3Reliability
If individual attributes are considered in movement time evaluation, then false positives and negatives are reduced, but the computational requirements and data processing complexity increase
Solution Approach 1:
The patent applies preliminary action by determining attributes and selecting appropriate reference values before performing the suspicion degree calculation. This preparatory step ensures that the correct attribute-specific reference values are ready when needed, improving evaluation reliability while managing data processing complexity through advance preparation and structured data organization.
Data Source
AI summary
A first camera and a second camera are installed in a facility. An analyzing apparatus detects the same person from a first captured image generated by the first camera and a second captured image generated by the second camera. The analyzing apparatus computes a movement time which is required for the detected person to move between a captured area of the first camera and a captured area of the second camera. The analyzing apparatus determines an attributed of the person using at least one of the first captured image and the second captured image. The analyzing apparatus acquires a reference value corresponding to the determined attribute, compares the computed movement time with the acquired reference value, and outputs output information based on a result of the comparison.