Behavior Recognition Device Using Room Layout and Sensor Position
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing behavior recognition techniques fail to accurately identify behaviors dependent on specific spaces within a building, as they do not consider room layout information, leading to suboptimal recognition accuracy.
Innovation Solution
A behavior recognition device that acquires target behavior information, room layout information, and image sensor position data to select candidate behaviors and use corresponding recognizers to calculate feature values from image data, enhancing recognition accuracy by integrating space-specific information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing behavior recognition techniques are used without room layout information, then the system complexity is low, but the behavior recognition accuracy deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-acquiring and storing room layout information and image sensor position information before behavior recognition. The behavior selection unit uses these pre-acquired spatial data to determine candidate behaviors, avoiding the need to process complex spatial relationships in real-time during recognition, thus improving accuracy without proportionally increasing system complexity
Solution Approach 2:
The patent introduces room layout information and image sensor position information as intermediary data that mediates between the image data and behavior recognition. These intermediary spatial data enable the behavior selection unit to accurately determine candidate behaviors by providing contextual spatial information, thereby improving recognition accuracy while keeping the core recognition mechanism relatively simple
2Measurement precision
If multiple recognizers are used to calculate feature values for different candidate behaviors, then the behavior recognition accuracy improves, but the processing time increases
Solution Approach 1:
The patent applies segmentation by dividing the behavior recognition process into distinct stages: behavior selection based on spatial information, candidate behavior determination, and feature value calculation using multiple recognizers. This segmentation allows the system to use multiple recognizers only for selected candidate behaviors rather than all possible behaviors, improving accuracy while controlling processing time through staged processing
Solution Approach 2:
The patent applies partial action by using multiple recognizers selectively rather than universally. The behavior selection unit determines candidate behaviors based on room layout and sensor position, then only applies multiple recognizers to calculate feature values for these specific candidates. This partial application of multiple recognizers provides sufficient accuracy improvement without the excessive processing time that would result from applying all recognizers to all possible behaviors
Data Source
AI summary
Provided is a behavior recognition device that selects a candidate behavior as a recognition candidate from among target behaviors based on position information on an image sensor and room layout information, acquires image data detected by the image sensor, determines one or more recognizers corresponding to the candidate behavior, calculates a feature value of the image data using the one or more recognizers, and recognizes the candidate behavior based on the feature value.