Spatial Zone Segmentation for Noise Rejection in Occupancy Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Person detection systems face limitations due to output noise from sensors, making it difficult to distinguish a person from the ambient environment, particularly when the noise threshold is high, leading to either false negatives or false positives.
Innovation Solution
Implementing a processor-based system that divides the observed area into zones and applies a detection threshold value lower than the noise threshold, determining a trigger condition based on sequential ordered events, such as heat signature changes within these zones, to improve noise immunity and stability without relying on filters or oversampling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a high threshold is applied to sensor readings to reduce false positives, then system stability is improved, but detection sensitivity deteriorates (false negatives increase)
Solution Approach 1:
The patent divides the observed area into multiple zones and segments sensor readings into ordered events (e.g., entering a zone, leaving a zone, moving between zones). This segmentation allows the system to distinguish between random noise and coherent human movement patterns by analyzing sequences of events across zones rather than relying on a single threshold comparison.
Solution Approach 2:
The system performs preliminary actions by establishing a sequence of ordered events before triggering a detection. Instead of immediately reacting to a single sensor reading, the system pre-establishes a pattern of events (entering zones in sequence, directional movement) that must occur before confirming human presence, thereby filtering out random noise while maintaining sensitivity.
2Measurement precision
If a low threshold is applied to sensor readings to increase detection sensitivity, then detection capability is improved, but system stability deteriorates (false positives increase)
Solution Approach 1:
The patent segments the detection process into multiple ordered events across different zones. Even with a low threshold that may trigger on random noise, the requirement for a specific sequence of events (entering zone A, then zone B, then zone C in order) makes it extremely unlikely that random noise will satisfy all conditions, thereby maintaining system stability while preserving sensitivity.
Solution Approach 2:
The system requires continuous, sequential confirmation of events across multiple zones before triggering a detection. This continuity requirement ensures that transient noise spikes or isolated false readings cannot trigger alerts, as they would not sustain the required sequence of ordered events across different spatial zones.
3Reliability
If filters are applied to sensor data to reduce noise, then noise immunity is improved, but system response time deteriorates
Solution Approach 1:
Instead of applying temporal filters that delay responses, the patent segments space into zones and uses spatial ordering of events to achieve noise immunity. The system waits for a sequence of events across different spatial locations rather than filtering over time, thereby maintaining fast response while achieving robust noise rejection through spatial pattern recognition.
4Measurement precision
If oversampling is applied to sensor data to improve detection accuracy, then measurement precision is improved, but data processing complexity increases
Solution Approach 1:
The patent segments the observed area into zones and focuses processing on identifying ordered events across these zones rather than processing every individual sensor reading. This spatial segmentation approach reduces processing complexity by concentrating analysis on event sequences rather than continuously processing high-rate sensor data, while maintaining detection accuracy through the ordered event pattern recognition.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables reliable person detection with reduced noise interference, maintaining system stability and accuracy by preferring previous states and using directional preferences, effectively distinguishing human movement from random noise.
Implementation Method 1
in a system that uses a thermal sensor to detect a person's heat signature
Data Source
AI summary
Systems and methods for detecting or following a person within an observed area. One example system includes a sensor and a processor configured to: receive, from the sensor, a series of readings indicative of the person in the observed area, wherein the observed area is divided into a plurality of zones; determine a sequential occurrence of each of a plurality of ordered events by applying a detection threshold value to the series of readings, wherein the detection threshold value is lower than a noise threshold of the sensor, and wherein each of the plurality of ordered events is associated with at least one of the plurality of zones; and determine a trigger condition based on the sequential occurrence of each of the plurality of ordered events; and provide an instruction to a fixture based on the trigger condition.