Sensor Fusion for Biometric Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current monitoring systems using passive infrared detectors and optical cameras often experience false alarms due to lighting changes and ambient temperature issues, which can lead to missed detections of objects such as animals or people.
Innovation Solution
A system comprising sensors, processors, and memory that detect objects, classify them as authorized or foreign based on biometric profiles, and trigger alarms or enable services accordingly, using a combination of image and infrared data to generate biometric signatures for identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical cameras are used to detect objects, then visual identification capability is improved, but false detections increase due to lighting changes
Solution Approach 1:
The patent merges optical camera detection with passive infrared detector data. When lighting changes cause false detections in the camera, the infrared sensor provides complementary thermal information that helps distinguish genuine objects from lighting-induced artifacts, maintaining visual identification capability while reducing false detections.
Solution Approach 2:
The passive infrared detector acts as an intermediary validation mechanism for the optical camera. By introducing thermal signature data as an intermediary check, the system can verify whether camera-detected objects are genuine, filtering out false detections caused by lighting changes before final identification.
2Measurement precision
If multiple sensor types are combined for monitoring, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent designs the monitoring system with multi-functional processing capability that handles both infrared and optical data through a unified correlation analysis framework. This universal processing approach improves detection accuracy by leveraging multiple sensor types while managing system complexity through integrated rather than separate processing paths.
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
This approach reduces false alarms and improves the accuracy of object detection and identification, enabling effective monitoring and tracking of authorized and unauthorized objects within monitored areas.
Implementation Method 1
Infrared sensors detect heat signatures
Implementation Method 2
Cameras can also be used to trigger an alarm if enough pixels change to identify a foreign object
Data Source
AI summary
A system that incorporates teachings of the subject disclosure may include, for example, a method for detecting, by a system including at least one processor, a presence of an object from sensor data generated by a sensor device, retrieving, by the system, from a memory device a plurality of profiles biometrically descriptive of approved objects, asserting, by the system, an alarm responsive to determining from the sensor data that the detected object is not biometrically correlated to any of the plurality of profiles, classifying, by the system, the detected object as an authorized object responsive to determining from the sensor data that the detected object is biometrically correlated to at least one of the plurality of profiles, and notifying, by the system, at least one neighboring device responsive to asserting the alarm or responsive to classifying the detected object as the authorized object. Other embodiments are disclosed.