IR Depth Map False Alert Exclusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video camera systems face challenges in simplifying usage and optimizing the utilization of illumination emitters for accurate scene analysis and camera self-diagnosis, particularly in low-light conditions, leading to inefficiencies in motion detection and zone monitoring.
Innovation Solution
The implementation generates lookup tables for estimating spatial depth using IR illuminators and image sensors, allowing for the creation of depth maps and object classification, which enables more precise motion detection and automatic zone adjustments based on camera movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If motion detection is performed in all regions of the scene, then detection coverage is improved, but false positive alerts increase due to non-critical areas like ceilings and floors
Solution Approach 1:
The scene is divided into multiple depth layers using depth map data. Regions at extreme depths (ceilings, floors) are segmented and excluded from motion detection zones, while only regions within the valid monitoring depth range are subjected to motion analysis. This segmentation eliminates false positives from non-critical areas while preserving detection accuracy in monitored zones.
2Illumination intensity
If IR illuminators are used to illuminate the scene in low-light conditions, then visibility is improved, but depth estimation accuracy deteriorates due to incorrect depth mapping
Solution Approach 1:
Depth lookup tables are pre-computed and stored for various IR illuminator configurations and scene depths. Before performing depth estimation during low-light conditions, the system retrieves the appropriate pre-computed lookup table based on current illuminator states, ensuring accurate depth mapping without requiring real-time computation. This preliminary preparation eliminates depth estimation errors that would otherwise occur during IR illumination.
3Speed
If lookup tables are pre-computed for all possible scene configurations, then depth estimation speed is improved, but memory requirements increase
Solution Approach 1:
Instead of computing comprehensive lookup tables for all possible scene configurations, the system computes and stores only the portions of lookup tables corresponding to actual monitored regions and typical scene depths. The depth estimation process then uses only these locally-relevant table portions, reducing memory requirements while maintaining fast estimation speeds for the actual monitoring scenarios.
4Measurement precision
If zones are manually configured for motion detection, then detection precision in specific areas is improved, but system complexity increases due to camera movement compensation
Solution Approach 1:
The system automatically adjusts monitoring zones based on detected camera movement. When camera position or orientation changes are detected through depth map analysis, the system self-corrects zone coordinates and boundaries without requiring manual reconfiguration. This self-service capability maintains detection precision while eliminating the complexity of manual zone management during camera movement.
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 enhances the accuracy and efficiency of motion detection, reduces false positives, and improves user satisfaction by providing detailed scene analysis and camera self-diagnosis capabilities.
Implementation Method 1
surveillance cameras include infrared emitters in order to illuminate a scene when light from other sources is limited or absent
Implementation Method 2
a 2-dimensional array of image sensors (e.g., photodiodes) and a plurality of IR illuminators
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A process reduces false positive security alerts. The process is performed at a computing device having one or more processors, and memory storing one or more programs configured for execution by the one or more processors. The process computes a depth map for a scene monitored by a video camera using a plurality of IR images captured by the video camera and uses the depth map to identify a first region within the scene having historically above average false positive detected motion events. In some instances, the first region is a ceiling, a window, or a television. The process monitors a video stream provided by the video camera to identify motion events, excluding the first region, and generates a motion alert when there is detected motion in the scene outside of the first region and the detected motion satisfies threshold criteria.