Augmented Reality Security via Edge Neural Network Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented reality devices lack effective solutions for real-time monitoring and identification of abnormal conditions and behaviors in crowded areas, such as signs of intoxication, sickness, or suspicious activities, which can lead to delayed response times in security operations.
Innovation Solution
An augmented reality vigilance system utilizing smart glasses equipped with an artificial neural network that analyzes video feeds from cameras, detects anomalies, and provides real-time alerts to authorized users through augmented reality displays, enabling immediate attention and action.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If augmented reality devices implement real-time anomaly detection using artificial neural networks, then security monitoring capability is improved, but energy consumption increases
Solution Approach 1:
The system segments the anomaly detection workload by deploying artificial neural networks on distributed edge devices (smartphones, tablets, wearables) rather than concentrating all processing on a single centralized server. This distribution allows real-time local detection while managing energy consumption across multiple devices.
Solution Approach 2:
The patent introduces an intermediary layer of edge devices that process video feeds locally using artificial neural networks before transmitting results to centralized servers. This intermediary approach enables real-time detection at the edge while reducing the energy burden on any single device and optimizing overall system energy efficiency.
2Speed
If augmented reality devices process video feeds in real-time, then detection speed is improved, but computational burden increases
Solution Approach 1:
The system segments video processing across multiple edge devices, each running lightweight artificial neural networks that process specific portions of video feeds. This segmentation enables parallel real-time processing while distributing computational burden across the edge device ecosystem rather than overwhelming a single device.
Solution Approach 2:
The patent implements partial processing by having edge devices perform only the critical real-time anomaly detection functions using optimized artificial neural networks, while less time-sensitive processing is handled by centralized servers. This partial action approach achieves real-time detection speed for critical functions without requiring excessive computational resources on edge devices.
3Reliability
If augmented reality devices monitor crowded areas continuously, then situational awareness is improved, but false alarms increase
Solution Approach 1:
The system implements feedback loops where artificial neural networks on edge devices continuously analyze video feeds and adjust their detection parameters based on observed patterns. This feedback mechanism enables the system to learn from false alarms and improve its accuracy over time, maintaining high situational awareness while reducing false positive rates in crowded areas.
Solution Approach 2:
The patent applies preliminary action by pre-training artificial neural networks on diverse datasets representing various crowded场景 and normal behaviors before deployment. This preliminary training allows the systems to distinguish between normal crowd dynamics and actual anomalies more effectively, reducing false alarms while maintaining comprehensive situational awareness.
Data Source
AI summary
An augment reality security system to identify outliers in behaviors. For example, cameras can be each configured to capture images, compress the images, and provide compressed images having embeddings representative of features determined by an artificial neural network. A server computer can receive, from the plurality of cameras, compressed images to generate analytics of embeddings of features in the compressed images, identify from the analytics an anomaly associated with a first face, and determine metrics representative of features of the first face in an image. A pair of augmented reality glasses can have a computing unit to detect a second face in a view through the glasses, communicate with the server computer to determine a match of the second face with the first face based on the metrics, and generate an augmented reality display in the view to identify the first face.