Edge Audio Surveillance for Real-Time Sound Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Automated surveillance systems face challenges in low visibility conditions and latency issues, particularly with video and thermal cameras, and existing audio surveillance methods lack specificity and real-time analysis capabilities, making them ineffective for immediate risk detection and notification.

Innovation Solution

A method for real-time surveillance using audio streams processed on-site via edge computing, which identifies and notifies relevant sound types of interest directly from devices located near the monitored location, without relying on cloud servers, utilizing machine learning and edge AI for immediate detection and notification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If video surveillance is used for automated detection, then visual monitoring capability is improved, but detection reliability deteriorates in low visibility conditions (night, fog, low light)

Engineering Contradiction:

Improvevisual monitoring capabilityVSAvoiddetection reliability

Core Design Contradiction:

Illumination intensityVSReliability

Solution Approach 1:

The surveillance system is segmented into multiple independent sensing modalities (video cameras, thermal cameras, audio sensors) that operate separately but are integrated through a common processing platform. This allows each sensor type to function optimally in its preferred environmental conditions while compensating for others' weaknesses.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system changes the operational parameters by switching between different sensing modalities based on environmental conditions. When visibility is poor, the system transitions from optical-based video detection to thermal-based detection and audio-based detection, effectively changing the physical parameter used for surveillance.

Inventive Principle:

Principle #35Parameter changes

2Power

If cloud server processing is used for audio stream analysis, then computational capability is improved, but response time deteriorates due to upload latency

Engineering Contradiction:

Improvecomputational capabilityVSAvoidresponse time

Core Design Contradiction:

PowerVSLoss of time

Solution Approach 1:

The system adds a spatial dimension to the processing architecture by distributing computational resources between edge devices (on-premises servers or local computers) and cloud servers. This multi-dimensional architecture allows real-time processing locally while maintaining cloud connectivity for non-critical functions.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system performs preliminary audio processing and analysis locally at the edge device before potentially uploading selected data to the cloud. This preliminary action filters out most processing needs, allowing only exceptional cases to be transmitted, thereby minimizing latency for critical detections.

Inventive Principle:

Principle #10Preliminary action

3Difficulty of detecting and measuring

If existing audio surveillance methods are used, then audio detection capability is improved, but detection specificity deteriorates due to lack of location-specific sound identification

Engineering Contradiction:

Improveaudio detection capabilityVSAvoiddetection specificity

Core Design Contradiction:

Difficulty of detecting and measuringVSMeasurement precision

Solution Approach 1:

The system implements local quality by customizing the audio detection profile for each specific location. Different locations have different background noise characteristics and relevant sound types, so the system tailors the detection algorithms and sound libraries to match each location's unique acoustic environment, thereby improving specificity.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The detection system is made dynamic by allowing continuous adjustment of detection parameters, sound type priorities, and sensitivity thresholds based on location-specific requirements. The system can adapt to changing environmental conditions and update its detection profile over time to maintain optimal specificity.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS11568887B1Devices, systems, and methods for real time surveillance of audio streams and uses therefore

Publication Date: 2023.01.31 AGLOGICA HOLDINGS INC

AI summary

Various examples are provided for surveillance of an audio stream. In one example, a method includes identifying presence or absence of a sound type of interest at a location during a time period; selecting the sound type from a library of sound type information to provide a collection of sound type information; incorporating the collection on a device proximate to the location; acquiring an audio stream from the location by the device to provide a locational audio stream; analyzing the locational audio stream to determine whether a sound type in the collection is present in the audio stream; and generating a notification to a user or computer if a sound type in the collection is present. The device can acquire and process the audio stream. In another example, a bulk sound type information library can be generated by identifying sound types of interest including them based upon a confidence level.