Edge Image Surveillance Model Distribution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Cloud-based analysis of edge device data for event detection sacrifices response time and consumes bandwidth, making edge-based analysis advantageous but not effectively implemented in existing systems.

Innovation Solution

A method and system where a management device identifies and transmits event detection models to image capture devices within specific geographic areas and time periods, allowing these devices to process image data locally for event identification and notification, with model data stored and used only as needed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If cloud-based analysis is used for event detection, then processing and memory requirements of edge devices are reduced, but response time is sacrificed

Engineering Contradiction:

Improveprocessing and memory requirementsVSAvoidresponse time

Core Design Contradiction:

Device complexityVSLoss of time

Solution Approach 1:

The system implements different analysis approaches at different locations: edge devices perform local event detection using deployed models for immediate response, while cloud infrastructure provides model training and management. This local differentiation resolves the contradiction by enabling fast local detection without requiring full processing power at the edge.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

Event detection models are pre-trained and deployed to edge devices before actual event detection occurs. This preliminary preparation allows edge devices to perform rapid local inference without real-time cloud processing, improving response time while maintaining reduced edge device complexity through pre-computed models.

Inventive Principle:

Principle #10Preliminary action

2Device complexity

If cloud-based analysis is used for event detection, then edge device processing requirements are reduced, but bandwidth consumption increases

Engineering Contradiction:

Improveprocessing requirementsVSAvoidbandwidth consumption

Core Design Contradiction:

Device complexityVSLoss of energy

Solution Approach 1:

The system extracts and transfers only the essential model data to edge devices, enabling them to perform local inference. This extraction approach reduces bandwidth consumption compared to continuous video streaming to the cloud, while still providing adequate processing capability at the edge through the deployed models.

Inventive Principle:

Principle #2Taking out (Extraction)

3Loss of time

If edge-based analysis is implemented, then response time is improved, but processing and memory requirements of edge devices increase

Engineering Contradiction:

Improveresponse timeVSAvoidprocessing and memory requirements

Core Design Contradiction:

Loss of timeVSDevice complexity

Solution Approach 1:

The system changes the parameters of edge devices by deploying optimized models with appropriate precision and size characteristics. This allows edge devices to perform local analysis with improved response time while keeping processing and memory requirements manageable through careful model parameter selection and optimization.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12088964B2Mobile image surveillance systems

Publication Date: 2024.09.10 GEOTAB INC

AI summary

Detection models are distributed as needed to a select subset of image capture devices, based on geographic location and time of the image capture devices. Each image capture device which receives a model processes image data according to the model, and provides a notification of any detected instances of an event the model is trained to detect. Distribution and detection can be based on historical data or live data.