Dual Video Analytics Algorithm for Event Validation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video processing algorithms face a trade-off between computational power and accuracy, with simpler algorithms being less accurate and more comprehensive algorithms requiring more resources, necessitating a method to balance these factors for improved event recognition in video streams.

Innovation Solution

A dual-algorithm approach is employed, where a less computationally intensive first algorithm identifies candidate events, and a more comprehensive second algorithm validates these events by comparing against ground truth, with the first algorithm being retrained if validation fails, ensuring accurate event reporting and reducing false positives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by moving object

If a simpler video analytics algorithm is used, then computational power requirements are reduced, but event recognition accuracy deteriorates

Engineering Contradiction:

Improvecomputational power requirementsVSAvoidevent recognition accuracy

Core Design Contradiction:

Use of energy by moving objectVSMeasurement precision

Solution Approach 1:

The video analytics processing is segmented into two stages: a first simpler algorithm performs initial event detection to identify candidate events, reducing computational load on the overall system. A second more comprehensive algorithm then validates these candidates by comparing against ground truth. This segmentation allows the system to use less computational power while maintaining accuracy through the two-stage verification process.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If a more comprehensive video analytics algorithm is used, then event recognition accuracy is improved, but computational power requirements increase

Engineering Contradiction:

Improveevent recognition accuracyVSAvoidcomputational power requirements

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system applies the more comprehensive second video analytics algorithm only partially - specifically, only to candidate events identified by the first algorithm rather than to all video data. This partial application of the comprehensive algorithm reduces overall computational power requirements while still achieving high accuracy for the events that are detected and validated.

Inventive Principle:

Principle #16Partial or excessive action

3Productivity

If a simpler video analytics algorithm is used, then processing speed is improved, but false positive rate increases

Engineering Contradiction:

Improveprocessing speedVSAvoidfalse positive rate

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The first simpler video analytics algorithm acts as an intermediary that performs rapid initial detection at high processing speed. The second more comprehensive algorithm then serves as a mediator to validate the candidates and filter out false positives by comparing against ground truth. This intermediary approach maintains high processing speed while reducing the false positive rate through the validation stage.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20250111673A1System and method for identifying events in a video stream

Publication Date: 2025.04.03 HONEYWELL INTERNATIONAL INC

AI summary

Events may be identified in a video stream using a first video analytics algorithm and the identified events may be validated using a second video analytics algorithm that is more comprehensive than the first video analytics algorithm. The event as identified by the first video analytics algorithm may be reported when the identified event is validated by the second video analytics algorithm.