Self-adaptive AI for Real-time Video Surveillance

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional surveillance systems fail to prevent crimes in advance due to inaccuracies in facial recognition, inability to detect unknown threats, and lack of real-time analysis of behavioral patterns, leading to ineffective crime prevention and public safety management.

Innovation Solution

A self-adaptive AI system that generates space-time-behaviour datasets from video data to identify predefined events and causes, determining suspicious activities by comparing these datasets with predefined parameters, providing early warning signals for security breaches.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If facial recognition technology is used to identify criminals, then evidence collection is improved, but crime prevention capability deteriorates

Engineering Contradiction:

Improveidentification accuracyVSAvoidcrime prevention capability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The system performs preliminary analysis of behavioral patterns, spatial relationships, and temporal sequences before crimes occur. By generating space-time-behaviour datasets and identifying suspicious activity patterns in advance, the system enables preventive intervention rather than just post-crime identification.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system transitions from traditional 2D image-based facial recognition to a multi-dimensional space-time-behaviour analysis framework. This includes spatial position, temporal sequences, behavioral patterns, and contextual relationships, enabling detection of suspicious activities that precede crimes.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Ease of manufacture

If conventional image-based techniques are used to analyze events, then implementation simplicity is improved, but detection accuracy of suspicious activities deteriorates

Engineering Contradiction:

Improveimplementation simplicityVSAvoidbehavioral analysis accuracy

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The system segments video data into discrete space-time-behaviour datasets for analysis. Each dataset contains segmented spatial coordinates, temporal markers, and behavioral features that can be independently processed and analyzed to identify suspicious patterns.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces space-time-behaviour datasets as an intermediary representation between raw video data and crime detection. This intermediary layer extracts meaningful features such as spatial relationships, temporal sequences, and behavioral patterns that bridge simple image processing and complex threat detection.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If real-time monitoring of all activities is implemented, then detection capability is improved, but system complexity deteriorates

Engineering Contradiction:

Improvethreat detection capabilityVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system applies partial monitoring by focusing on specific space-time-behaviour parameters that are most indicative of suspicious activities. Rather than analyzing every pixel and movement, it selectively monitors key behavioral features, spatial relationships, and temporal patterns that are most relevant to crime prevention.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system changes the parameters of analysis from raw pixel data to extracted space-time-behaviour features. By transforming video data into standardized datasets with specific parameters (spatial coordinates, temporal markers, behavioral descriptors), the system simplifies real-time processing while maintaining detection effectiveness.

Inventive Principle:

Principle #35Parameter changes

4Speed

If facial recognition is applied for real-time monitoring, then identification speed is improved, but accuracy in detecting disguised faces deteriorates

Engineering Contradiction:

Improveidentification speedVSAvoiddisguised face detection accuracy

Core Design Contradiction:

SpeedVSMeasurement precision

Solution Approach 1:

The system uses multi-functional space-time-behaviour analysis that can identify both known individuals and suspicious behaviors regardless of facial appearance. The analysis encompasses facial features, body language, spatial movements, and temporal patterns, making it universally applicable to both identified and unidentified subjects.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20230386212A1Method and system for monitoring activities and events in real-time through self-adaptive ai

Publication Date: 2023.11.30 SKYLARK LABS INC

AI summary

This disclosure relates to method and system for monitoring activities and events in real-time. The method includes receiving video data of an area from each of one or more cameras. The video data includes a plurality of frames. For each frame of the plurality of frames, the method further includes generating in real-time, a space-time-behaviour dataset corresponding to the frame; identifying in real-time, at least one of an event from a plurality of predefined events, or an associated cause of an event from a plurality of predefined causes, based on the space-time-behaviour dataset; determining in real-time, a set of cause parameters or a set of event parameters, based on the space-time-behaviour dataset; and determining in real-time, whether at least one of the cause or the event corresponds to suspicious activity based on the comparison with parameters of the plurality of predefined causes and the plurality of predefined events.