Progressive Analysis for Occluded Data Assessment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current surveillance systems, both video and extra-video, face challenges in detecting occluded data and combining analysis results from multiple sensors, leading to incomplete detection and reduced confidence in analysis, especially when dealing with moving targets and occlusions.

Innovation Solution

A method and system for progressive analysis that combines sensor data from multiple views and sensors to resolve occluded areas and increase confidence in non-occluded data, using algorithmic procedures for tracking, segmentation, and voting mechanisms to provide holistic assessment and visualization of surveillance results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple sensors surveying the same scene from different views are employed, then the ability to detect occluded areas is improved, but the device complexity increases

Engineering Contradiction:

Improvedetection completenessVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system divides the surveillance scene into multiple viewports, each handled by a dedicated sensor. The analysis is segmented by tracking targets across individual viewports and then combining results. This allows complex multi-sensor data to be processed through simpler, modular units before integration.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent combines analysis results from multiple sensors viewing the same scene from different views into a single comprehensive data set. The voting algorithm merges individual sensor assessments to produce a unified detection result, resolving occluded areas through consensus of multiple perspectives.

Inventive Principle:

Principle #5Merging (Combining)

2Productivity

If algorithmic surveillance processes are implemented, then productivity is improved, but the ability to handle occluded data is worsened

Engineering Contradiction:

Improvesurveillance efficiencyVSAvoidoccluded data detection

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs preliminary actions by analyzing multiple frames and sensors in advance to identify occluded areas before they become critical. The voting algorithm is prepared to resolve occlusions when sufficient data becomes available, maintaining continuous surveillance without waiting for complete visibility.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms where analysis results from one viewport or sensor feed back into the overall assessment. When a target is detected in one view, the system waits for subsequent frames or alternative sensors to confirm or resolve occluded portions, continuously updating the assessment based on new information.

Inventive Principle:

Principle #23Feedback

3Loss of time

If continuous surveillance is maintained, then loss of time is reduced, but the ability to resolve occluded areas is worsened

Engineering Contradiction:

Improvesurveillance downtimeVSAvoidocclusion resolution

Core Design Contradiction:

Loss of timeVSReliability

Solution Approach 1:

The system maintains continuous surveillance by keeping all sensors active and continuously analyzing the scene. Even when areas are occluded, the system continues to process data from visible areas and prepare to resolve occlusions when they clear, ensuring uninterrupted surveillance coverage.

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The system performs partial analysis on visible areas while waiting for occluded areas to become visible. The voting algorithm is prepared with preliminary results from unoccluded areas, allowing the system to maintain surveillance continuity while gradually building confidence in the complete assessment as occluded areas resolve.

Inventive Principle:

Principle #16Partial or excessive action

4Reliability

If redundant analysis of non-occluded data is performed, then confidence in analysis results is improved, but use of energy is worsened

Engineering Contradiction:

Improveanalysis confidenceVSAvoidenergy consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system applies redundant analysis selectively to non-occluded areas where confidence building is most valuable. The voting algorithm focuses computational resources on areas with sufficient data availability, performing multiple analyses only where the target is visible and detectable, rather than exhaustively analyzing all areas uniformly.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS8913129B2Method and system of progressive analysis for assessment of occluded data and redundant analysis for confidence efficacy of non-occluded data

Publication Date: 2014.12.16 THERMAL MATRIX USA

AI summary

A method of progressive analysis for assessment of occluded data is disclosed. In a particular embodiment, the method includes capturing sensor data from at least one sensor, identifying a target of interest using the sensor data, and determining a location and orientation of the target of interest using a last computed direction of travel of the target. The method also includes segmenting the sensor data of the target of interest into segments for analysis and determining whether a particular segment of the segments is occluded. In addition, the method includes determining whether additional sensor data from the at least one sensor confirms that the particular occluded segment is no longer occluded allowing unresolved analysis to be supplemented with progressive analysis to provide incremental resolution to the analysis.