Tracking Fingerprint Verification for Adaptive Object Tracking

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional tracking systems face difficulties in maintaining object tracking when characteristics such as size, illumination, or occlusion change, leading to incorrect model modifications and loss of object tracking.

Innovation Solution

An image processing system with a first tracking system (TME) that generates and stores tracking fingerprints based on feature data, allowing for adaptive tracking by comparing generated fingerprints with stored ones to control the operation of a second tracking system, preventing incorrect model updates and maintaining object tracking even under changes or occlusion.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If an adaptive tracking system updates the model based on subsequent images, then the system can track objects with changing characteristics, but the system may modify the model based on incorrect information when the object is occluded or substantially different

Engineering Contradiction:

Improveadaptability to object characteristic changesVSAvoidaccuracy of model modifications

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces a fingerprint verification mechanism as an intermediary between the adaptive tracking system and the model updating process. Before the adaptive tracker modifies the object model based on image data, the system first generates a fingerprint from the image data and compares it against stored fingerprints. This verification step acts as a mediator that prevents incorrect model updates by validating whether the detected object truly matches the expected object characteristics, thereby resolving the contradiction between adaptability and reliability.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent implements a feedback mechanism where the fingerprint comparison results are fed back into the model updating decision process. The system continuously monitors the fingerprint similarity between current image data and stored fingerprints, and uses this feedback information to control whether the adaptive tracker should update the model. This feedback loop ensures that only reliable, verified object detections trigger model modifications, preventing propagation of incorrect information while maintaining adaptability to legitimate changes.

Inventive Principle:

Principle #23Feedback

2Productivity

If a non-adaptive tracking system uses a single model, then the system is simple and fast, but the system loses track of the object when characteristics change substantially

Engineering Contradiction:

Improvetracking speed and efficiencyVSAvoidability to handle object characteristic changes

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent segments the tracking function into two independent components: a fast non-adaptive tracker that maintains a single model for efficient processing, and a fingerprint verification system that handles adaptability. The non-adaptive tracker continues to operate at full speed using its simple single-model approach, while the fingerprint system independently verifies object identity and controls model updates only when necessary. This segmentation allows the system to maintain high productivity while achieving adaptability through the fingerprint verification mechanism.

Inventive Principle:

Principle #1Segmentation

3Adaptability or versatility

If the adaptive tracking system modifies the model based on incorrect information, then the system tracks incorrect objects, but the system cannot recover when the correct object is no longer occluded

Engineering Contradiction:

Improveresponse to occlusion and object reappearanceVSAvoidloss of correct object identity information

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent applies preliminary action by performing fingerprint verification before the adaptive tracker modifies the model. This advance verification step ensures that the system only updates the model when the detected object is confirmed to be the correct object, preventing incorrect information from being stored. When an occluded object reappears, the fingerprint comparison allows the system to recognize the correct object identity and recover tracking, preventing loss of correct object information.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10872425B2System and method to improve object tracking using tracking fingerprints

Publication Date: 2020.12.22 THE BOEING CO

AI summary

A method includes receiving image data at a first tracking system. The image data represents a region in an image of a sequence of images. The method also includes generating a first tracking fingerprint based on the image data. The method further includes comparing the first tracking fingerprint and a second tracking fingerprint. The method also includes providing an output from the first tracking system to a second tracking system based on a result of the comparison of the first tracking fingerprint and the second tracking fingerprint. The output includes an instruction associated with an object model stored at the second tracking system.