Hybrid Camera Object Tracking with Spectral Material Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object tracking systems in videos face challenges with identifying objects in the presence of similar-sized, shaped, or colored objects, partial occlusions, and shadows, leading to performance degradation, and require high computational complexity and cost for spectral imaging.
Innovation Solution
A hybrid camera system that combines a conventional camera for high spatial and frame rate visual information with a spectral sensor for multi-spectral data collection, allowing for material identification and reduced computational complexity by analyzing spectral images to track objects across the video scene.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If spectral imaging is used for material identification and object tracking, then object identification accuracy is improved, but cost and computational complexity increase significantly
Solution Approach 1:
The patent divides the tracking task into two segments: conventional cameras handle motion detection and rough object localization, while spectral imaging is applied only to identified object regions for material verification. This segmentation allows spectral imaging to be used selectively rather than continuously, reducing computational complexity while maintaining identification accuracy.
Solution Approach 2:
Instead of applying spectral imaging to the entire scene or continuously, the system applies spectral analysis only to specific regions containing objects of interest that have been pre-identified by conventional tracking algorithms. This partial application of spectral imaging reduces data processing requirements while maintaining the ability to identify materials accurately.
2Measurement precision
If spectral imaging is used for material identification, then material detection capability is improved, but data storage capacity requirements increase
Solution Approach 1:
The patent extracts only the essential spectral features needed for material identification from the full spectral data, rather than storing complete spectral images. By extracting key spectral signatures and comparing them against known material databases, the system maintains material detection capability while significantly reducing the storage requirements.
Solution Approach 2:
The system performs spectral analysis only on selected regions of interest containing potential objects, rather than processing and storing spectral data for the entire scene. This selective approach reduces the volume of data that needs to be stored while maintaining the ability to identify materials in objects of interest.
3Device complexity
If conventional tracking methods are used, then system simplicity is maintained, but tracking reliability degrades in presence of similar objects, shadows, and occlusions
Solution Approach 1:
The patent merges conventional tracking methods with spectral imaging in a hybrid system. Conventional cameras provide continuous motion tracking and object localization, while spectral imaging is integrated to provide material identification and verification. This combination allows the system to maintain simplicity of conventional tracking while adding spectral capability to improve reliability in challenging scenarios.
Solution Approach 2:
The patent substitutes spectral analysis for mechanical/visual tracking in specific scenarios where conventional methods fail. When objects have similar appearance, cast shadows, or are partially occluded, the system uses spectral signatures to distinguish and track the correct object, replacing reliance on visual appearance with reliance on material properties.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system provides robust object identification and tracking with reduced computational complexity, enabling efficient material analysis and alert signaling for objects of interest, such as explosives, without increasing cost or complexity.
Implementation Method 1
The spectral image comprises different spectral planes each having pixel locations corresponding to a reflectance obtained at a wavelength band of interest
Data Source
AI summary
What is disclosed is a system and method for identifying materials comprising an object captured in a video and for using the identified materials to track that object as it moves across the captured video scene. In one embodiment, a multi-spectral or hyper-spectral sensor is used to capture a spectral image of an object in an area of interest. Pixels in the spectral planes of the spectral images are analyzed to identify a material comprising objects in that area of interest. A location of each of the identified objects is provided to an imaging sensor which then proceeds to track the objects as they move through a scene. Various embodiments are disclosed.