Video Range and Velocity Estimation Using Anthropometric Measures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing range and velocity measurement technologies are computationally intensive and unsuitable for portable devices, as they primarily capture two-dimensional velocity vectors, making them inefficient for use in firearm applications where resources are scarce.
Innovation Solution
A method that estimates target range using anthropometric measurements, such as head-and-shoulder outlines, and computes radial and angular velocity components separately, leveraging a priori knowledge and filtering techniques to reduce computational load and enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical flow methods are used to compute velocity in the image plane, then velocity measurement is achieved, but computational complexity increases and only two-dimensional velocity vectors are captured
Solution Approach 1:
The velocity vector is segmented into two independent components: radial velocity (computed from change in object size across frames) and angular velocity (computed from image plane motion). This segmentation allows each component to be calculated with simpler algorithms, reducing overall computational complexity while maintaining complete 3D velocity measurement.
Solution Approach 2:
The method transitions from computing only 2D image plane velocity to computing complete 3D velocity vectors by adding the radial component. This dimensional expansion is achieved through a separate ranging module that measures object size changes, providing the third dimension without requiring complex full 3D optical flow computations.
2Measurement precision
If complex range and velocity computations are performed, then measurement accuracy is improved, but power consumption increases making them unsuitable for portable devices
Solution Approach 1:
The ranging function is extracted as a separate module that operates independently from the velocity computation. By extracting ranging to compute object size changes and use anthropometric knowledge, the system reduces the computational burden on the main processing unit while maintaining accurate range and velocity measurements suitable for portable devices.
Solution Approach 2:
The system uses simple, low-cost computational models based on anthropometric knowledge (standard body dimensions) rather than expensive complex 3D modeling. These simplified models provide sufficient accuracy for military applications while consuming minimal computational resources and power.
3Device complexity
If anthropometric measurements are used for range estimation, then computational load is reduced, but measurement precision may be affected
Solution Approach 1:
The system incorporates feedback by comparing detected object size changes with expected anthropometric dimensions to validate and refine range estimates. This feedback mechanism ensures that simplified anthropometric models maintain sufficient measurement precision for military applications while keeping computational requirements low.
Data Source
AI summary
A system and method calculate a range and velocity of an object in image data. The range calculation includes detecting a contour of the object from the image data, forming a template from the image data based on the contour; and calculating a range to the object using pixel resolution and dimension statistics of the object. A three-dimensional velocity of the object is determined by calculating a radial component and an angular component of the velocity. The radial velocity component is calculated by determining the range of the object in two or more image frames, determining a time differential between the two or more image frames, and calculating the radial velocity as a function of the range of the object in the two or more image frames and the time differential between the two or more image frames. The angular component is calculated using spatial-temporal derivatives as a function of a motion constraint equation.