Optical Flow Field Construction Using Motion Sensors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current systems for constructing optical flow fields are complex, expensive, and unsuitable for real-time video imaging applications, particularly in aerial video imaging, due to their inability to efficiently handle the motion of mobile platforms and the status of imaging devices.

Innovation Solution

A method and system that categorize the motion of a mobile platform and the status of an imaging device into predefined scenarios, allowing for the pre-construction and selection of corresponding optical flow fields, using motion sensors and a gimbal to acquire and classify movements and rotations, enabling efficient construction of optical flow fields for various imaging scenarios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional estimation methods are used to obtain optical flow fields, then measurement precision can be achieved, but device complexity and processing time increase significantly

Engineering Contradiction:
Improveoptical flow field accuracyVSAvoidalgorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces complex image processing algorithms with a direct sensor measurement approach. Motion sensors (accelerometers, gyroscopes) directly measure platform motion, and the optical flow field is constructed by combining these motion measurements with simple geometric relationships, eliminating the need for complex pixel-level image analysis algorithms.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces motion sensors as an intermediary between the platform and the optical flow field construction. The sensors provide direct measurement data about platform motion, which serves as a bridge to simplify the optical flow calculation process and avoid complex image processing.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If traditional estimation methods are used to obtain optical flow fields, then measurement precision can be achieved, but processing speed decreases making it unsuitable for real-time applications

Engineering Contradiction:
Improveoptical flow field accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces computationally intensive image processing with direct sensor measurements and simple mathematical operations. Motion sensors provide real-time data about platform velocity and orientation, which can be directly transformed into optical flow field parameters without requiring frame-by-frame image analysis.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent performs preliminary measurement of platform motion using motion sensors before optical flow field construction is needed. This advance measurement of platform velocity and orientation allows the optical flow field to be constructed rapidly when required, enabling real-time processing.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If texture decomposition technologies are used, then optical flow field measurement precision is improved, but the method becomes inapplicable to real-time aerial video imaging applications

Engineering Contradiction:
Improveoptical flow field accuracyVSAvoidreal-time processing capability
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces texture decomposition image processing with direct inertial measurement. Motion sensors provide direct measurements of platform motion that are independent of image texture characteristics, eliminating the computational burden of texture analysis while maintaining measurement accuracy.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

4Measurement precision

If complex algorithms are used to construct optical flow fields, then measurement precision is maintained, but processing time increases making it unsuitable for mobile platforms

Engineering Contradiction:
Improveoptical flow field accuracyVSAvoidconstruction time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces complex image processing algorithms with direct sensor measurements. Motion sensors provide immediate data about platform motion, and the optical flow field is constructed through simple coordinate transformations rather than iterative image analysis, dramatically reducing construction time.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent performs preliminary measurement of platform motion parameters using motion sensors before optical flow field construction. This advance acquisition of velocity and orientation data allows the optical flow field to be constructed rapidly when needed, minimizing construction time.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3329461B1System and method for constructing optical flow fields
Publication Date: 2021.04.07 SZ DJI TECH CO LTD
  • EP3329461B1 patent drawingFigure 1
  • EP3329461B1 patent drawingFigure 2
  • EP3329461B1 patent drawingFigure 3

AI summary

A system (300) for constructing an optical flow field and methods (200) for making and using same. The system (300) can construct the optical flow field that corresponds with a selected video frame. The optical flow field can be constructed based on a first motion of a mobile platform (110) having an imaging device (116) and a status of the imaging device (116). The first motion and the status can be determined with measurements of sensors installed on the mobile platform (110) and/or the imaging device (116) installed on the mobile platform (110). The first motion can include a first rotation, a horizontal movement and a vertical movement of the mobile platform (110). The status can include a rotation of the imaging device (116) and/or an orientation of the imaging device (116) relative to the mobile platform (110). The system (300) advantageously can be applied to a wide range of video applications, including video applications involving mobile platforms (110), such as unmanned aerial vehicles.