UAV Visual Tracking via Pixel Set Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) face challenges in tracking a moving human subject through a crowd, as existing systems struggle to distinguish the subject from others with similar visual characteristics, and maintain a consistent perspective while accounting for changes in direction and velocity.
Innovation Solution
The UAV is equipped with a camera and a visual recognition system that generates pixel sets from images, allowing it to analyze and compare reference and tracking images to identify and track the subject, adjusting rotor speeds via an attitude control system to maintain a desired orientation relative to the subject.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the UAV uses basic image capture without advanced processing, then the system complexity is low, but the ability to distinguish the subject from others in a crowd is insufficient
Solution Approach 1:
The patent segments the subject identification process into multiple components: capturing images from multiple orientations, generating pixel sets from each image, comparing pixel sets to identify the subject, and tracking subject movement. This segmentation allows the system to achieve high identification accuracy by breaking down the complex task into manageable processing stages, each handling a specific aspect of subject recognition.
Solution Approach 2:
The patent captures images from multiple orientations of the UAV (different angles and positions) to create a multi-dimensional view of the subject. By generating pixel sets from these various orientations and comparing them, the system enhances subject identification accuracy beyond what a single viewpoint could provide, effectively adding dimensional information to the recognition process.
2Adaptability or versatility
If the UAV maintains a fixed orientation relative to the subject, then the perspective consistency is good, but the system cannot account for changes in subject direction or velocity
Solution Approach 1:
The patent implements a feedback mechanism where the UAV continuously captures images, generates pixel sets, compares them to track subject movement, and uses this information to adjust its orientation. This closed-loop feedback system allows the UAV to adapt to changes in subject direction and velocity while maintaining reliable tracking, as the system constantly monitors and responds to subject movement patterns.
Solution Approach 2:
The patent transitions from a static fixed-orientation approach to a dynamic tracking system where the UAV's orientation changes in response to subject movement. The system dynamically adjusts the UAV's position and angle by comparing pixel sets from consecutive images, enabling it to follow the subject's changing direction and velocity while maintaining consistent perspective through active adaptation.
3Measurement precision
If the UAV captures images at multiple orientations to improve tracking, then the tracking accuracy improves, but the data processing complexity increases
Solution Approach 1:
The patent extracts only the essential information needed for tracking by generating pixel sets from captured images and comparing these compressed representations rather than processing the full high-resolution images. This extraction approach maintains tracking precision by preserving key subject features while significantly reducing the computational complexity of processing multiple oriented images.
Data Source
AI summary
An unmanned aerial vehicle (UAV) may select a subject to track or follow and capture images including the subject. A visual recognition system of the UAV may generate a profile of reference images portraying the subject breaking the reference images down into pixel sets corresponding to the subject and his/her components and distinguishing characteristics (body parts, clothing, facial features, accessories). The visual recognition system may break the incoming stream of incoming images into pixel sets, analyzing the pixel sets to distinguish the subject from his/her surroundings (i.e., in a crowd) and determine movement of the subject and the current orientation of the UAV to the subject. The UAV may then change its heading, velocity, or position based on any difference between the current orientation and the desired or predetermined orientation between the UAV and the subject.