Drone Adaptive Positioning for Face Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current face recognition systems using drone-mounted imaging sensors face challenges in achieving reliable and robust identification due to factors like head pose, environmental obstructions, and illumination, which affect the quality and accuracy of facial images, leading to potential misclassification and security vulnerabilities.
Innovation Solution
A method and system that dynamically adjust the drone's position to optimize image capture based on identified positioning properties such as head pose, environmental parameters, and obstructions, using machine learning models to iteratively improve the visibility and quality of facial images until a sufficient confidence threshold is met, and optionally incorporating additional authentication methods for enhanced security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the drone uses a fixed position for capturing facial images, then the device complexity is reduced, but the face recognition reliability deteriorates due to varying head poses, environmental obstructions, and illumination conditions
Solution Approach 1:
The patent implements dynamic positioning of the drone by continuously adjusting its spatial coordinates based on real-time analysis of positioning properties (head pose, obstructions, illumination). The drone transitions from a static capture position to a dynamic adaptive positioning system that modifies its location iteratively to optimize facial image quality, directly resolving the contradiction between fixed simplicity and reliability.
Solution Approach 2:
The system employs feedback mechanisms where machine learning models analyze captured images and positioning properties, then feed this information back to adjust the drone's position. This closed-loop control system continuously refines the drone's location based on image quality metrics and environmental factors, improving face recognition reliability without requiring overly complex manual positioning systems.
2Measurement precision
If the drone captures images from multiple positions to improve face recognition accuracy, then the measurement precision is improved, but the loss of time increases due to multiple iterations and repositioning
Solution Approach 1:
The system performs preliminary analysis of positioning properties (head pose, potential obstructions, illumination conditions) before capturing the actual facial image. By pre-assessing these factors and adjusting the drone's position in advance, the system reduces the need for multiple retry iterations, thereby improving facial image quality while minimizing time loss.
Solution Approach 2:
The patent changes multiple parameters simultaneously including drone spatial position, capture angle, and timing, based on machine learning model predictions. This multi-parameter optimization allows the system to achieve high measurement precision in fewer iterations by coordinating position adjustments with capture timing, reducing overall processing time.
3Measurement precision
If the system uses complex machine learning models to achieve high accuracy face recognition, then the measurement precision is improved, but the computational resources required increase
Solution Approach 1:
The system applies partial action by using simplified machine learning models for initial assessments and only deploying more complex analysis when necessary. The machine learning models process partial features first (basic face detection, then progressive refinement), allowing the system to achieve high accuracy while minimizing computational energy consumption through staged processing.
Solution Approach 2:
The machine learning models perform preliminary analysis of positioning properties and image quality metrics before full face recognition processing. This preliminary filtering allows the system to identify and correct suboptimal capture conditions early, reducing the need for computationally intensive reprocessing and lowering overall energy consumption while maintaining high accuracy.
4Manufacturing precision
If the drone approaches closer to the target person to capture higher quality images, then the manufacturing precision of image quality is improved, but the object-generated harmful factors increase due to potential disturbances and security concerns
Solution Approach 1:
The system optimizes multiple parameters including drone distance, altitude, and capture angle to achieve high image quality without requiring excessive proximity to the target person. By adjusting these parameters in combination, the system maintains manufacturing precision (image quality) while minimizing the harmful effects of close approach, such as disturbance or security concerns.
Data Source
AI summary
Presented herein are systems, methods and apparatuses for increasing reliability of face recognition in analysis of images captured by drone mounted imaging sensors, comprising: recognizing a target person in one or more iterations, each iteration comprising: identifying one or more positioning properties of the target person based on analysis of image(s) captured by imaging sensor(s) mounted on a drone operated to approach the target person, instructing the drone to adjust its position to an optimal facial image capturing position selected based on the positioning property(s), receiving facial image(s) of the target person captured by the imaging sensor(s), receiving a face classification associated with a probability score from machine learning model(s) trained to recognize the target person, and initiating another iteration in case the probability score does not exceed a certain threshold. Finally, the face classification may be outputted for use by one or more face recognition based systems.