Person Detection via Wireless Signal and Camera Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing monitoring systems in retail stores, airports, and smart areas face challenges in accurately identifying individuals using facial recognition algorithms, especially in crowded environments, and current wireless tracking methods provide coarse-grained location data due to reliance on RSSI, which lacks accuracy and requires multiple infrastructure anchors.
Innovation Solution
A system combining a camera and wireless transceiver that uses CSI data from a single system unit with multiple antennas to capture motion signatures, fusing camera image data and wireless packet data to identify individuals by comparing motion trajectories and pseudo-identifying people using Wi-Fi or Bluetooth signals, reducing the need for multiple anchors and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If facial recognition algorithms are used for person identification, then identification capability is provided, but accuracy deteriorates in crowded environments with thousands of people
Solution Approach 1:
The system segments the identification process into multiple independent components: wireless signal-based detection for initial person detection and tracking, camera-based capture for visual data, and fused identification for final recognition. This segmentation allows each component to operate independently with optimized accuracy, avoiding the overload of single-algorithm approaches in crowded environments.
Solution Approach 2:
The system merges wireless signal data (packet data with amplitude information) and camera image data into a unified identification framework. By combining motion information from wireless signals with visual data from cameras, the system achieves more robust and accurate person identification in crowded environments than either modality alone could provide.
2Loss of information
If RSSI-based wireless tracking is used, then location data is obtained, but accuracy deteriorates and multiple infrastructure anchors are required
Solution Approach 1:
The system changes the parameter used for wireless tracking from RSSI (received signal strength indicator) to CSI (channel state information), specifically utilizing amplitude information from wireless packets. This parameter change provides more accurate and fine-grained location data without requiring multiple infrastructure anchors, as CSI contains richer spatial information about the wireless channel.
3Area of stationary object
If cameras are mounted in ceiling looking downward, then coverage area is maximized, but identification accuracy deteriorates due to inability to capture facial features
Solution Approach 1:
The system introduces wireless signals as an intermediary for obtaining motion information. By using CSI-based motion signatures from wireless packets, the system can track person movement and infer identification without requiring the camera to directly capture facial features, thus maintaining ceiling-mounted coverage while achieving accurate identification through fused data analysis.
Data Source
AI summary
An apparatus comprising a wireless transceiver configured to communicate packet data with a mobile device associated with one or more persons in a vicinity of the wireless transceiver, and a controller in communication with the wireless transceiver and a camera, the controller configured to receive a plurality of packet data from one or more person, wherein the packet data includes at least amplitude information associated with the wireless channel communicating with the wireless transceiver and receive images from the camera containing trajectory of motion of individuals, performs detection, tracking, and pseudo-identification of individuals by fusing motion trajectories from wireless signals and camera images.