Edge Server Visitor Analysis Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques face challenges in integrating data on visitor numbers and demographic information from separate systems, and there are concerns about privacy protection when processing and transmitting videos containing sensitive personal information.
Innovation Solution
A method and device that use an edge computing environment to extract feature data from videos, generate detection data using an artificial neural network-based model, and integrate location and appearance data of visitors, reducing the need for external server communication and minimizing privacy risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If separate systems are built for counting visitors and identifying demographic information, then both functions can be performed, but it becomes difficult to integrate and utilize information from the two systems
Solution Approach 1:
The patent merges the visitor counting system and demographic identification system into a single integrated system. The server simultaneously performs both functions by processing video data to detect visitor entry/exit events and analyze demographic information (gender, age) from the same video feed, eliminating the need for separate systems and enabling seamless data integration.
Solution Approach 2:
The server is designed with multi-functionality to handle both visitor counting and demographic analysis tasks. It can identify visitors as they enter or exit, count total visitors, and simultaneously extract demographic information from video data, making a single system capable of performing multiple functions that were previously required separate systems.
2Loss of information
If video capturing appearances (faces) of visitors is transmitted or processed externally, then demographic information can be identified, but there is a risk of legal issues related to privacy protection
Solution Approach 1:
The patent extracts only the necessary demographic information (gender, age estimates) from the video data while not retaining or transmitting the actual facial images or personal identifiable information. The server processes the video locally to extract demographic features and then discards the original video data, keeping only the anonymized demographic statistics, thus eliminating privacy risks while maintaining information utility.
Solution Approach 2:
The server acts as an intermediary that processes video data locally to extract demographic information without transmitting sensitive personal information to external systems. By performing all processing in-house and only retaining anonymized demographic data, the server mediates between the need for demographic information and privacy protection requirements.
3Reliability
If analysis is performed on a server, then comprehensive processing can be done, but time is required for communication between device and server
Solution Approach 1:
The system performs preliminary action by pre-processing and analyzing video data on the server before any external communication is needed. The server continuously monitors video feeds, detects visitor events, and extracts demographic information in advance, so that when analysis is needed, the data is already prepared and available immediately without requiring real-time communication delays.
Data Source
AI summary
A method for analyzing a visitor on the basis of a video in an edge computing environment is provided. The method includes the steps of: extracting feature data from a captured video of an offline space; generating detection data on a location and an appearance of an object contained in the captured video from the feature data using an artificial neural network-based detection model; and integrating detection data of a location and an appearance of a target object.