Face Recognition Monitoring System Hash Table Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As the number of monitoring targets increases, the load and processing time for face feature matching in monitoring systems using face recognition technology also increase, leading to inefficiencies in matching processing.
Innovation Solution
The monitoring system employs a configuration with a first and second face feature quantity list, where the second list is a subset or empty set of the first, allowing for efficient matching by only comparing extracted face features against the relevant list based on the imaging device capturing the image, thereby reducing processing load and time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the number of monitoring targets increases, then the monitoring coverage and detection capability are improved, but the processing time and computational load for face feature matching increase
Solution Approach 1:
The patent divides the single large face feature quantity list into multiple smaller sub-lists stored in hash tables. Each sub-list contains face feature quantities for a specific range of monitoring targets. This segmentation reduces the number of comparisons needed during matching operations, thereby decreasing processing time while maintaining the ability to monitor all targets across the expanded coverage.
Solution Approach 2:
The patent performs preliminary organization of face feature quantities into structured hash tables with predefined ranges and sub-lists before the actual monitoring process. This pre-processing step creates an optimized data structure that enables faster retrieval and comparison during runtime, reducing the computational burden when processing images with multiple monitoring targets.
2Adaptability or versatility
If the number of monitoring targets increases, then the monitoring coverage is improved, but the computational load for matching processing increases
Solution Approach 1:
The patent segments the comprehensive face feature quantity list into multiple smaller sub-lists organized in hash tables. Each hash table contains face feature quantities for a specific range of targets, allowing the system to quickly identify and compare only the relevant subset rather than processing the entire list. This significantly reduces computational load while maintaining full monitoring coverage.
Solution Approach 2:
The patent implements local optimization by creating hash tables with different granularities and organizing face feature quantities into range-based sub-lists. This allows the system to apply different matching strategies for different portions of the data, comparing images against only the locally relevant subset of face feature quantities rather than the entire database, thereby reducing overall computational requirements.
3Device complexity
If a single comprehensive face feature quantity list is used, then the system is simple to manage, but the matching processing efficiency decreases
Solution Approach 1:
The patent divides the single comprehensive face feature quantity list into multiple smaller sub-lists organized within hash tables. Each hash table manages a specific range of monitoring targets, making the data structure more manageable despite the increased number of components. This segmentation dramatically improves matching efficiency by reducing the search space from the entire list to only the relevant subset for each comparison.
Solution Approach 2:
The patent introduces a new organizational dimension by arranging face feature quantities in hash tables with range-based sub-lists rather than using a single flat list. This multi-dimensional organization (hash table structure with indexed ranges) allows for efficient retrieval and comparison operations, transforming the data access pattern from linear search to hashed range-based search, thereby improving processing efficiency.
Data Source
AI summary
A monitoring system for monitoring a target is described. The monitoring system extracts features of a person's face from an image obtained from the face of the person and determines whether the person's face matches a person included in a list containing facial features of multiple individuals.