Facial Recognition Image Selection for Photo Kiosks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in selecting images from large unclassified collections, particularly in photo kiosk systems, due to limited input devices and lack of permanent storage, making manual searching time-consuming and inefficient.
Innovation Solution
A method using facial recognition to analyze images, update a count variable for unique faces, and select images based on a predetermined condition, grouping them by temporal and geographical metadata to reduce the number of images presented to the user.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual searching is used to select images from a large collection, then users can find specific photographs, but the process becomes tedious and time-consuming
Solution Approach 1:
The system performs automatic facial recognition and image selection without requiring manual user intervention. The computer autonomously analyzes photographs, identifies faces, counts occurrences, and selects images containing persons of interest, making the system serve itself rather than requiring continuous user guidance.
Solution Approach 2:
The manual mechanical process of users viewing and selecting images is replaced with an automated computer-based facial recognition system. The system uses optical and computational methods to automatically identify faces and select images, substituting human manual operation with automated processing.
2Quantity of substance
If the photo kiosk system stores large collections of unclassified photographs, then users can access more images, but the lack of permanent storage and limited input devices makes searching more difficult
Solution Approach 1:
The system replaces manual searching operations with automated facial recognition processing. Instead of users manually navigating through images using limited input devices, the computer automatically analyzes all stored photographs, identifies faces, and selects relevant images based on facial occurrence counts.
Solution Approach 2:
The system changes the organizational parameter from unclassified storage to classification based on facial occurrence frequency. By counting how many times each face appears across the collection and selecting images where faces meet predetermined occurrence thresholds, the system transforms the search space into a manageable subset.
3Productivity
If automated facial recognition is implemented, then image selection time is reduced, but the system complexity increases
Solution Approach 1:
The system segments the image collection processing into distinct stages: facial recognition in each image, face matching across images, counting occurrences, and final selection. This segmentation allows the complex task to be broken into manageable computational steps that can be processed efficiently.
Solution Approach 2:
The system introduces a predetermined occurrence threshold parameter to filter and select images. By setting a minimum face occurrence count, the system automatically reduces the image collection to only those photographs containing persons of interest, simplifying the final selection process while maintaining high productivity.
Data Source
AI summary
Methods and systems for selecting one or more images from a plurality of captured images are disclosed. One method comprises the steps of: analysing the plurality of captured images to recognise faces in the captured images using a facial recognition process (step 304); updating a count variable relating to a unique face each time the face is recognised in the captured images; selecting one or more unique faces when the related count variable satisfies a predetermined condition (step 308); and selecting a reduced number of the images which together include at least one instance of each of the selected faces (step 310).