Magnetic Elevator Button Sensing for Hygienic Noncontact Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Elevator buttons pose hygiene risks due to direct manipulation, particularly in infectious environments, and existing noncontact methods like high-specification video processing are costly and inaccurate in indoor conditions.
Innovation Solution
A magnetic field sensing-based system collects sensor data, assigns weight values to a Z-axis, adjusts weight function widths, and applies kernel window sizes to enhance button recognition accuracy using machine learning algorithms like KNN, minimizing hardware costs and improving recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If direct contact button manipulation is used, then ease of operation is improved, but hygiene is worsened due to infection risks
Solution Approach 1:
The patent replaces mechanical contact-based button operation with a magnetic field sensing system. Magnetic sensors detect the approach of a user's hand or finger without physical contact, triggering button activation through magnetic field changes rather than mechanical pressure. This substitution eliminates direct contact while maintaining operational ease.
Solution Approach 2:
The patent introduces magnetic field detection as an intermediary between the user and the button system. Instead of direct hand-to-button contact, the user's hand creates magnetic field disturbances that are detected by magnetic sensors, serving as a noncontact mediator that preserves hygiene while enabling operation.
2Ease of operation
If high-specification video processing is used for noncontact recognition, then noncontact operation is achieved, but cost and accuracy in indoor conditions worsen
Solution Approach 1:
The patent changes the detection parameter from optical (video processing) to magnetic field sensing. Magnetic field parameters are less affected by indoor lighting conditions, camera quality, and image processing complexity. This parameter change improves recognition accuracy and reliability in indoor environments while reducing hardware requirements.
Solution Approach 2:
The patent replaces expensive high-specification video processing hardware with simpler, lower-cost magnetic field sensors. The magnetic sensing approach requires basic magnetic sensors rather than high-resolution cameras and complex processing units, significantly reducing system cost while maintaining noncontact operation capability.
3Measurement precision
If magnetic field sensing with machine learning is used, then recognition accuracy is improved, but device complexity increases
Solution Approach 1:
The patent implements preliminary machine learning training offline to create optimized recognition models. During actual operation, the pre-trained model quickly classifies magnetic field patterns without requiring complex real-time computation. This preliminary action separates the complex learning process from the operational phase, reducing real-time device complexity while maintaining high accuracy.
Solution Approach 2:
The system uses collected magnetic field data from actual button interactions to continuously refine and retrain the machine learning model. The system serves itself by automatically learning from operational data, improving accuracy over time without requiring external intervention or complex manual calibration procedures.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system provides accurate, noncontact elevator button operation with reduced hardware costs and enhanced recognition, minimizing malfunctions and maintaining hygiene by reducing direct contact.
Implementation Method 1
collecting a magnetic field sensor value corresponding to each of a plurality of buttons on the basis of an action of a user for button manipulation
Data Source
AI summary
This application relates to a magnetic field sensing-based noncontact button apparatus, an elevator control panel, and an operating method thereof. In one aspect, the operating method includes collecting a magnetic field sensor value corresponding to each of a plurality of buttons based on an action of a user for button manipulation. The method also includes assigning a weight value, used for activating a button, to a Z-axis value of the magnetic field sensor value corresponding to a button input direction of the user for each button and collecting number of uses of each button. The method further includes adjusting a variation width of a weight function based on the Z-axis value based on the number of uses of each button to construct a surface data set corresponding to each button, and setting the surface data set to learning data to learn an algorithm for activating the button.