Gait Identification Device Footwear Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying individuals based on gait struggle with accuracy due to variations in the waveform caused by different types of footwear.
Innovation Solution
An identification device that includes a detection unit for identifying walking events from a walking waveform, a waveform processing unit for normalizing the waveform based on detected events, and an identification unit for recognizing users based on the normalized waveform.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If load information is used for identification, then personal authentication can be performed without hardware, but identification accuracy deteriorates due to waveform variations caused by different footwear
Solution Approach 1:
The patent applies preliminary action by detecting walking events (heel contact, toe off) before performing the identification process. By pre-identifying key gait events in the waveform, the system can segment and normalize the gait data, removing variations caused by footwear before comparison. This preliminary event detection enables subsequent accurate identification despite initial waveform differences.
Solution Approach 2:
The patent applies parameter changes by transforming the raw load waveform into a normalized representation based on detected walking events. The system changes the parameter reference frame from absolute load values to relative gait phase positions, allowing comparison of gait patterns independent of footwear-induced amplitude and timing variations.
2Device complexity
If raw walking waveform is used for identification, then the identification process is simple, but footwear type causes waveform differences that reduce identification reliability
Solution Approach 1:
The patent applies segmentation by dividing the continuous walking waveform into distinct segments based on detected walking events such as heel contact and toe off. By segmenting the gait cycle into measurable phases, the system can analyze specific portions of the waveform that are less affected by footwear variations, improving reliability while maintaining manageable processing complexity.
Solution Approach 2:
The patent introduces walking events (heel contact, toe off) as intermediary markers between the raw waveform and the identification process. These events serve as reference points that mediate the comparison process, allowing the system to align and compare gait patterns from different users regardless of footwear-induced waveform variations.
Data Source
AI summary
Provided is an identification device for identifying an individual on the basis of gait irrespective of the type of footwear, the identification device comprising a detection unit that detects a walking event on the basis of a walking waveform of a user, a waveform processing unit that normalizes the walking waveform on the basis of the detected walking event and generates a normalized waveform, and an identification unit that identifies the user on the basis of the normalized waveform.


