Step Counting Accuracy via Axis Characteristic Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing step counting methods have low accuracy due to similarities in acceleration data from walking and other activities, leading to erroneous step counting.
Innovation Solution
A method that acquires acceleration datasets from an acceleration sensor, determines characteristic values, compares axes to identify a characteristic axis, and judges the validity of step counting data using a reference axis, performing step counting only when data is valid, and adjusts the reference axis based on motion state changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If acceleration data is used to count steps, then step counting function is provided, but accuracy of step counting decreases due to similarity between walking and other activities
Solution Approach 1:
The patent segments the acceleration data analysis by dividing it into three orthogonal components (x, y, z axes). Instead of analyzing acceleration data as a single value, the method separately processes each axis component, calculates characteristic values for each, and identifies the characteristic axis with the maximum characteristic value. This segmentation allows the system to distinguish between different motion patterns more effectively, as walking typically produces characteristic acceleration patterns on specific axes while other activities produce different patterns.
2Measurement precision
If simple acceleration threshold method is used, then device complexity is reduced, but step counting accuracy decreases
Solution Approach 1:
The patent transforms the acceleration data by calculating characteristic values from the raw acceleration datasets. This parameter transformation converts the original acceleration values into new characteristic parameters that better represent the motion state. By changing the parameter representation (from raw acceleration to characteristic values derived from multiple data points), the system achieves more accurate step detection without requiring complex hardware modifications.
Data Source
AI summary
A step counting judgment method and device and a computer-readable storage medium are disclosed. The step counting judgment method comprises: acquiring an acceleration dataset corresponding to each axis of an acceleration sensor in a current step counting cycle; determining a characteristic value corresponding to each axis according to the acceleration dataset corresponding to each axis; comparing characteristic values of different axes, and determining that an axis corresponding to a maximum value in the characteristic values is a characteristic axis; judging validity of step counting data according to the characteristic axis and a reference axis; and performing a step counting operation when the step counting data is valid data. Aims to solve the problem of low step counting accuracy of step counting judgment devices, the present disclosure provides the step counting judgment method and device and the computer-readable storage medium.