CN107462258A - A kind of step-recording method based on mobile phone 3-axis acceleration sensor - Google Patents
A kind of step-recording method based on mobile phone 3-axis acceleration sensor Download PDFInfo
- Publication number
- CN107462258A CN107462258A CN201710569013.4A CN201710569013A CN107462258A CN 107462258 A CN107462258 A CN 107462258A CN 201710569013 A CN201710569013 A CN 201710569013A CN 107462258 A CN107462258 A CN 107462258A
- Authority
- CN
- China
- Prior art keywords
- data
- calculation window
- peak
- axis
- acceleration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000001133 acceleration Effects 0.000 title claims abstract description 128
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000004364 calculation method Methods 0.000 claims abstract description 93
- 238000009499 grossing Methods 0.000 claims abstract description 19
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 5
- 238000005070 sampling Methods 0.000 claims description 84
- 230000006870 function Effects 0.000 claims description 19
- 230000033001 locomotion Effects 0.000 claims description 11
- 238000013145 classification model Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 abstract description 10
- 238000007635 classification algorithm Methods 0.000 abstract description 4
- 230000011218 segmentation Effects 0.000 abstract description 3
- 238000007781 pre-processing Methods 0.000 abstract description 2
- 238000013480 data collection Methods 0.000 abstract 1
- 238000005457 optimization Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C22/00—Measuring distance traversed on the ground by vehicles, persons, animals or other moving solid bodies, e.g. using odometers, using pedometers
- G01C22/006—Pedometers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
- H04M1/724—User interfaces specially adapted for cordless or mobile telephones
- H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
- H04M1/72457—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions according to geographic location
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M2250/00—Details of telephonic subscriber devices
- H04M2250/12—Details of telephonic subscriber devices including a sensor for measuring a physical value, e.g. temperature or motion
Landscapes
- Engineering & Computer Science (AREA)
- Environmental & Geological Engineering (AREA)
- Human Computer Interaction (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
本发明公开了一种基于手机三轴加速度传感器的计步方法，该方法包含以下步骤：数据采集，根据计算窗口采集三轴加速度传感器的三轴数据；数据预处理，对数据进行平均值平滑，并求取合加速度；特征提取，分别提取数据的均值和方差，合加速度的波峰个数，利用聚类算法将合加速度聚类，提取聚类中心作为合加速度特征；位置判别，利用分类算法对特征进行分类得到类别标签；波形重构，对数据进行分割，重构成四元组，将数据分割成一个个完整波；步数计算，根据类别获取对应判别阈值，利用阈值对重构数据进行判别，当满足条件时对应类别步数加1。采用本发明的计步方法，可识别手机所处的不同位置，计步更加准确，具有较强的抗干扰性。
The invention discloses a step counting method based on a three-axis acceleration sensor of a mobile phone. The method comprises the following steps: data collection, collecting three-axis data of the three-axis acceleration sensor according to a calculation window; data preprocessing, performing average value smoothing on the data, And obtain the combined acceleration; feature extraction, extract the mean value and variance of the data, and the number of peaks of the combined acceleration, use the clustering algorithm to cluster the combined acceleration, and extract the cluster center as the combined acceleration feature; position discrimination, use the classification algorithm to Classification of features to obtain category labels; waveform reconstruction, segmentation of data, reconstruction into quadruples, and segmentation of data into complete waves; calculation of steps, obtaining corresponding discrimination thresholds according to categories, and using thresholds to discriminate reconstructed data , when the condition is met, the step number of the corresponding category is increased by 1. By adopting the step counting method of the invention, different positions of the mobile phone can be identified, the step counting is more accurate, and the step counting method has strong anti-interference performance.
Description
技术领域technical field
本发明涉及一种基于手机三轴加速度传感器的计步方法，属于消费应用电子技术领域。The invention relates to a step counting method based on a three-axis acceleration sensor of a mobile phone, belonging to the technical field of consumer application electronics.
背景技术Background technique
随着人民生活水平的提高，人们对如何进行合理的身体锻炼已经越来越重视，对身体状况进行及时监控，及时获取运动数据成为广大热爱运动的群体的基本需求。另一方面，在物联网的蓬勃发展的今天，越来越多的传感器设备被应用到可穿戴设备中。智能手机作为典型的智能可穿戴设备以及智能的监控设备走入了大众的生活中，在智能手机这个移动平台上出现了许多健康监控和运动指导的应用软件，这为获取健康信息提供了有力的物质和技术的支持。计步器就是相关应用的典型例子。With the improvement of people's living standards, people have paid more and more attention to how to carry out reasonable physical exercise. Timely monitoring of physical conditions and timely acquisition of exercise data have become the basic needs of the majority of sports-loving groups. On the other hand, with the vigorous development of the Internet of Things today, more and more sensor devices are applied to wearable devices. As a typical smart wearable device and smart monitoring device, smart phones have entered the lives of the public. Many health monitoring and exercise guidance applications have appeared on the mobile platform of smart phones, which provide a powerful way to obtain health information. Material and technical support. Pedometers are a typical example of related applications.
目前存在很多统计使用者行走步数的设备，但是它们都有一些缺陷。现有的大多通过加速度传感器获取加速度信号，然后设定简单的阈值判断合加速度中波峰值是否有效，从而实现计步功能。由于传感器采集过程中容易出现伪波峰，该方法可能出现误判，导致计步精度偏低。此外也有利用运动的周期性特点，判断前后两个波的合加速度波形相似度是否在某些阈值范围内等等。由于运动者的运动形式多种多样，比如跑步，行走，同时在运动的过程中，当采集装置放置在身体的不同位置时，加速度变化不同，波形的相似度可能不够高。因此，仅使用合加速度数据以及提取波形的周期性特性进行步数统计容易出现误计或漏计的情况，现有方法还有不足的地方。There are many devices for counting the number of steps taken by users at present, but they all have some defects. Most of the existing ones obtain the acceleration signal through the acceleration sensor, and then set a simple threshold to judge whether the peak value of the combined acceleration is valid, so as to realize the step counting function. Since false peaks are prone to appear during the sensor acquisition process, this method may cause misjudgment, resulting in low step counting accuracy. In addition, the periodic characteristics of motion are also used to judge whether the similarity of the combined acceleration waveform of the two waves before and after is within a certain threshold range and so on. Due to the various forms of exercise of the athlete, such as running and walking, and during the exercise, when the acquisition device is placed in different positions of the body, the acceleration changes differently, and the similarity of the waveform may not be high enough. Therefore, only using the combined acceleration data and extracting the periodic characteristics of the waveform to count the number of steps is prone to miscounting or missing counting, and the existing methods still have shortcomings.
发明内容Contents of the invention
本发明所要解决的技术问题是提供一种基于手机三轴加速度传感器的计步方法，该方法提取三轴加速度数据以及合加速度数据的特征，利用分类算法识别需要计步的数据，通过重构数据去除伪波峰以及伪波谷带来的影响，此外，识别手机在运动过程中所处在的不同位置，对不同位置的数据设置不同的阈值进行过滤，最终提高计步精准度，具有较高的抗干扰性。The technical problem to be solved by the present invention is to provide a method for counting steps based on the three-axis acceleration sensor of the mobile phone. Remove the influence of false peaks and false troughs. In addition, identify the different positions of the mobile phone during the movement, and set different thresholds to filter the data at different positions, and finally improve the accuracy of step counting, with high resistance intrusive.
本发明为解决上述技术问题采用以下技术方案The present invention adopts the following technical solutions to solve the above-mentioned technical problems
本发明提供一种基于手机三轴加速度传感器的计步方法，具体步骤如下：The present invention provides a step counting method based on a three-axis acceleration sensor of a mobile phone, and the specific steps are as follows:
步骤1，根据设定采样频率与采样时间窗口确定采样计算窗口，采集手机三轴加速度传感器的三轴加速度数据，每个采样计算窗口内的每一轴数据按照采集顺序编号为0,1,…,n-1，其中，n表示每个采样计算窗口内每一个轴的的数据个数；Step 1. Determine the sampling calculation window according to the set sampling frequency and sampling time window, and collect the three-axis acceleration data of the three-axis acceleration sensor of the mobile phone. The data of each axis in each sampling calculation window are numbered 0, 1,... ,n-1, where n represents the number of data of each axis in each sampling calculation window;
步骤2，在每个采样计算窗口内，对采集到的三轴加速度数据分别进行平滑处理，并计算平滑后的合加速度；Step 2, within each sampling calculation window, smooth the collected three-axis acceleration data respectively, and calculate the smoothed resultant acceleration;
步骤3，在每个采样计算窗口内，对步骤2中平滑后的三轴加速度数据以及合加速度进行特征提取，其中，提取x轴加速度的平均值xMean和方差xVariance作为x轴数据特征，提取y轴加速度的平均值yMean和方差yVariance作为y轴数据特征，提取z轴加速度的平均值zMean和方差zVariance作为z轴数据特征，提取合加速度波峰个数peakCount作为波峰的出现特征；利用聚类算法将合加速度数据聚成3类，分别对每类合加速度数据求取平均值，并将3个平均值按照由大到小进行排序作为合加速度特征<clusterPeak，clusterMean，clusterThrough>，clusterPeak>clusterMean>clusterThrough；Step 3, in each sampling calculation window, perform feature extraction on the smoothed three-axis acceleration data and resultant acceleration in step 2, where the mean xMean and variance xVariance of the x-axis acceleration are extracted as the x-axis data features, and y The average value yMean and variance yVariance of the axial acceleration are used as the data characteristics of the y-axis, the average value zMean and the variance zVariance of the z-axis acceleration are extracted as the data characteristics of the z-axis, and the peak count peakCount of the combined acceleration is extracted as the appearance characteristics of the peaks; The combined acceleration data is grouped into 3 categories, and the average value is calculated for each type of combined acceleration data, and the three average values are sorted from large to small as the combined acceleration feature <clusterPeak, clusterMean, clusterThrough>, clusterPeak>clusterMean>clusterThrough ;
步骤4，根据步骤3中提取到的特征，判断每个采样计算窗口内采集到的数据的类别；Step 4, according to the features extracted in step 3, judge the category of the data collected in each sampling calculation window;
步骤5，若步骤4中的判别结果为静止噪声或运动噪声，则舍弃当前采样计算窗口内的数据，否则，对当前采样计算窗口内的合加速度数据进行波形重构；其中，波形重构的方法具体如下：Step 5, if the judgment result in step 4 is static noise or motion noise, discard the data in the current sampling calculation window, otherwise, perform waveform reconstruction on the resultant acceleration data in the current sampling calculation window; where, the waveform reconstruction The method is as follows:
5.1，计算当前采样计算窗口内合加速度的平均值，利用平均值将当前采样计算窗口内的合加速度分割为多个波峰所在区域和波谷所在区域，其中，波峰所在区域与波谷所在区域交替出现；5.1, calculate the average value of the resultant acceleration in the current sampling calculation window, and use the average value to divide the resultant acceleration in the current sampling calculation window into areas where multiple peaks are located and areas where troughs are located, wherein the area where the peaks are located and the area where the troughs are located alternately appear;
5.2，搜索每一个波峰所在区域，获取当前波峰所在区域中最大的波峰，作为该区域的真波峰peak，并记录真波峰在当前采样计算窗口内对应数据编号peakIndex；5.2. Search the area where each peak is located, obtain the largest peak in the area where the current peak is located, and use it as the real peak peak in this area, and record the corresponding data number peakIndex of the real peak in the current sampling calculation window;
5.3，搜索每一个波谷所在的区域，获取每个波谷所在区域中最小的波谷，作为该区域的真波谷，同时记录真波谷值在当前采样计算窗口内对应数据编号；5.3. Search the area where each trough is located, and obtain the smallest trough in the area where each trough is located, as the real trough of the area, and record the corresponding data number of the true trough value in the current sampling calculation window;
5.4，真波峰peak与其前最后一个真波谷troughLeft、其后第一个真波谷troughRight以及半波长度halfWaveLength组成重构波形四元组<peak,troughLeft,troughRight,halfWaveLength>，完成一个波形的重构，halfWaveLength＝max{|peakIndex-troughIndexLeft|,|peakIndex-troughIndexRight|}，troughIndexLeft表示troughLeft在当前采样计算窗口内对应数据编号，troughIndexRight表示troughRight在当前采样计算窗口内对应数据编号；5.4, the real peak peak and the last real trough troughLeft before it, the first real trough troughRight after it, and the half-wave length halfWaveLength form a reconstructed waveform quadruple <peak, troughLeft, troughRight, halfWaveLength> to complete the reconstruction of a waveform. halfWaveLength＝max{|peakIndex-troughIndexLeft|,|peakIndex-troughIndexRight|}, troughIndexLeft indicates that troughLeft corresponds to the data number in the current sampling calculation window, and troughIndexRight indicates that troughRight corresponds to the data number in the current sampling calculation window;
步骤6，根据当前采样计算窗口内采集到数据的类别，设定阈值四元组<peakThreshold,troughThreshold,maxWaveLength,minWaveLength>，对步骤5中得到的重构波形进行逐一判决：Step 6, according to the category of data collected in the current sampling calculation window, set the threshold quadruple <peakThreshold, troughThreshold, maxWaveLength, minWaveLength>, and judge the reconstructed waveforms obtained in step 5 one by one:
6.1，根据当前采样计算窗口内采集到数据的类别，从预设阈值四元组中获取对应的阈值四元组<peakThreshold,troughThreshold,maxWaveLength,minWaveLength>，其中，peakThreshold作为真波峰的阈值，troughThreshold作为真波谷的阈值，maxWaveLength以及minWaveLength分别为halfWaveLength的最大可取值以及最小可取值；6.1. According to the category of data collected in the current sampling calculation window, obtain the corresponding threshold quadruple <peakThreshold, troughThreshold, maxWaveLength, minWaveLength> from the preset threshold quadruple, where peakThreshold is the threshold of the true peak, and troughThreshold is the The threshold of the true wave trough, maxWaveLength and minWaveLength are the maximum and minimum values of halfWaveLength respectively;
6.2，将重构波形的波形四元组<peak,troughLeft,troughRight,halfWaveLength>逐一与阈值四元组进行比较：首先，判断真波峰，若peak>peakThreshold则进入下一步，否则比较下一个重构波形的波形四元组；然后，判断真波谷，若troughLeft<troughThreshold且troughRight<troughThreshold则进入下一步，否则比较下一个重构波形的波形四元组；最后，判断半波长，若halfWaveLength>minWaveLength且halfWaveLength<maxWaveLength则当前重构波形计为一步，且对应数据类别的总步数加一；否则比较下一个重构波形的波形四元组；6.2, compare the waveform quadruple <peak, troughLeft, troughRight, halfWaveLength> of the reconstructed waveform with the threshold quadruple one by one: first, judge the true peak, if peak>peakThreshold, go to the next step, otherwise compare the next reconstruction The waveform quadruple of the waveform; then, judge the true valley, if troughLeft<troughThreshold and troughRight<troughThreshold, go to the next step, otherwise compare the waveform quadruple of the next reconstructed waveform; finally, judge the half-wavelength, if halfWaveLength>minWaveLength and If halfWaveLength<maxWaveLength, the current reconstructed waveform is counted as one step, and the total number of steps corresponding to the data category is increased by one; otherwise, compare the waveform quadruple of the next reconstructed waveform;
步骤7，各个采样计算窗口内的步数根据不同数据类别分别相加，即实现计步。In step 7, the number of steps in each sampling calculation window is added according to different data categories, that is, step counting is realized.
作为本发明的进一步优化方案，步骤2中采用平均值平滑法对三轴数据的每一轴数据分别进行平滑处理，平滑公式为：As a further optimization scheme of the present invention, in step 2, the average value smoothing method is used to smooth each axis data of the three-axis data respectively, and the smoothing formula is:
其中，s为预设的平滑窗口大小且为大于零的偶数，ai+j表示平滑前采样计算窗口内编号为i+j的数据，ai′表示平滑后采样计算窗口内编号为i的数据，n表示采样计算窗口内的数据个数。Among them, s is the preset smoothing window size and is an even number greater than zero, a i+j represents the data numbered i+j in the sampling calculation window before smoothing, and a i ' represents the data numbered i in the sampling calculation window after smoothing Data, n represents the number of data in the sampling calculation window.
作为本发明的进一步优化方案，步骤5中步骤5.1中利用平均值将当前采样时间窗口内的合加速度分割为波峰所在区域和波谷所在区域，具体为：通过判断合加速度是否大于平均值进行分割，当合加速度大于平均值时其所在区域为波峰所在区域，当合加速度值小于平均值时其所在区域为波谷所在区域。As a further optimization scheme of the present invention, in step 5, in step 5.1, the resultant velocity in the current sampling time window is divided into the area where the peak is located and the area where the trough is located, by using the average value in step 5.1, specifically: by judging whether the resultant velocity is greater than the average value for segmentation, When the resultant acceleration is greater than the average value, the area where it is located is the area where the peak is located, and when the resultant acceleration value is smaller than the average value, the area where it is located is the area where the wave trough is located.
作为本发明的进一步优化方案，步骤5.4中当前一个采样计算窗口内最后的合加速度数据中无法重构一个完整波形四元组时，将该部分合加速度数据添加到下一个采样计算窗口的前部，以保证相邻采样计算窗口中计算的连续性。As a further optimization scheme of the present invention, in step 5.4, when a complete waveform quadruple cannot be reconstructed in the last combined acceleration data in the previous sampling calculation window, add this part of the combined acceleration data to the front part of the next sampling calculation window , to ensure the continuity of calculations in adjacent sampling calculation windows.
作为本发明的进一步优化方案，采样计算窗口的长度windowLength＝采样频率f*采样时间窗口的长度N。As a further optimization solution of the present invention, the length of the sampling calculation window windowLength=sampling frequency f*length N of the sampling time window.
作为本发明的进一步优化方案，步骤4中每个采样计算窗口内采集到的数据的类别的判断方法为：As a further optimization scheme of the present invention, the judgment method of the category of the data collected in each sampling calculation window in step 4 is:
4.1，根据数据类别分别采集手机三轴加速度数据，并按照步骤1至3的方法进行特征提取，将提取到的特征作为训练集，其中，数据类别包括a)静止噪声、b)手机在上衣口袋、c)手机在裤子口袋、d)手机在手中时行走、e)手机在手中时跑步、f)手机在其他位置、g)运动噪声；4.1. Collect the three-axis acceleration data of the mobile phone according to the data category, and perform feature extraction according to the method of steps 1 to 3, and use the extracted features as the training set, where the data category includes a) static noise, b) the mobile phone in the jacket pocket , c) mobile phone in trouser pocket, d) walking with mobile phone in hand, e) running with mobile phone in hand, f) mobile phone in other position, g) motion noise;
4.2，构建分类模型，并利用步骤4.1中的训练集对分类模型进行训练学习；4.2, constructing a classification model, and using the training set in step 4.1 to train and learn the classification model;
4.3，将每个采样计算窗口内采集到的数据提取到的特征输入训练学习后的分类模型，分类模型的输出即为其对应的类别。4.3. Input the feature extracted from the data collected in each sampling calculation window into the trained classification model, and the output of the classification model is its corresponding category.
作为本发明的进一步优化方案，步骤4.2中的分类模型为三层神经网络，其中，第一层和第二层的激活函数均为带泄露线性整流函数，第三层采用归一化指数函数softmax函数，采用带动量的随机梯度下降算法对三层神经网络进行训练，损失函数为交叉熵损失函数。As a further optimization scheme of the present invention, the classification model in step 4.2 is a three-layer neural network, wherein the activation functions of the first layer and the second layer are both linear rectification functions with leakage, and the third layer uses a normalized exponential function softmax The function uses the stochastic gradient descent algorithm with momentum to train the three-layer neural network, and the loss function is the cross-entropy loss function.
本发明采用以上技术方案与现有技术相比，具有以下技术效果：Compared with the prior art, the present invention adopts the above technical scheme and has the following technical effects:
1、本发明方法采集三轴加速度传感器的三轴数据，提取数据特征，借助分类算法进行手机位置状态的判别，可以识别手机在运动过程中所处的不同位置，重构数据识别出完整波，对不同的位置类别分别进行计步，使得计步更加精准；1. The method of the present invention collects the three-axis data of the three-axis acceleration sensor, extracts the data features, and uses the classification algorithm to discriminate the position state of the mobile phone, which can identify the different positions of the mobile phone during the movement process, and reconstruct the data to identify the complete wave. Count steps for different location categories to make step counting more accurate;
2、本发明方法识别手机在运动过程中所处的不同位置，可以对不同的位置类别进行识别并进行计步，满足使用者对不同应用场景的需求；2. The method of the present invention identifies the different positions of the mobile phone during the movement process, and can identify different position categories and perform step counting to meet the needs of users for different application scenarios;
3、本发明方法通过重构数据识别出真正的波峰以及真正的波谷，去除伪波峰以及伪波谷的影响，具有较强的抗干扰性；3. The method of the present invention identifies real peaks and real troughs by reconstructing data, removes the influence of false peaks and false troughs, and has strong anti-interference;
4、本发明方法基于手机平台，可以利用随身携带的手机作为数据采集、计算、存储和展示设备，普通智能手机中都有三轴加速度传感器，实用性强，具有很强的可移植性。4. The method of the present invention is based on the mobile phone platform, and the portable mobile phone can be used as a data acquisition, calculation, storage and display device. Common smart phones have three-axis acceleration sensors, which are practical and have strong portability.
附图说明Description of drawings
图1为本发明实施例的系统流程图；Fig. 1 is the system flowchart of the embodiment of the present invention;
图2为本发明实施例的三轴加速度示意图；Fig. 2 is a schematic diagram of triaxial acceleration according to an embodiment of the present invention;
图3为本发明实施例的波形重构流程图；Fig. 3 is the waveform reconstruction flowchart of the embodiment of the present invention;
图4为本发明实施例的波谷解析流程图；Fig. 4 is the trough analysis flowchart of the embodiment of the present invention;
图5为本发明实施例的步数计算流程图。FIG. 5 is a flow chart of step count calculation in an embodiment of the present invention.
具体实施方式detailed description
下面结合附图对本发明的技术方案做进一步的详细说明：Below in conjunction with accompanying drawing, technical scheme of the present invention is described in further detail:
如图1所示，给出了本发明实施例的系统流程，包括以下步骤：As shown in Figure 1, the system flow of the embodiment of the present invention is provided, including the following steps:
101、数据采集：包括设置手机三轴加速度传感器的采样频率以及采样时间窗口，计算采样计算窗口，并存储采集到的三轴加速度的值。每个采样计算窗口内的每一个轴的数据按照采集顺序编号为0,1,…,n-1，其中，n表示每个采样计算窗口内每一个轴的数据个数。101. Data acquisition: including setting the sampling frequency and sampling time window of the three-axis acceleration sensor of the mobile phone, calculating the sampling calculation window, and storing the collected values of the three-axis acceleration. The data of each axis in each sampling calculation window is numbered 0, 1, ..., n-1 according to the collection sequence, where n represents the number of data of each axis in each sampling calculation window.
计算采样计算窗口，公式为：windowLength＝f*N。其中，windowLength表示采样计算窗口的总长度，单位为个；f表示采样频率，单位为赫兹；N表示采样的时间长度，单位为秒；举例来说，当f＝50赫兹，N＝4时，得到windowLength＝4*50＝200个，此时，当加速度传感器在x轴，y轴以及z轴分别采集到200个对应方向上的数据时，当前计算窗口关闭，将数据传递到下一步骤，开启新的计算窗口接受新数据。Calculate the sampling calculation window, the formula is: windowLength=f*N. Among them, windowLength represents the total length of the sampling calculation window, in unit; f represents the sampling frequency, in hertz; N represents the sampling time length, in seconds; for example, when f=50 Hz, N=4, Get windowLength=4*50=200. At this time, when the acceleration sensor collects data in 200 corresponding directions on the x-axis, y-axis and z-axis respectively, the current calculation window is closed, and the data is passed to the next step. Open a new calculation window to accept new data.
所述的三轴加速度传感器的三轴如图2所示，分别记为x轴、y轴以及z轴。以屏幕朝上水平放置时的状态为例，所述x轴采集加速度传感器左右运动时该方向加速度变化情况；所述y轴采集加速度传感器前后运动时该方向上加速度变化情况；所述z轴采集加速度传感器上下运动时该方向上加速度变化情况。The three axes of the three-axis acceleration sensor are shown in FIG. 2 , which are denoted as x-axis, y-axis and z-axis respectively. Taking the state when the screen is placed horizontally upwards as an example, the x-axis collects acceleration changes in this direction when the acceleration sensor moves left and right; the y-axis collects acceleration changes in this direction when the acceleration sensor moves back and forth; the z-axis collects Acceleration changes in this direction when the acceleration sensor moves up and down.
102、数据预处理：包括采用平均值平滑法对101中采集得到数据进行平滑处理。根据预设的平滑窗口系数s对三轴加速度中的x轴加速度、y轴加速度以及z轴加速度分别进行平滑，平滑公式如下：102. Data preprocessing: including smoothing the data collected in step 101 using the average smoothing method. Smooth the x-axis acceleration, y-axis acceleration, and z-axis acceleration in the three-axis acceleration according to the preset smoothing window coefficient s. The smoothing formula is as follows:
其中，s为预设的平滑窗口大小且为大于零的偶数，ai+j表示平滑前采样计算窗口内编号为i+j的数据，ai′表示平滑后采样计算窗口内编号为i的数据，n表示采样计算窗口内的数据个数。Among them, s is the preset smoothing window size and is an even number greater than zero, a i+j represents the data numbered i+j in the sampling calculation window before smoothing, and a i ' represents the data numbered i in the sampling calculation window after smoothing Data, n represents the number of data in the sampling calculation window.
将平滑后的x轴加速度、y轴加速度以及z轴加速度计算合加速度值，公式如下：Calculate the combined acceleration value from the smoothed x-axis acceleration, y-axis acceleration, and z-axis acceleration. The formula is as follows:
其中，xf、yf、zf分别为平滑后的x轴加速度、y轴加速度以及z轴加速度。Wherein, x f , y f , and z f are the smoothed x-axis acceleration, y-axis acceleration, and z-axis acceleration, respectively.
103、特征提取：提取当前计算窗口内加速度的特征，包括：分别提取三轴加速度传感器中x轴，y轴以及z轴加速度值的特征；提取三轴合加速度值的特征，提取波峰的出现特征。103. Feature extraction: extracting the characteristics of the acceleration in the current calculation window, including: extracting the characteristics of the x-axis, y-axis and z-axis acceleration values in the three-axis acceleration sensor respectively; extracting the characteristics of the three-axis combined acceleration value and extracting the appearance characteristics of the peak .
分别提取三轴加速度传感器x轴，y轴以及z轴加速度的特征，包括：计算窗口内提取x轴加速度值的平均值xMean以及方差xVariance，分别作为当前计算窗口内x轴方向上运动的大小特征以及混乱程度；类似的，提取y轴加速度值的平均值yMean以及方差yVariance，提取z轴加速度值的平均值zMean以及方差zVariance。提取当前计算窗口内合加速度波峰个数作为波峰的出现特征，记为peakCount；同时，提取三轴合加速度值的特征。Extract the characteristics of the x-axis, y-axis and z-axis acceleration of the three-axis acceleration sensor respectively, including: extracting the average xMean and variance xVariance of the x-axis acceleration values in the calculation window, respectively, as the size characteristics of the movement in the x-axis direction in the current calculation window And the degree of confusion; similarly, extract the mean yMean and variance yVariance of the y-axis acceleration values, and extract the mean zMean and variance zVariance of the z-axis acceleration values. Extract the number of combined acceleration peaks in the current calculation window as the appearance feature of the peak, which is recorded as peakCount; at the same time, extract the characteristics of the three-axis combined acceleration value.
提取三轴合加速度值的特征，包括：利用预先设置聚类个数的聚类算法将合加速度数据聚成3类，分别将属于同一类的数据求取平均合加速度值，对得到的3个平均值进行从大到小排序，得到3个聚类中心<clusterPeak，clusterMean，clusterThrough>作为合加速度特征，clusterPeak表示波峰的均值特征，clusterMean表示平均值附近的特征，clusterThrough表示波谷的均值特征。Extract the characteristics of the three-axis combined acceleration value, including: clustering the combined acceleration data into three categories by using a clustering algorithm with a preset number of clusters, calculating the average combined acceleration value for the data belonging to the same category, and calculating The average value is sorted from large to small, and the three cluster centers <clusterPeak, clusterMean, clusterThrough> are obtained as the combined velocity feature, clusterPeak represents the mean value feature of the peak, clusterMean represents the feature near the mean value, and clusterThrough represents the mean value feature of the trough.
提取波峰的出现特征peakCount，计算当前计算窗口内合加速度值中波峰的出现次数。其中的波峰满足如下公式：Extract the appearance feature peakCount of the peak, and calculate the number of occurrences of the peak in the combined acceleration value in the current calculation window. The peaks satisfy the following formula:
ah-1<ah a h-1 <a h
ah+1<ah a h+1 <a h
其中，ah表示采样计算窗口内数据编号为h的合加速度值，ah-1表示采样计算窗口内数据编号为h-1的合加速度值，ah+1表示采样计算窗口内数据编号为h+1的合加速度值。Among them, a h represents the resultant acceleration value with data number h in the sampling calculation window, a h-1 represents the resultant acceleration value with data number h-1 in the sampling calculation window, and a h+1 represents the data number in the sampling calculation window as The resultant acceleration value of h+1.
以上构成特征值组The above constitutes the eigenvalue group
<xMean,xVariance,yMean,yVariance,zMean,zVariance,peakCount,clusterPeak,clusterMean,clusterThrough>。<xMean,xVariance,yMean,yVariance,zMean,zVariance,peakCount,clusterPeak,clusterMean,clusterThrough>.
104、数据类别的判别：根据提取的x轴的特征，y轴的特征，z轴的特征，合加速度的特征以及波峰的出现特征组成当前采集的计算窗口的特征值组，利用分类算法进行对采集到的数据的类别进行判别。其中，数据的类别包括：a)静止噪声、b)手机在上衣口袋、c)手机在裤子口袋、d)手机在手中时行走、e)手机在手中时跑步、f)手机在其他位置、g)运动噪声，b到f这5种类别表示手机的位置，f是当手机放置于包内等其他位置时的统称。104. Discrimination of data categories: According to the extracted features of the x-axis, y-axis, z-axis, combined acceleration and peak appearance, the feature value group of the currently collected calculation window is formed, and the classification algorithm is used for classification Classification of the collected data. Among them, the categories of data include: a) static noise, b) mobile phone in jacket pocket, c) mobile phone in trouser pocket, d) walking with mobile phone in hand, e) running with mobile phone in hand, f) mobile phone in other locations, g ) motion noise, the five categories b to f indicate the position of the mobile phone, and f is a general term for when the mobile phone is placed in other places such as a bag.
所述的分类算法，在本发明中使用的是多层神经网络，网络分为3层，第一层和第二层的激活函数为带泄露线性整流函数，第三层采用归一化指数函数softmax函数。在其训练学习的过程中，首先根据不同的类别分别采集相应的三轴数据，进行101、102、103的处理提取特征，然后进行类别标记，即按照a到g的顺序根据所处类别进行独热码编码得到类别标签，将特征数据与对应的类别标签输入到多层神经网络中，采用带动量的随机梯度下降算法进行训练，损失函数为交叉熵损失函数。The classification algorithm used in the present invention is a multilayer neural network, the network is divided into 3 layers, the activation functions of the first layer and the second layer are linear rectification functions with leakage, and the third layer adopts a normalized exponential function softmax function. In the process of training and learning, firstly collect the corresponding three-axis data according to different categories, perform 101, 102, 103 processing to extract features, and then perform category marking, that is, according to the order of a to g, according to the category in which it is located. The category label is obtained by hot code encoding, the feature data and the corresponding category label are input into the multi-layer neural network, and the stochastic gradient descent algorithm with momentum is used for training, and the loss function is the cross-entropy loss function.
105、波形重构：根据104中的类别判别结果，当判别结果为a或者g时，表示当前采样计算窗口内的数据为无效数据，舍去当前采样计算窗口内的数据。否则，重构当前采样计算窗口内的数据，计算当前计算窗口内合加速度值的平均值，利用该平均值解析出真波峰和真波谷，重构得到波形四元组，将计算窗口内的数据解析成多个四元组的集合，即可得到重构的多个完整的波形四元组。105. Waveform reconstruction: According to the category discrimination result in 104, when the discrimination result is a or g, it means that the data in the current sampling calculation window is invalid data, and discard the data in the current sampling calculation window. Otherwise, reconstruct the data in the current sampling calculation window, calculate the average value of the combined acceleration value in the current calculation window, use the average value to analyze the true peak and true wave trough, reconstruct the waveform quadruple, and calculate the data in the window By parsing it into a set of multiple quadruples, multiple complete waveform quadruples can be reconstructed.
重构获取波形四元组：计算当前计算窗口内合加速度值的平均值，利用平均值分割出波峰所在区域以及波谷所在区域，当合加速度值大于平均值时，当前所在区域为需要搜索的波峰区域，当合加速度值小于平均值时，当前所在区域为需要搜索的波谷区域；搜索波峰所在区域，获取该波峰区域中最大的波峰，作为当前区域的真正波峰peak，并记录真波峰所在的数据编号peakIndex；搜索波谷所在的区域，获取该波谷区域最小的波谷，作为当前区域的真正波谷。Reconstruct and obtain the waveform quadruple: Calculate the average value of the combined acceleration value in the current calculation window, and use the average value to divide the area where the peak is located and the area where the trough is located. When the combined acceleration value is greater than the average value, the current area is the peak that needs to be searched Area, when the combined acceleration value is less than the average value, the current area is the valley area to be searched; search the area where the peak is located, obtain the largest peak in the peak area, as the real peak peak of the current area, and record the data where the true peak is located Number peakIndex; search the area where the trough is located, and obtain the smallest trough in the trough area as the real trough of the current area.
一个完整波由三个区域构成，分别是左波谷区域、波峰区域以及右波谷区域。波峰区域与波谷区域交替出现，从左波谷区域穿越平均值到达波峰区域，波峰区域穿越平均值到达右波谷区域，一个完整波两次穿越平均值。A complete wave consists of three regions, namely the left trough region, the crest region and the right trough region. The peak area and the trough area appear alternately, from the left trough area crossing the mean to reach the peak area, the peak area crossing the mean to the right trough area, a complete wave crosses the mean twice.
为了去除伪波峰的影响，定义一个波的真波峰为：在两次穿越平均值中，一个波峰区域内，众多波峰中波峰最大的值，波峰满足如下公式：In order to remove the influence of false peaks, the real peak of a wave is defined as: in the average value of two crossings, within a peak area, the peak value among many peaks is the largest, and the peak satisfies the following formula:
ah-1<ah a h-1 <a h
ah+1<ah a h+1 <a h
类似的，定义一个真波谷为：在两次穿越平均值中，一个波谷区域内，众多波谷中取波谷最小的值，波谷满足以下公式：Similarly, a true trough is defined as: In the average value of two crossings, within a trough area, the minimum value of the trough is taken among many troughs, and the trough satisfies the following formula:
ak-1>ak a k-1 >a k
ak+1>ak a k+1 >a k
其中，ak表示采样计算窗口内数据编号为k的合加速度值，ak-1表示采样计算窗口内数据编号为k-1的合加速度值，ak+1表示采样计算窗口内数据编号为k+1的合加速度值。Among them, a k represents the resultant acceleration value with data number k in the sampling calculation window, a k-1 represents the resultant acceleration value with data number k-1 in the sampling calculation window, and a k+1 represents the data number in the sampling calculation window as The resultant acceleration value of k+1.
具体操作包括：在一个波峰搜索区域中，获取波峰的最大值，记录真波峰所在的数据编号，计为peakIndex；类似的，在一个波谷搜索区域中，获取最小的一个波谷，记录真波谷的所在的数据编号，最终得到一个四元组<peak,troughLeft,troughRight,halfWaveLength>，其中各变量分别为：真波峰peak，遇到真波峰前的最后一个真波谷troughLeft，真波峰过后的第一个真波谷troughRight，halfWaveLength表示半波的长度。The specific operations include: in a peak search area, obtain the maximum value of the peak, record the data number where the true peak is located, and count it as peakIndex; similarly, in a valley search area, obtain the smallest valley, and record the location of the true valley Finally, a quaternion <peak, troughLeft, troughRight, halfWaveLength> is obtained, in which the variables are: true peak peak, the last true wave troughLeft before encountering the true peak, the first true wave after the true wave peak The trough troughRight, halfWaveLength indicates the length of the half wave.
halfWaveLength的计算公式为：The calculation formula of halfWaveLength is:
halfWaveLength＝max{|peakIndex-troughIndexLeft|,|peakIndex-troughIndexRight|}halfWaveLength＝max{|peakIndex-troughIndexLeft|,|peakIndex-troughIndexRight|}
其中，troughIndexLeft表示troughLeft所在的数据编号，troughIndexRight表示troughRight所在的数据编号。Among them, troughIndexLeft indicates the data number where troughLeft is located, and troughIndexRight indicates the data number where troughRight is located.
以上完成一个波的重构，在一个采样计算窗口中，可以重构得到多个重构波形四元组。为了保证计算窗口内相邻完整波数据的连续性，一个四元组的troughRight为下一个四元组的troughLeft。The reconstruction of one wave is completed above, and multiple reconstructed waveform quadruples can be reconstructed in one sampling calculation window. In order to ensure the continuity of adjacent complete wave data in the calculation window, the troughRight of a quadruple is the troughLeft of the next quadruple.
完整波的搜索从真波谷开始，当无法完整地得到上面的四元组时，认为当前波不完整，需要保存当前残波数据，等待下一个计算窗口的数据到来，将合加速度数据中，troughLeft所在区域及其以后的数据添加到下一个计算窗口数据的前部，继续搜索真波峰以及真波谷。当得到一个完整波的波形四元组时，进行下一步骤的计算。The search of the complete wave starts from the real trough. When the above quadruple cannot be obtained completely, the current wave is considered to be incomplete, and the current residual wave data needs to be saved. Waiting for the arrival of the data in the next calculation window, the combined acceleration data, troughLeft The data in the area and beyond are added to the front of the data in the next calculation window, and the search for true peaks and true troughs is continued. When the waveform quadruple of a complete wave is obtained, the calculation of the next step is performed.
如图3所示为步骤105进行波形重构的具体流程：As shown in Figure 3, the specific process of performing waveform reconstruction in step 105:
301、判断合加速度值是否大于均值。如果大于均值，表示进入波峰搜索区域，则开始寻找波形四元组<peak,troughLeft,troughRight,halfWaveLength>中的真波峰信息；否则，开始寻找真波谷信息。301. Determine whether the resultant acceleration value is greater than the average value. If it is greater than the average value, it means entering the peak search area, then start to search for the true peak information in the waveform quadruple <peak,troughLeft,troughRight,halfWaveLength>; otherwise, start to search for the true valley information.
302、判断当前值的前一个值是否小于均值。如果是进行信息保存；否则，表示上一个波的信息已经保存好了，无需在进行信息保存。302. Determine whether the previous value of the current value is smaller than the mean value. If it is to save information; otherwise, it means that the information of the previous wave has been saved, and there is no need to save information.
303、信息保存。当小于均值时，说明当前正在穿越均值，到达波峰所在的区域。此时，将真波谷信息保存为上一个波的四元组信息中的真波谷值troughRight以及数据编号troughIndexRight，计算上一个波中的halfWaveLength，公式为：303. Information preservation. When it is less than the mean value, it means that it is currently crossing the mean value and reaching the area where the peak is located. At this time, save the real trough information as the real trough value troughRight and the data number troughIndexRight in the quadruple information of the previous wave, and calculate the halfWaveLength in the previous wave. The formula is:
max{|peakIndex-troughIndexLeft|,|peakIndex-troughIndexRight|}max{|peakIndex-troughIndexLeft|,|peakIndex-troughIndexRight|}
以上完成上一个波的特征采集。同时，将波谷信息保存为当前波的真左波谷值troughLeft，记录数据编号troughIndexLeft，将原有波谷区域清空。The feature collection of the previous wave is completed above. At the same time, save the trough information as the true left trough value troughLeft of the current wave, record the data number troughIndexLeft, and clear the original trough area.
304、判断当前合加速度值是否为波峰。如果不是波峰，舍去数据，读取下一个合加速度值。304. Determine whether the current combined acceleration value is a peak. If it is not a wave peak, discard the data and read the next resultant acceleration value.
305、当前合加速度为波峰值，需要判断当前合加速度值是否为波峰域中最大值。将当前合加速度值与已有的波峰域中的波峰进行判断，查找最大值。如果不是最大值，舍去数据，读取下一个合加速度值。305. The current combined acceleration is the peak value, and it is necessary to judge whether the current combined acceleration value is the maximum value in the peak domain. Judgment is made between the current resultant velocity value and the peak in the existing peak domain to find the maximum value. If it is not the maximum value, discard the data and read the next combined acceleration value.
306、当前合加速度为波形四元组所潜在的真波峰peak值，保存peakIndex，并将当前波峰加入波峰域中。306. The current combined acceleration is the potential true peak peak value of the waveform quadruple, save the peakIndex, and add the current peak to the peak domain.
307、当前合加速度值不大于均值，是波谷所在的区域，进行真波谷相关特征的提取。307. The current resultant acceleration value is not greater than the average value, which is the area where the trough is located, and the relevant features of the true trough are extracted.
如图4所示为步骤307进行波谷解析的具体流程：As shown in Figure 4, the specific process of performing valley analysis in step 307:
401、判断是否为当前波中第一次进入波谷所在区域，如果是，那么需要保存当前波的真波峰信息。否则，表示当前波的真波峰信息已经保存，直接进行波谷搜索。401. Judging whether it is the first time in the current wave to enter the area where the trough is located, and if yes, it is necessary to save the real peak information of the current wave. Otherwise, it means that the real peak information of the current wave has been saved, and the valley search is performed directly.
402、当前是从真波峰穿越平均值进入到波谷所在的波段，需要将当前完整波的真波峰特征信息保存到四元组中，即保存真波峰的值peak，以及真波峰所在的数据编号peakIndex，同时清空波峰区域。402. Currently, the true peak crosses the average value and enters the wave band where the trough is located. It is necessary to save the true peak feature information of the current complete wave into a quadruple, that is, save the value peak of the true peak and the data number peakIndex where the true peak is located. , while clearing the crest area.
403、判断当前合加速度值是否为波谷，如果不是，则输入下一个合加速度值。403. Determine whether the current combined acceleration value is a valley, and if not, input the next combined acceleration value.
404、判断当前的波谷是否为波谷域中的最小值，将当前的合加速度值与域中的已有波谷值进行对比，选出最小值作为最终候选的真波谷值。404. Determine whether the current valley is the minimum value in the valley domain, compare the current resultant acceleration value with existing valley values in the domain, and select the minimum value as the final candidate true valley value.
405、保存当前得到的候选的最小波谷值，将波谷的信息进行保存，包括波谷值以及波谷对应的数据编号，将波谷值加入到波谷域当中。输入下一个合加速度值进行判断。405. Save the currently obtained candidate minimum valley value, save the valley information, including the valley value and the data number corresponding to the valley, and add the valley value to the valley domain. Input the next resultant acceleration value for judgment.
106、步数计算：根据105波形重构得到的四元组数据，以及104中判别所得到的数据类别信息，利用每种类别的对应的阈值四元组进行判别，当满足阈值四元组的要求，则在对应类别的总计步数加1。106. Calculation of the number of steps: According to the quadruple data obtained by waveform reconstruction in 105, and the data category information obtained by discrimination in 104, use the corresponding threshold quadruples of each category for discrimination, and when the threshold quadruples are satisfied If required, add 1 to the total number of steps in the corresponding category.
根据步骤104得到的数据类别信息，获取相应类别的阈值四元组<peakThreshold,troughThreshold,maxWaveLength,minWaveLength>，将所得到的波形四元组特征<peak,troughLeft,troughRight,halfWaveLength>与通过阈值四元组进行比较，筛选出符合要求的四元组，进行步数计算。According to the data category information obtained in step 104, the threshold value quadruple <peakThreshold, troughThreshold, maxWaveLength, minWaveLength> of the corresponding category is obtained, and the obtained waveform quadruple feature <peak, troughLeft, troughRight, halfWaveLength> is combined with the threshold value quadruple Groups are compared, the quadruples that meet the requirements are screened out, and the number of steps is calculated.
如图5所示为步骤106进行步数计算的具体流程：As shown in Figure 5, the specific process of step calculation in step 106 is as follows:
501、步骤104获取数据类别type。501. Step 104 acquires the data category type.
502、根据获取的类别type，当类别对应为静止噪声或者运动噪声时，舍去。否则，取出对应的阈值四元组<peakThreshold,troughThreshold,maxWaveLength,minWaveLength>。使用阈值对重构得到的四元组数据进行判别。502. According to the acquired category type, when the category corresponds to static noise or motion noise, discard it. Otherwise, take out the corresponding threshold quadruple <peakThreshold, troughThreshold, maxWaveLength, minWaveLength>. Use a threshold to discriminate the reconstructed quadruple data.
503、判别真波峰值是否在预定范围内，如果peak<peakThreshold，那么舍去当前重构数据，判断下一个重构波形四元组。503. Determine whether the peak value of the true wave is within a predetermined range, and if peak<peakThreshold, discard the current reconstructed data, and determine the next reconstructed waveform quadruplet.
504、判断真左波谷是否在预定的范围内，如果troughLeft>troughThreshold，那么当前重构数据不是有效的数据，舍去。504. Determine whether the true left trough is within a predetermined range, if troughLeft>troughThreshold, then the current reconstructed data is not valid data, discard it.
505、判断真右波谷是否在预定的范围内，如果troughRight>troughThreshold，那么表示当前数据为无效数据，舍去。505. Determine whether the true right trough is within a predetermined range. If troughRight>troughThreshold, it means that the current data is invalid data and discarded.
506、判断半波长是否在设定的范围内，如果halfWaveLength<minWaveLength，那么当前数据为无效数据，舍去。506 . Determine whether the half-wavelength is within the set range. If halfWaveLength<minWaveLength, then the current data is invalid data and discarded.
507、判断半波长的是否超出预先设定的范围，如果halfWaveLength>maxWaveLength，那么当前数据不符合要求，舍去。507. Determine whether the half-wavelength exceeds the preset range, if halfWaveLength>maxWaveLength, then the current data does not meet the requirements, and discard it.
508、当前重构四元组数据为有效数据，属于正确的计步数据，将type类型的数据的计步总数加上一步。508. The currently reconstructed quadruple data is valid data and belongs to correct step counting data, and one step is added to the total number of steps counting of the data of type type.
以上所述，仅为本发明中的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉该技术的人在本发明所揭露的技术范围内，可理解想到的变换或替换，都应涵盖在本发明的包含范围之内，因此，本发明的保护范围应该以权利要求书的保护范围为准。The above is only a specific implementation mode in the present invention, but the scope of protection of the present invention is not limited thereto. Anyone familiar with the technology can understand the conceivable transformation or replacement within the technical scope disclosed in the present invention. All should be covered within the scope of the present invention, therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims (7)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710569013.4A CN107462258B (en) | 2017-07-13 | 2017-07-13 | Step counting method based on mobile phone three-axis acceleration sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710569013.4A CN107462258B (en) | 2017-07-13 | 2017-07-13 | Step counting method based on mobile phone three-axis acceleration sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107462258A true CN107462258A (en) | 2017-12-12 |
| CN107462258B CN107462258B (en) | 2020-06-16 |
Family
ID=60546251
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710569013.4A Expired - Fee Related CN107462258B (en) | 2017-07-13 | 2017-07-13 | Step counting method based on mobile phone three-axis acceleration sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107462258B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109190553A (en) * | 2018-08-30 | 2019-01-11 | 上海奥孛睿斯科技有限公司 | Step detection method |
| CN109350072A (en) * | 2018-11-15 | 2019-02-19 | 北京航空航天大学 | A cadence detection method based on artificial neural network |
| CN109540143A (en) * | 2018-11-27 | 2019-03-29 | 中国广核电力股份有限公司 | The unconventional direction of action recognition methods of pedestrian of more sensing sources dynamic peak value fusions |
| CN110595501A (en) * | 2019-10-09 | 2019-12-20 | 成都乐动信息技术有限公司 | Running distance correction method based on three-axis sensor |
| CN110721456A (en) * | 2019-10-09 | 2020-01-24 | 成都乐动信息技术有限公司 | Pedal frequency detection method based on three-axis sensor |
| CN110852260A (en) * | 2019-11-08 | 2020-02-28 | 青岛联合创智科技有限公司 | Drowning behavior identification method based on accelerometer |
| CN111238527A (en) * | 2020-01-15 | 2020-06-05 | 桂林市优创电子科技有限公司 | Step counting method based on three-axis acceleration sensor |
| CN111358472A (en) * | 2020-03-27 | 2020-07-03 | 山东建筑大学 | Gait detection algorithm, device and system |
| CN111442785A (en) * | 2020-03-27 | 2020-07-24 | 广东工业大学 | Inertial-based pedometer method, device and storage medium |
| CN112067015A (en) * | 2020-09-03 | 2020-12-11 | 青岛歌尔智能传感器有限公司 | Step counting method and device based on convolutional neural network and readable storage medium |
| CN112923922A (en) * | 2021-03-04 | 2021-06-08 | 香港理工大学深圳研究院 | Method, system and storage medium for counting steps and determining position information of pedestrian |
| CN113008231A (en) * | 2021-04-30 | 2021-06-22 | 东莞市小精灵教育软件有限公司 | Motion state identification method and system, wearable device and storage medium |
| CN113378801A (en) * | 2021-08-11 | 2021-09-10 | 深圳市梦瑶科技有限公司 | Intelligent bracelet accurate step counting method based on neural network |
| CN114034313A (en) * | 2021-11-10 | 2022-02-11 | 广东乐心医疗电子股份有限公司 | Step counting method and device |
| CN114139669A (en) * | 2021-11-19 | 2022-03-04 | 歌尔科技有限公司 | Rope skipping counting method, device and medium thereof |
| CN114739412A (en) * | 2022-04-11 | 2022-07-12 | 河海大学 | Pedestrian gait real-time detection method and device based on smart phone |
| CN114815695A (en) * | 2022-04-28 | 2022-07-29 | 广东万颗子智控科技有限公司 | Remote control instruction generation method and device, remote control equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6441846B1 (en) * | 1998-06-22 | 2002-08-27 | Lucent Technologies Inc. | Method and apparatus for deriving novel sports statistics from real time tracking of sporting events |
| CN104567912A (en) * | 2015-02-02 | 2015-04-29 | 河海大学 | Method for realizing pedometer on Android mobile phone |
| CN104983489A (en) * | 2015-07-28 | 2015-10-21 | 河北工业大学 | Road condition identifying method for lower limb prosthesis walking |
| CN104990562A (en) * | 2015-06-29 | 2015-10-21 | 合肥工业大学 | Step counting method based on autocorrecting computing |
| CN106767888A (en) * | 2016-11-15 | 2017-05-31 | 皖西学院 | A kind of meter based on Wave crest and wave trough detection walks algorithm |
-
2017
- 2017-07-13 CN CN201710569013.4A patent/CN107462258B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6441846B1 (en) * | 1998-06-22 | 2002-08-27 | Lucent Technologies Inc. | Method and apparatus for deriving novel sports statistics from real time tracking of sporting events |
| CN104567912A (en) * | 2015-02-02 | 2015-04-29 | 河海大学 | Method for realizing pedometer on Android mobile phone |
| CN104990562A (en) * | 2015-06-29 | 2015-10-21 | 合肥工业大学 | Step counting method based on autocorrecting computing |
| CN104983489A (en) * | 2015-07-28 | 2015-10-21 | 河北工业大学 | Road condition identifying method for lower limb prosthesis walking |
| CN106767888A (en) * | 2016-11-15 | 2017-05-31 | 皖西学院 | A kind of meter based on Wave crest and wave trough detection walks algorithm |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109190553A (en) * | 2018-08-30 | 2019-01-11 | 上海奥孛睿斯科技有限公司 | Step detection method |
| CN109350072A (en) * | 2018-11-15 | 2019-02-19 | 北京航空航天大学 | A cadence detection method based on artificial neural network |
| CN109540143A (en) * | 2018-11-27 | 2019-03-29 | 中国广核电力股份有限公司 | The unconventional direction of action recognition methods of pedestrian of more sensing sources dynamic peak value fusions |
| CN109540143B (en) * | 2018-11-27 | 2022-12-02 | 中国广核电力股份有限公司 | Pedestrian unconventional action direction identification method based on multi-sensing-source dynamic peak fusion |
| CN110595501A (en) * | 2019-10-09 | 2019-12-20 | 成都乐动信息技术有限公司 | Running distance correction method based on three-axis sensor |
| CN110721456A (en) * | 2019-10-09 | 2020-01-24 | 成都乐动信息技术有限公司 | Pedal frequency detection method based on three-axis sensor |
| CN110595501B (en) * | 2019-10-09 | 2022-10-04 | 成都乐动信息技术有限公司 | Running distance correction method based on three-axis sensor |
| CN110721456B (en) * | 2019-10-09 | 2020-12-25 | 成都乐动信息技术有限公司 | Pedal frequency detection method based on three-axis sensor |
| CN110852260A (en) * | 2019-11-08 | 2020-02-28 | 青岛联合创智科技有限公司 | Drowning behavior identification method based on accelerometer |
| CN110852260B (en) * | 2019-11-08 | 2023-05-30 | 青岛联合创智科技有限公司 | Drowning behavior recognition method based on accelerometer |
| CN111238527B (en) * | 2020-01-15 | 2023-09-29 | 桂林市优创电子科技有限公司 | Step counting method based on triaxial acceleration sensor |
| CN111238527A (en) * | 2020-01-15 | 2020-06-05 | 桂林市优创电子科技有限公司 | Step counting method based on three-axis acceleration sensor |
| CN111442785A (en) * | 2020-03-27 | 2020-07-24 | 广东工业大学 | Inertial-based pedometer method, device and storage medium |
| CN111442785B (en) * | 2020-03-27 | 2021-01-26 | 广东工业大学 | Inertial-based pedometer method, device and storage medium |
| CN111358472B (en) * | 2020-03-27 | 2022-08-23 | 山东建筑大学 | Gait detection algorithm, device and system |
| CN111358472A (en) * | 2020-03-27 | 2020-07-03 | 山东建筑大学 | Gait detection algorithm, device and system |
| CN112067015A (en) * | 2020-09-03 | 2020-12-11 | 青岛歌尔智能传感器有限公司 | Step counting method and device based on convolutional neural network and readable storage medium |
| CN112067015B (en) * | 2020-09-03 | 2022-11-22 | 青岛歌尔智能传感器有限公司 | Step counting method and device based on convolutional neural network and readable storage medium |
| CN112923922A (en) * | 2021-03-04 | 2021-06-08 | 香港理工大学深圳研究院 | Method, system and storage medium for counting steps and determining position information of pedestrian |
| WO2022227202A1 (en) * | 2021-04-30 | 2022-11-03 | 东莞市小精灵教育软件有限公司 | Motion state recognition method and system, and wearable device and storage medium |
| CN113008231A (en) * | 2021-04-30 | 2021-06-22 | 东莞市小精灵教育软件有限公司 | Motion state identification method and system, wearable device and storage medium |
| CN113378801A (en) * | 2021-08-11 | 2021-09-10 | 深圳市梦瑶科技有限公司 | Intelligent bracelet accurate step counting method based on neural network |
| CN114034313A (en) * | 2021-11-10 | 2022-02-11 | 广东乐心医疗电子股份有限公司 | Step counting method and device |
| CN114034313B (en) * | 2021-11-10 | 2024-04-02 | 广东乐心医疗电子股份有限公司 | Step counting method and device |
| CN114139669A (en) * | 2021-11-19 | 2022-03-04 | 歌尔科技有限公司 | Rope skipping counting method, device and medium thereof |
| CN114139669B (en) * | 2021-11-19 | 2024-10-22 | 歌尔科技有限公司 | Rope skipping counting method, device and medium thereof |
| CN114739412A (en) * | 2022-04-11 | 2022-07-12 | 河海大学 | Pedestrian gait real-time detection method and device based on smart phone |
| CN114739412B (en) * | 2022-04-11 | 2024-05-03 | 河海大学 | A real-time pedestrian gait detection method and device based on smart phone |
| CN114815695A (en) * | 2022-04-28 | 2022-07-29 | 广东万颗子智控科技有限公司 | Remote control instruction generation method and device, remote control equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107462258B (en) | 2020-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107462258B (en) | Step counting method based on mobile phone three-axis acceleration sensor | |
| Yang et al. | Deep convolutional neural networks on multichannel time series for human activity recognition. | |
| Chen et al. | A deep learning approach to human activity recognition based on single accelerometer | |
| CN103970271B (en) | The daily routines recognition methods of fusional movement and physiology sensing data | |
| Rui et al. | Segmenting visual actions based on spatio-temporal motion patterns | |
| WO2021115064A1 (en) | Fitness exercise recognition method based on wearable sensor | |
| CN106096662A (en) | Human motion state identification based on acceleration transducer | |
| CN112464738A (en) | Improved naive Bayes algorithm user behavior identification method based on mobile phone sensor | |
| Zhao et al. | Detection and recognition of human body posture in motion based on sensor technology | |
| CN108549900A (en) | Tumble detection method for human body based on mobile device wearing position | |
| CN111401435A (en) | A method of human motion pattern recognition based on sports bracelet | |
| WO2023035093A1 (en) | Inertial sensor-based human body behaviour recognition method | |
| CN110286774A (en) | A Sign Language Recognition Method Based on Wrist Motion Sensor | |
| Shan et al. | Deep analysis for smartphone-based human activity recognition | |
| CN107506781A (en) | A kind of Human bodys' response method based on BP neural network | |
| CN115393956A (en) | CNN-BiLSTM fall detection method with improved attention mechanism | |
| Geng | Research on athlete’s action recognition based on acceleration sensor and deep learning | |
| Li et al. | Multiresolution fusion convolutional network for open set human activity recognition | |
| Fu et al. | Ping pong motion recognition based on smart watch | |
| CN114881079A (en) | Method and system for abnormal detection of human motion intention for wearable sensor | |
| CN109271889A (en) | A kind of action identification method based on the double-deck LSTM neural network | |
| Liu et al. | Preimpact fall detection for elderly based on fractional domain | |
| Fu et al. | Unconstrained workout activity recognition on unmodified commercial off-the-shelf smartphones | |
| CN109350072B (en) | Step frequency detection method based on artificial neural network | |
| CN111914724A (en) | Continuous Chinese sign language identification method and system based on sliding window segmentation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200616 |