CN107145844A - A kind of signal detecting method merged based on sub- window - Google Patents
A kind of signal detecting method merged based on sub- window Download PDFInfo
- Publication number
- CN107145844A CN107145844A CN201710263060.6A CN201710263060A CN107145844A CN 107145844 A CN107145844 A CN 107145844A CN 201710263060 A CN201710263060 A CN 201710263060A CN 107145844 A CN107145844 A CN 107145844A
- Authority
- CN
- China
- Prior art keywords
- window
- signal
- sub
- msubsup
- mrow
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of signal detecting method merged based on sub- window, the present invention detects signal using the less sub- window of size, and is iterated merging, finally merges into female window comprising complete hand signal.Every sub- window rule of thumb threshold decision whether as actuating signal a part.When not detecting actuating signal within the time more than 0.4 second, above multiple continuous overlapping sub- windows are just merged into a female window, finally, the actuating signal of each female window kind covers multiple overlapping sub- windows, and it is imperfect and detect the signal of multiple redundancies that female window avoids signal detection well.Time complexity of the present invention is low, and the computer speed of service is not influenceed, compared with conventional window function signal detection algorithm, and the algorithm effectively prevent imperfect and detection redundant signals the situation of signal detection, substantially increase accuracy of detection.
Description
Technical field
The invention belongs to signal detection technique field, it is related to a kind of signal detecting method merged based on sub- window, especially relates to
And a kind of method of intermediate frequency or low frequency acceleration signal detection to human action.
Background technology
Judge that the acceleration of human action can be detected by window function and using the average energy of random noise as threshold value
Spend signal.But the problem of window width size influences signal detection precision occurs.If it is too small that window width is set, it certainly will occur
The incomplete situation of signal detection occurs, if window width setting is excessive, will cover two or more actions in same window
Signal, causes to examine the action for not measuring and having performed.Therefore, it is badly in need of wanting a kind of relatively low accurate signal detection of time complexity
Method.
The content of the invention
In order to solve the technical problem present in existing signal detection, the invention provides a kind of method complexity is low, easy
In the signal detecting method for understanding and using.
The technical solution adopted in the present invention is：1. a kind of signal detecting method merged based on sub- window, it is characterised in that
Comprise the following steps：
Step 1：Set sub- window of the width as N number of sampled point；
Step 2：If the value of three axles of acceleration is xi、yi、zi, calculate the resultant acceleration signal of three axles in every sub- window
Energy value EI；
Wherein, I represents the index of current sub- window；
Step 3：If signal energy is more than noise threshold in sub- window, regards it as a part for signal and record,
Next sub- window of detection is then proceeded to, if still greater than the threshold value, just remaining and being merged with the sub- window above recorded；When
When detecting the time interval containing action next time more than preset time, then it is assumed that the signal detection is finished；
Step 4：All sub- windows above recorded are merged, the window after merging is referred to as female window, the letter in female window
Number with regard to being used as the human action signal to be detected.
The present invention detects signal using the less sub- window of size, and is iterated merging, finally merges into comprising complete
Female window of hand signal, efficiently solves signal detection in traditional window detection method imperfect and detect asking for redundant signals
Topic.
Brief description of the drawings
Fig. 1 is illustrated with the signal detection based on sub- window merging method based on conventional window functional based method in the embodiment of the present invention
Figure.
Fig. 2 be in the embodiment of the present invention based on conventional window functional based method with based on sub- window merging method signal detection performance ratio
Relatively illustrate.
Fig. 3 is the time complexity signal based on sub- window merging method detection signal in the embodiment of the present invention.
Embodiment
Understand for the ease of those of ordinary skill in the art and implement the present invention, below in conjunction with the accompanying drawings and embodiment is to this hair
It is bright to be described in further detail, it will be appreciated that implementation example described herein is merely to illustrate and explain the present invention, not
For limiting the present invention.
The present invention mainly solves the technical problem present in existing signal detection；Provide a method that complexity it is low,
The signal detecting method that should be readily appreciated that and use, solves signal detection in traditional window detection method imperfect and detect redundancy
The problem of signal, this method is to detect signal using the less sub- window of size, and is iterated merging, finally merges into and includes
Female window of complete hand signal.Then detected using a wide rectangular window of only N number of sampled point, sliding window Duplication choosing
50% is selected, sub- window is called.And calculate the energy value of the resultant acceleration signal of three axles in every sub- window.If signal energy in sub- window
When amount is more than noise threshold, a part for signal can be regarded it as and recorded, next sub- window of detection be then proceeded to, if still
More than the threshold value, just remain and merged with the sub- window that above records, one-time detection was arrived containing the time acted instantly
When interval was more than 0.4 second (usual people's judged reflecting time for 0.15s to 0.4s or more), you can think that the signal detection is complete
Finish, and all sub- windows above recorded are merged, the window after merging is referred to as female window, the signal in female window is just used as will
The human action signal detected.
The specific method flow of the present invention is as follows：
Input:The value x of three axles of accelerationi,yi,zi.
Output:The data X, Y, Z of three axles of female window where acceleration signal
Every sub- window rule of thumb threshold decision whether as actuating signal a part.When in the time more than 0.4 second
Actuating signal is not detected inside, and above multiple continuous overlapping sub- windows are just merged into a female window, finally, each mother's window kind
Actuating signal cover multiple overlapping sub- windows, it is imperfect and detect multiple redundancies that female window avoids signal detection well
Signal.
As shown in figure 1, the present invention can be very good to avoid because signal detection is imperfect or same window in comprising multiple
The situation of signal occurs, so as to do basis for follow-up feature extraction.
The present embodiment merges signal detection algorithm with sub- window respectively to ten hands of human body based on conventional window function check algorithm
Exemplified by the gesture classification of motion, by selecting SVM to be used as grader.And set respectively sub- window merge the N of algorithm neutron window size for 6,
8th, 10,12,14,16,18,20,22,24,26,28, classification performance is then solved respectively.A kind of window width is often selected, in training set
In, verify classifier performance using ten folding cross validation methods, i.e., training set be divided into ten parts, in turn will wherein nine parts for instructing
Practice model, portion verifies Model Identification precision as test set.Finally using the accuracy of identification of test as judging algorithm
Foundation, as shown in Figure 2：
It was found from Fig. 2 (a), using signal detection algorithm, when group window window width is 18, accuracy of identification highest is average
Accuracy of identification is 92.14%, and variance is 8.77%.Thus selection N is 18.When using common window detection algorithm, identification essence
Degree increases with the increase of window width size, and variance reduces with the increase of window width size, because when window width is too small
When, signal easily occurs detecting incomplete phenomenon, causes accuracy of identification to reduce, even if when window width size increases to a conjunction
Suitable value, its accuracy of identification also only has 86% or so, and sub- window merges algorithm and can not only solved the above problems well, makes identification
Stable accuracy is 90% or so, and variance is also at reduced levels value, with very high stability.It was found from Fig. 2 (b), in window width
For 18 when, the algorithm is no more than 0.06ms to the average time-consuming maximum of different gestures, and its time-consuming standard deviation is no more than 0.01ms, such as
Shown in Fig. 3, thus computing power is also had little to no effect.
Therefore, this algorithm has the following advantages that：Time complexity is low, algorithm is it can be readily appreciated that signal detection effect is good.Pass through
The detection algorithm that this seed window of design merges detects the different actuating signal of human body.Do not influenceing the feelings of the computer speed of service
Under condition, compared with conventional signal detection algorithm, the algorithm effectively prevent the incomplete phenomenon of signal detection, substantially increase
Accuracy of detection.
It should be appreciated that the part that this specification is not elaborated belongs to prior art.
It should be appreciated that the above-mentioned description for preferred embodiment is more detailed, therefore it can not be considered to this
The limitation of invention patent protection scope, one of ordinary skill in the art is not departing from power of the present invention under the enlightenment of the present invention
Profit is required under protected ambit, can also be made replacement or be deformed, each fall within protection scope of the present invention, this hair
It is bright scope is claimed to be determined by the appended claims.
Claims (3)
1. a kind of signal detecting method merged based on sub- window, it is characterised in that comprise the following steps：
Step 1：Set sub- window of the width as N number of sampled point；
Step 2：If the value of three axles of acceleration is xi、yi、zi, calculate the energy of the resultant acceleration signal of three axles in every sub- window
Value EI；
<mrow>
<msub>
<mi>E</mi>
<mi>I</mi>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<msubsup>
<mi>F</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<mo>;</mo>
</mrow>
<mrow>
<msub>
<mi>F</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<msqrt>
<mrow>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<mo>+</mo>
<msubsup>
<mi>y</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<mo>+</mo>
<msubsup>
<mi>z</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
</mrow>
</msqrt>
<mo>;</mo>
</mrow>
Wherein, I represents the index of current sub- window；
Step 3：If signal energy is more than noise threshold in sub- window, regards it as a part for signal and record, then
Continue to detect next sub- window, if still greater than the threshold value, just remaining and merging with the sub- window above recorded；When next
It is secondary detect containing action time interval exceed preset time when, then it is assumed that the signal detection is finished；
Step 4：All sub- windows above recorded are merged, the window after merging is referred to as female window, the signal in female window is just
It is used as the human action signal to be detected.
2. the signal detecting method according to claim 1 merged based on sub- window, it is characterised in that：It is sub described in step 1
The Duplication of window is set to 50%.
3. the signal detecting method according to claim 1 merged based on sub- window, it is characterised in that：When being preset in step 4
Between be 0.4 second.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710263060.6A CN107145844B (en) | 2017-04-20 | 2017-04-20 | A kind of signal detecting method merged based on sub- window |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710263060.6A CN107145844B (en) | 2017-04-20 | 2017-04-20 | A kind of signal detecting method merged based on sub- window |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107145844A true CN107145844A (en) | 2017-09-08 |
| CN107145844B CN107145844B (en) | 2019-06-25 |
Family
ID=59774853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710263060.6A Active CN107145844B (en) | 2017-04-20 | 2017-04-20 | A kind of signal detecting method merged based on sub- window |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107145844B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110569800A (en) * | 2019-09-10 | 2019-12-13 | 武汉大学 | detection method of handwriting signal |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101975940A (en) * | 2010-09-27 | 2011-02-16 | 北京理工大学 | Segmentation combination-based adaptive constant false alarm rate target detection method for SAR image |
| CN103514664A (en) * | 2012-06-14 | 2014-01-15 | 索尼公司 | Parking charging achievement device, system and method |
| CN205942282U (en) * | 2016-07-27 | 2017-02-08 | 广东石油化工学院 | Intelligence life information watch |
-
2017
- 2017-04-20 CN CN201710263060.6A patent/CN107145844B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101975940A (en) * | 2010-09-27 | 2011-02-16 | 北京理工大学 | Segmentation combination-based adaptive constant false alarm rate target detection method for SAR image |
| CN103514664A (en) * | 2012-06-14 | 2014-01-15 | 索尼公司 | Parking charging achievement device, system and method |
| CN205942282U (en) * | 2016-07-27 | 2017-02-08 | 广东石油化工学院 | Intelligence life information watch |
Non-Patent Citations (1)
| Title |
|---|
| 覃方君等: "基于新息自适应卡尔曼滤波的加速度计信号降噪", 《数据采集与处理》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110569800A (en) * | 2019-09-10 | 2019-12-13 | 武汉大学 | detection method of handwriting signal |
| CN110569800B (en) * | 2019-09-10 | 2022-04-12 | 武汉大学 | Detection method of handwriting signal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107145844B (en) | 2019-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103971102B (en) | Static gesture recognition method based on finger contour and decision-making trees | |
| CN108875067A (en) | text data classification method, device, equipment and storage medium | |
| EP2477136B1 (en) | Method for continuously verifying user identity via keystroke dynamics | |
| CN106095996A (en) | Method for text classification | |
| US20160012348A1 (en) | Touch classification | |
| CN105844216A (en) | Detection and matching mechanism for recognition of handwritten letters using WiFi signals | |
| CN103795612A (en) | Method for detecting junk and illegal messages in instant messaging | |
| CN101359373B (en) | Method and device for recognizing degraded character | |
| CN105572651A (en) | CFAR detection method based on clutter background statistical recognition | |
| CN107273825A (en) | Personal identification method is merged based on the physiological signal for improving canonical correlation analysis | |
| CN103474061A (en) | Automatic distinguishing method based on integration of classifier for Chinese dialects | |
| CN105631519A (en) | Convolution nerve network acceleration method based on pre-deciding and system | |
| CN103675787A (en) | One-dimension range profile optimal orthogonal nolinear subspace identification method for radar targets | |
| CN107885849A (en) | A kind of moos index analysis system based on text classification | |
| CN105389486A (en) | Authentication method based on mouse behavior | |
| Abdu-Aguye et al. | Detecting adversarial attacks in time-series data | |
| CN107145844A (en) | A kind of signal detecting method merged based on sub- window | |
| CN104915009A (en) | Gesture prediction method and system | |
| CN106776751A (en) | The clustering method and clustering apparatus of a kind of data | |
| Yu et al. | Mental workload classification via online writing features | |
| CN109886164B (en) | Abnormal gesture recognition and processing method | |
| CN109977737A (en) | A kind of character recognition Robust Method based on Recognition with Recurrent Neural Network | |
| CN104298980A (en) | Fingerprint matching method and device based on smart card | |
| CN103472946A (en) | Method for dynamically adjusting reporting rate of touch screen | |
| CN106295683A (en) | A kind of outlier detection method of time series data based on sharpness |
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 |