CN107280673A - A kind of infrared imaging breath signal detection method based on key-frame extraction technique - Google Patents
A kind of infrared imaging breath signal detection method based on key-frame extraction technique Download PDFInfo
- Publication number
- CN107280673A CN107280673A CN201710408606.2A CN201710408606A CN107280673A CN 107280673 A CN107280673 A CN 107280673A CN 201710408606 A CN201710408606 A CN 201710408606A CN 107280673 A CN107280673 A CN 107280673A
- Authority
- CN
- China
- Prior art keywords
- frame
- msub
- mrow
- infrared
- video
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/0826—Detecting or evaluating apnoea events
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
Abstract
The invention discloses a kind of infrared imaging breath signal detection method based on key-frame extraction technique.Infrared image video is obtained to face's imaging using infrared thermoviewer first, secondly air-breathing is extracted with start frame when exhaling as key frame, real-time respiratory cycle parameter is obtained finally according to time series of the key frame in infrared video.The method of the present invention is effective and feasible, can realize the non-contact detection to breath signal.
Description
Technical field
The invention belongs to image processing field, particularly a kind of infrared imaging breath signal based on key-frame extraction technique
Detection method.
Background technology
Breath signal includes abundant physiologic information, is that can reflect one of signal of interest of human body situation.Exhale
The detection for inhaling signal has great significance for respiratory disorders such as diagnosis OSASs (OSAHS).
Traditional breath signal detection method is contact, needs to stick electrode with measured during detection.It is this
The breathing detection method of contact is primarily present following shortcoming：1) contact electrode can to patient physiology with psychologically bringing one
Fixed sense of discomfort and pressure, this will influence the physiological characteristic of patient, reduce the accuracy of testing result；2) to special crowd
(as burnt, scalding, infectious disease, skin disease, infant patients), the method for contact is not applied to simultaneously；3) complex operation is cumbersome, takes
Arduously.
Therefore, be badly in need of a kind of contactless breath signal detection method at present to solve to be mentioned above various asks
Topic, patient is freed from pain, and can effectively improve the accuracy of breath signal detection, but there is no in the prior art
Associated description.
The content of the invention
Carried it is an object of the invention to the deficiency existed for contact breathing detection there is provided a kind of based on key frame
The infrared imaging breath signal detection method of technology is taken, breath signal is obtained by infrared thermography.
The technical solution for realizing the object of the invention is：A kind of infrared imaging breathing letter based on key-frame extraction technique
Number detection method, comprises the following steps：
Step 1, using infrared thermoviewer face is imaged, obtains infrared image video；
Step 2, by infrared image Video segmentation into frame of video；
Step 3, the whole two field picture positioning nostril region of scanning；Specially：
M maximum point of image pixel intensities change is used as characteristic point in step 3-1, selected whole two field picture；
Wherein
Step 3-2, detection and tracking feature point, so as to position nostril region.
Step 4, the image entropy for determining n-th frame frame of video nostril region in video, image entropy HnFormula is as follows：
In formula, pkRepresent that gray value is the ratio shared by k pixel in image；
Step 5, the change waveform for obtaining the image entropy of the nostril region of all frame of video in infrared video, and it is carried out
Bandpass filtering；
Step 6, the waveform progress valley detection that changes to the image entropy after bandpass filtering, frame where choosing trough point is crucial
Frame；Specially：
Step 6-1, the first derivative for determining image entropy change curve every, selected first derivative are used as pre-selection for 0 point
Point, be specially：The point for meeting equation below is pre- reconnaissance：
Step 6-2, the wherein corresponding image entropy progress second order derivation of pre- reconnaissance to being determined in step 6-1, second dervative
Point more than 0 is trough point, is specially：The point for meeting equation below is trough point：
Frame where step 6-3, selection trough point is key frame.
Step 7, real-time respiratory cycle parameter obtained according to time series of the key frame in infrared video.Specifically
For：
Step 7-1, key frame corresponding time series in infrared video is determined, be specially：According to above-mentioned steps from red
One group of key frame is extracted in outer video, if the n-th key frame in this group of key frame is the i-th frame in infrared image video,
The frame corresponding time series in infrared video is ti, then have：
Wherein FsTo record the sample frequency of infrared video；
Step 7-2, using the time interval being separated by between key frame as respiration cycle, be specially：If once exhaling
The cycle T of suction is the difference of the time series of N+2 and n-th key frame in the one group of key frame extracted, if key frame group
In n-th key frame be the i-th frame in infrared image video, its time series is ti, the N+2 key frame is infrared image
Jth frame in video, its time series is tj, cycle T formula is as follows：
Compared with prior art, its remarkable advantage is the present invention：1) present invention realizes breathing using infrared thermoviewer
Non-contact detection, compared with traditional contact measurement mode, substantially reduce measured's sense of discomfort；2) detection is conducive to
As a result accuracy and operation more facilitates；3) face is imaged using infrared thermoviewer, is obtaining breath signal
Meanwhile, protect the privacy of measured.
The present invention is described further with reference to Figure of description.
Brief description of the drawings
Fig. 1 realizes the step block diagram of breathing detection for the present invention.
Fig. 2 is the video frame images Entropy Changes oscillogram after bandpass filtering.
Fig. 3 is the result figure that valley detection is carried out to the change waveform after bandpass filtering.
Embodiment
With reference to Fig. 1, the infrared imaging breath signal detection algorithm step of the invention based on key-frame extraction technique is as follows：
Step 1, using infrared thermoviewer to face imaging obtain infrared image video；
Step 2, by infrared image Video segmentation into frame of video；
Step 3, the whole two field picture positioning nostril region of scanning；Specially：
M maximum point of image pixel intensities change is used as characteristic point in step 3-1, selected whole two field picture；
Wherein
Step 3-2, detection and tracking feature point, so that nostril region is positioned, can be using Harris, KLT, SUSAN etc.
Algorithm is detected and tracking feature point.
Step 4, the image entropy for determining n-th frame frame of video nostril region in video, image entropy HnFormula is as follows：
In formula, pkRepresent that gray value is the ratio shared by k pixel in image；
Step 5, the change waveform for obtaining the image entropy of the nostril region of all frame of video in infrared video, and it is carried out
Bandpass filtering；Specially：The frequency of eupnea is 0.2~0.33HZ, and breathing is divided into expiration and two processes of air-breathing, each
During, the image entropy of nasal area has the process of a first increases and then decreases, therefore image entropy changes the cycle of waveform about
For the half of respiratory cycle, frequency is about twice of respiratory rate, is 0.4~0.66HZ.Therefore the passband frequency of bandpass filter
When rate is 0.3~0.8HZ, it can effectively filter out to record in infrared video and be exhaled with the noise signal produced in transmitting procedure, reservation
Inhale signal；
Step 6, due to being breathed in infrared video when exhale and the image entropy of start frame of air-breathing is minimum point, it is therefore right
Filtered image entropy change waveform carries out valley detection, and frame where choosing trough point is key frame.Specially：
Step 6-1, the first derivative for determining image entropy change curve every, selected first derivative are used as pre-selection for 0 point
Point, be specially：The point for meeting equation below is pre- reconnaissance：
Step 6-2, the wherein corresponding image entropy progress second order derivation of pre- reconnaissance to being determined in step 6-1, second dervative
Point more than 0 is trough point, is specially：The point for meeting equation below is trough point：
Frame where step 6-3, selection trough point is key frame.
Step 7, real-time respiratory cycle parameter obtained according to time series of the key frame in infrared video.Specifically
For：It is separated by the key frame time interval as respiratory cycle.Specially：
Step 7-1, key frame corresponding time series in infrared video is determined, be specially：According to above-mentioned steps from red
One group of key frame is extracted in outer video, if the n-th key frame in this group of key frame is the i-th frame in infrared image video,
The frame corresponding time series in infrared video is ti, then have：
Wherein FsTo record the sample frequency of infrared video；
Step 7-2, using the time interval being separated by between key frame as respiration cycle, be specially：If once exhaling
The cycle T of suction is the difference of the time series of N+2 and n-th key frame in the one group of key frame extracted, if key frame group
In n-th key frame be the i-th frame in infrared image video, its time series is ti, the N+2 key frame is infrared image
Jth frame in video, its time series is tj, cycle T formula is as follows：
The present invention realizes the non-contact detection of breathing using infrared thermoviewer, with traditional contact measurement mode phase
Than substantially reducing measured's sense of discomfort.
Further detailed description is done to the present invention with reference to embodiment.
Embodiment
The infrared image video of recording 1 minute as described in step 1, sample frequency is that 25 frames are per second, and a nostril area is about
For 0.6cm2, actual photographed area is about 900cm2, both ratios are about 6.67e-4, and total pixel number per two field picture is (582
× 776), then M value is about 301 (582 × 776 × 6.67e-4).While infrared video is recorded, detected with breathing zone
The respiratory rate of measured is used as reference value.It is time series of the key frame selected in infrared video with reference to table 1, in table,
It is separated by the time interval as respiratory cycle between key frame.The average period that each respiratory cycle can averagely be breathed is
3.1863s, respiratory rate (inverse of average respiratory cycle) is 0.3138HZ, and the respiratory rate that breathing zone is measured is 0.3129HZ, two
Person is basically identical.
Table 1
From the foregoing, it will be observed that the present invention can relatively accurately detect respiratory cycle parameter, the present invention is effective and feasible.
Claims (4)
1. a kind of infrared imaging breath signal detection method based on key-frame extraction technique, it is characterised in that including following step
Suddenly：
Step 1, using infrared thermoviewer face is imaged, obtains infrared image video；
Step 2, by infrared image Video segmentation into frame of video；
Step 3, the whole two field picture positioning nostril region of scanning；
Step 4, the image entropy for determining n-th frame frame of video nostril region in video, image entropy HnFormula is as follows：
<mrow>
<msub>
<mi>H</mi>
<mi>n</mi>
</msub>
<mo>=</mo>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mn>255</mn>
</munderover>
<msub>
<mi>p</mi>
<mi>k</mi>
</msub>
<mi>log</mi>
<mi> </mi>
<msub>
<mi>p</mi>
<mi>k</mi>
</msub>
</mrow>
In formula, pkRepresent that gray value is the ratio shared by k pixel in image；
Step 5, the change waveform for obtaining the image entropy of the nostril region of all frame of video in infrared video, and band logical is carried out to it
Filtering；
Step 6, the waveform progress valley detection that changes to the image entropy after bandpass filtering, selection trough point place frame is key frame；
Step 7, real-time respiratory cycle parameter obtained according to time series of the key frame in infrared video.
2. the infrared imaging breath signal detection method according to claim 1 based on key-frame extraction technique, its feature
It is, whole two field picture positioning nostril region is scanned in step 3；Specially：
M maximum point of image pixel intensities change is used as characteristic point in step 3-1, selected whole two field picture；
Wherein
Step 3-2, detection and tracking feature point, so as to position nostril region.
3. the infrared imaging breath signal detection method according to claim 1 based on key-frame extraction technique, its feature
It is, step 6 carries out valley detection to the image entropy change waveform after bandpass filtering, and frame where choosing trough point is key frame；
Specially：
Step 6-1, the first derivative for determining image entropy change curve every, selected first derivative are used as pre- reconnaissance, tool for 0 point
Body is：The point for meeting equation below is pre- reconnaissance：
<mrow>
<mfrac>
<mrow>
<msub>
<mi>dH</mi>
<mi>n</mi>
</msub>
</mrow>
<mrow>
<mi>d</mi>
<mi>n</mi>
</mrow>
</mfrac>
<mo>=</mo>
<mn>0</mn>
</mrow>
Step 6-2, the corresponding image entropy progress second order derivation of pre- reconnaissance to being determined in step 6-1, wherein second dervative are more than 0
Point be trough point, be specially：The point for meeting equation below is trough point：
<mrow>
<mfrac>
<mrow>
<msup>
<mi>d</mi>
<mn>2</mn>
</msup>
<msub>
<mi>H</mi>
<mi>n</mi>
</msub>
</mrow>
<mrow>
<msup>
<mi>dn</mi>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
<mo>></mo>
<mn>0</mn>
</mrow>
Frame where step 6-3, selection trough point is key frame.
4. the infrared imaging breath signal detection method according to claim 1 based on key-frame extraction technique, its feature
It is, step 7 is to obtain real-time respiratory cycle parameter according to time series of the key frame in infrared video；Specially：
Step 7-1, key frame corresponding time series in infrared video is determined, be specially：Regarded according to above-mentioned steps from infrared
One group of key frame is extracted in frequency, if the n-th key frame in this group of key frame is the i-th frame in infrared image video, the frame
Corresponding time series is t in infrared videoi, then have：
<mrow>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<mi>i</mi>
<msub>
<mi>F</mi>
<mi>s</mi>
</msub>
</mfrac>
</mrow>
Wherein FsTo record the sample frequency of infrared video；
Step 7-2, using the time interval being separated by between key frame as respiration cycle, be specially：If respiration
Cycle T is the difference of the time series of N+2 and n-th key frame in the one group of key frame extracted, if in key frame group
N-th key frame is the i-th frame in infrared image video, and its time series is ti, the N+2 key frame is infrared image video
In jth frame, its time series be tj, cycle T formula is as follows：
<mrow>
<mi>T</mi>
<mo>=</mo>
<msub>
<mi>t</mi>
<mi>j</mi>
</msub>
<mo>-</mo>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<mi>j</mi>
<msub>
<mi>F</mi>
<mi>s</mi>
</msub>
</mfrac>
<mo>-</mo>
<mfrac>
<mi>i</mi>
<msub>
<mi>F</mi>
<mi>s</mi>
</msub>
</mfrac>
<mo>.</mo>
</mrow>
2
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710408606.2A CN107280673B (en) | 2017-06-02 | 2017-06-02 | A kind of infrared imaging breath signal detection method based on key-frame extraction technique |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710408606.2A CN107280673B (en) | 2017-06-02 | 2017-06-02 | A kind of infrared imaging breath signal detection method based on key-frame extraction technique |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107280673A true CN107280673A (en) | 2017-10-24 |
| CN107280673B CN107280673B (en) | 2019-11-15 |
Family
ID=60094883
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710408606.2A Active CN107280673B (en) | 2017-06-02 | 2017-06-02 | A kind of infrared imaging breath signal detection method based on key-frame extraction technique |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107280673B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111402201A (en) * | 2020-02-23 | 2020-07-10 | 中国科学院西安光学精密机械研究所 | Non-contact respiration signal measuring method capable of resisting motion interference |
| WO2021120007A1 (en) * | 2019-12-17 | 2021-06-24 | 深圳先进技术研究院 | Infrared image sequence-based sleep quality evaluation system and method |
| CN113100748A (en) * | 2021-03-30 | 2021-07-13 | 联想(北京)有限公司 | Respiratory frequency determination method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050043674A1 (en) * | 2001-11-22 | 2005-02-24 | Blair Edward Duncan | Medicament dispensing system |
| CN102973273A (en) * | 2012-11-29 | 2013-03-20 | 中国人民解放军第四军医大学 | Sleep respiratory function monitoring system based on infrared radiation detection |
| CN103217256A (en) * | 2013-03-20 | 2013-07-24 | 北京理工大学 | Local gray level-entropy difference leak detection locating method based on infrared image |
| CN103810499A (en) * | 2014-02-25 | 2014-05-21 | 南昌航空大学 | Application for detecting and tracking infrared weak object under complicated background |
| CN104055498A (en) * | 2014-04-30 | 2014-09-24 | 首都医科大学 | Non-contact human respiration and heart beat signal detection method based on infrared sequence image |
| CN104268844A (en) * | 2014-10-17 | 2015-01-07 | 中国科学院武汉物理与数学研究所 | Small target infrared image processing method based on weighing local image entropy |
| CN106716515A (en) * | 2014-09-11 | 2017-05-24 | 株式会社电装 | Driver state determination apparatus |
-
2017
- 2017-06-02 CN CN201710408606.2A patent/CN107280673B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050043674A1 (en) * | 2001-11-22 | 2005-02-24 | Blair Edward Duncan | Medicament dispensing system |
| CN102973273A (en) * | 2012-11-29 | 2013-03-20 | 中国人民解放军第四军医大学 | Sleep respiratory function monitoring system based on infrared radiation detection |
| CN103217256A (en) * | 2013-03-20 | 2013-07-24 | 北京理工大学 | Local gray level-entropy difference leak detection locating method based on infrared image |
| CN103810499A (en) * | 2014-02-25 | 2014-05-21 | 南昌航空大学 | Application for detecting and tracking infrared weak object under complicated background |
| CN104055498A (en) * | 2014-04-30 | 2014-09-24 | 首都医科大学 | Non-contact human respiration and heart beat signal detection method based on infrared sequence image |
| CN106716515A (en) * | 2014-09-11 | 2017-05-24 | 株式会社电装 | Driver state determination apparatus |
| CN104268844A (en) * | 2014-10-17 | 2015-01-07 | 中国科学院武汉物理与数学研究所 | Small target infrared image processing method based on weighing local image entropy |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021120007A1 (en) * | 2019-12-17 | 2021-06-24 | 深圳先进技术研究院 | Infrared image sequence-based sleep quality evaluation system and method |
| CN111402201A (en) * | 2020-02-23 | 2020-07-10 | 中国科学院西安光学精密机械研究所 | Non-contact respiration signal measuring method capable of resisting motion interference |
| CN111402201B (en) * | 2020-02-23 | 2023-04-11 | 中国科学院西安光学精密机械研究所 | Non-contact respiration signal measuring method capable of resisting motion interference |
| CN113100748A (en) * | 2021-03-30 | 2021-07-13 | 联想(北京)有限公司 | Respiratory frequency determination method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107280673B (en) | 2019-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Reyes et al. | Tidal volume and instantaneous respiration rate estimation using a volumetric surrogate signal acquired via a smartphone camera | |
| US10004410B2 (en) | System and methods for measuring physiological parameters | |
| US9443289B2 (en) | Compensating for motion induced artifacts in a physiological signal extracted from multiple videos | |
| US9336594B2 (en) | Cardiac pulse rate estimation from source video data | |
| US9436984B2 (en) | Compensating for motion induced artifacts in a physiological signal extracted from a single video | |
| CN107358220A (en) | A kind of human heart rate and the contactless measurement of breathing | |
| KR102054213B1 (en) | Respiratory measurement system using thermovision camera | |
| CN104545888B (en) | A kind of sleep apnea acquisition analysis system based on dynamic electrocardiogram and respiratory wave acquisition | |
| KR101752873B1 (en) | Method and system for extracting heart information of time domain | |
| CN104173051A (en) | Automatic noncontact respiration assessing system and assessing method | |
| CN107280673B (en) | A kind of infrared imaging breath signal detection method based on key-frame extraction technique | |
| CN203290905U (en) | Non-contact automatic respiration measuring system | |
| CN107616795A (en) | A kind of contactless respiratory rate detection method in real time based on camera | |
| CN106999062A (en) | The method that heart information is extracted based on human body fine motion | |
| CN112294282A (en) | Self-calibration method of emotion detection device based on RPPG | |
| CN112806966A (en) | Non-interference type early warning system and method for apnea in sleep | |
| US10292594B2 (en) | Method and system for contactless detection of cardiac activity | |
| Luguern et al. | Remote photoplethysmography combining color channels with SNR maximization for respiratory rate assessment | |
| KR101996027B1 (en) | Method and system for extracting Heart Information of Frequency domain by using pupil size variation | |
| CN112869737B (en) | Non-contact human body blood oxygen saturation detection method | |
| Lin et al. | Step count and pulse rate detection based on the contactless image measurement method | |
| Liu et al. | A principal component analysis based data fusion method for estimation of respiratory volume | |
| Frigola et al. | Vision based respiratory monitoring system | |
| KR20170004548A (en) | Method and system for extracting heart information of time domain | |
| Maji et al. | Contactless heart rate monitoring using A standard RGB camera |
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 |