CN-122004808-A - Closed space infrared induction head physiological monitoring system

CN122004808ACN 122004808 ACN122004808 ACN 122004808ACN-122004808-A

Abstract

The invention relates to the technical field of physiological characteristic non-sensing monitoring and multi-mode sensor information fusion, in particular to a closed space infrared sensing head physiological monitoring system which comprises a multi-mode detection acquisition module, an optical path dynamic compensation module, a thermal inertia separation processing module and a cross-mode signal self-healing module. The invention abandons simple data stacking of the traditional multisensor, establishes a strict hardware-level clock synchronization mechanism at the front end, combines a depth displacement vector obtained by time-of-flight ranging with a square inverse attenuation law of illumination intensity, executes closed-loop optical path dynamic gain compensation, eliminates motion artifact interference caused by space displacement from an optical object bottom, simultaneously deeply digs the thermal inertia physical difference between the environment in a closed cabin and the physiological activity of a human body, and precisely peels off a thermal noise baseline under extremely severe environment by constructing a time sequence double-channel filter of a specific time window, thereby realizing high-fidelity extraction of weak respiratory rhythm.

Inventors

ZHOU GUANGYIN
CHEN RONG
LIU ZIHAO
ZHOU SHICHANG
TIAN FUDE
You can
WU CHUYUE
GAO AO
LIU DANPENG
XU GANG

Assignees

重庆智能工程职业学院

Dates

Publication Date: 20260512
Application Date: 20260326

Claims (10)

1. An enclosed-space infrared-sensing head physiological monitoring system, comprising: The multi-mode detection acquisition module is used for synchronously acquiring a three-dimensional depth displacement vector, a near infrared photoelectric pulse wave signal and a far infrared thermal radiation pattern sequence of the face of the target in the closed space; The optical path dynamic compensation module is connected with the multi-mode detection acquisition module and is used for dynamically adjusting the exposure parameters or digital gain of near infrared detection by taking the three-dimensional depth displacement vector as a compensation reference so as to output a reference pulse wave signal for eliminating motion artifacts; The thermal inertia separation processing module is used for constructing a time sequence dual-channel filter aiming at the far infrared thermal radiation diagram sequence based on the physical thermal inertia difference between the cabin environment background and the target physiological activity, subtracting and stripping the low-frequency environment thermal noise baseline from the total thermal radiation diagram, and extracting a pure target respiratory rhythm signal; And the cross-modal signal self-healing module is respectively connected with the optical path dynamic compensation module and the thermal inertia separation processing module and is used for monitoring the signal-to-noise ratio of the reference pulse wave signal in real time, extracting the phase characteristics of the respiratory rhythm signal when the signal-to-noise ratio is lower than a preset safety threshold value, substituting the phase characteristics into a preset cardiopulmonary coupling rhythm model, and carrying out mathematical interpolation and curve smoothing on the missing heart rate data fragments so as to output continuous uninterrupted physiological characteristic parameters.
2. The closed space infrared sensing head physiology monitoring system according to claim 1, wherein the multi-modality detection acquisition module comprises: The system comprises a time-of-flight ranging sensor, a near infrared camera component with a central wavelength of 850nm or 940nm, an uncooled infrared thermal imaging component and a global clock generator; The global clock generator is respectively and electrically connected with the flight time ranging sensor, the near infrared camera shooting assembly and the uncooled infrared thermal imaging assembly and is used for sending hardware trigger pulses with the same frequency to the flight time ranging sensor, the near infrared camera shooting assembly and the uncooled infrared thermal imaging assembly, so that the acquired three-dimensional depth displacement vector, the near infrared photoelectric pulse wave signal and the far infrared thermal radiation pattern sequence are kept in strict frame synchronization alignment on a time axis.
3. The closed space infrared sensing head physiological monitoring system according to claim 2, wherein the multi-modal detection acquisition module, when acquiring the three-dimensional depth displacement vector, specifically performs the following steps: Acquiring initial three-dimensional point cloud data of a target face by using the time-of-flight ranging sensor, extracting a rigid characteristic region of the face as a reference anchor point, and recording initial three-dimensional coordinates of the rigid characteristic region : Tracking the rigid feature region in real time at the current time during continuous monitoring Three-dimensional coordinates of (a) ; To eliminate speckle high frequency noise of a ranging sensor, a sliding window smoothing strategy is adopted to calculate the current moment Is a three-dimensional depth displacement vector of (2) : Wherein, the For a preset number of sliding window frames, Is that And (5) three-dimensional coordinates of the rigid characteristic region at the moment.
4. A closed space infrared sensing head physiological monitoring system according to claim 3, wherein the optical path dynamic compensation module comprises a depth projection unit and a gain calculation unit; The depth projection unit is used for carrying out the three-dimensional depth displacement vector Projecting the target amount of the distance variation in the optical axis direction to the physical optical axis direction of the near infrared camera shooting assembly ; The gain calculation unit is used for calculating the current moment based on the inverse square attenuation physical law of the illumination intensity Dynamic gain compensation coefficient of (a) The specific calculation formula is as follows: Wherein, the The amplification gain is preset for the system of the target face at the initial reference moment, For initial three-dimensional coordinates An initial absolute distance from the near infrared camera assembly.
5. The closed space infrared sensing head physiological monitoring system according to claim 4, wherein said optical path dynamic compensation module compensates said dynamic gain compensation coefficient And the signal multiplier is directly multiplied by the amplitude of the original photoelectric pulse wave signal acquired currently so as to output the reference pulse wave signal for eliminating the spatial displacement modulation artifact.
6. The closed space infrared sensing head physiological monitoring system according to claim 5, wherein the thermal inertia separation processing module, when extracting the target respiratory rhythm signal, specifically comprises the following steps: Positioning a target subnasal characteristic region from the far infrared thermal radiation diagram sequence, extracting a time domain change sequence of temperature average values of all pixel points in the region, and recording the time domain change sequence as an original thermal radiation sequence ; Based on large thermal inertia physical characteristics of closed space environment, a large-scale sliding time window is constructed, and a slow-change baseline of environmental thermal noise is extracted The calculation formula is as follows: Wherein, the Is the length of a preset background baseline time window, and The value of (2) is greater than 3 times the target minimum effective breathing period.
7. The closed space infrared sensing head physiological monitoring system according to claim 6, wherein said original thermal radiation sequence is subtracted from said slowly varying baseline by time domain differences to obtain a primary sequence of respiratory characteristics that eliminates environmental thermal interference : Primary sequence of the respiratory characteristics Is input into a preset physiological band-pass filter, and the pass band frequency of the band-pass filter is limited as follows Wherein And And respectively corresponding to the lower limit and the upper limit of the normal respiratory rate of the human body, and outputting the smooth pure target respiratory rhythm signal after filtering.
8. The closed space infrared sensing head physiological monitoring system according to claim 7, wherein the cross-modal signal self-healing module, when monitoring the signal-to-noise ratio of the reference pulse wave signal in real time, specifically comprises: Setting a pulse wave quality evaluation time window, extracting the alternating current component amplitude and the direct current component amplitude of a reference pulse wave signal in the window in real time, and calculating the ratio of the alternating current component amplitude and the direct current component amplitude as a current signal quality index; Comparing the current signal quality index with a preset safety threshold, if the current quality index is continuous And if the calculation period is lower than the safety threshold, judging that the near infrared photoelectric pulse wave signal encounters serious shielding or severe motion interference, judging that data loss occurs by the system, and triggering a cross-mode interpolation compensation mechanism.
9. The closed space infrared sensing head physiological monitoring system according to claim 8, wherein the specific mathematical interpolation and curve smoothing steps of the preset cardiopulmonary coupling rhythm model adopted in the cross-modal interpolation compensation mechanism include: Extracting average heart rate in a historical normal time window before data loss occurs And synchronized average respiration rate Calculating individualized heart and lung reference coupling coefficient : Extracting the instantaneous frequency of the pure target respiratory rhythm signal continuously output by the thermal inertia separation processing module in the time segment of data missing Respiratory phase characteristics 。
10. The closed space infrared sensing head physiological monitoring system according to claim 9, wherein said cardiopulmonary reference coupling coefficient is utilized based on physiological modulation characteristics of respiratory sinus arrhythmia Estimating instantaneous heart rate node sequences within a missing time segment : Wherein, the A periodic adjustment function for following alternating respiratory inhalation and exhalation phases; taking real effective heart rate data before and after the data missing segment as boundary anchor points, and taking the estimated instantaneous heart rate node sequence And (3) taking the curve as an intermediate node, adopting a cubic spline interpolation algorithm to perform curve fitting, and outputting a physiological characteristic parameter curve which is continuous and smooth and has no obvious step.

Description

Closed space infrared induction head physiological monitoring system Technical Field The invention relates to the technical field of physiological characteristic non-sensing monitoring and multi-mode sensor information fusion, in particular to a closed space infrared sensing head physiological monitoring system. Background With the development of intelligent technologies of closed spaces such as intelligent cabins, hyperbaric oxygen cabins and noninductive sleep monitoring, non-contact physiological characteristic monitoring (such as heart rate and respiratory rhythm detection) is becoming an industrial research hotspot due to the advantage of no wearing burden. Currently, the mainstream non-contact monitoring system mainly relies on near-infrared imaging technology based on photoplethysmography (rpg) and far-infrared thermal imaging technology based on thermodynamic distribution. However, in the actual working condition of a complex enclosed space, the existing non-contact physiological monitoring system faces the following technical defects that are difficult to overcome: 1. In a closed environment where dynamic jolts exist or the target is allowed to freely move, the target object is extremely likely to undergo spatial displacement in the depth direction such as body tilting forward, body tilting backward and the like. Such physical displacement can result in severe amplitude modulation (i.e., motion artifacts) of the reflected light intensity received by the near infrared camera due to the strong attenuation of the illumination intensity with increasing propagation distance. In the prior art, blind source separation or smoothing algorithm in a pure software layer is mostly adopted for post digital filtering, so that the method is difficult to cope with distance mutation with larger amplitude, and real physiological pulsation high-frequency characteristics are easily filtered out together, so that serious distortion of heart rate extraction is caused. 2. Far infrared thermal imaging techniques are highly susceptible to rapid changes in ambient temperature within an enclosed space (e.g., vehicle portholes are exposed to heat or cabin air conditioning warms are blown directly) when extracting facial breathing rhythms. The overall baseline drift amplitude of ambient temperature often exceeds the weak alternating fluctuations in temperature produced by the human respiratory airflow in the subnasal region. The existing system lacks an effective separation mechanism for large-scale environmental thermal noise and tiny physiological thermal radiation, and respiratory characteristic signals are easily and thoroughly submerged by environmental background temperature drift. 3. The multi-mode sensor in the existing system is often in an isolated working state, and simple comparison is only carried out on a final result layer, so that a deep coupling and complementary mechanism of data of a bottom layer is lacked. When a signal fault occurs to a certain main force sensor (such as a near infrared component) due to extreme physical shielding or intense movement of a target, the system can only output intermittent physiological parameters or directly trigger false alarm shutdown, and the rigidity requirement on continuous monitoring data in a high-level safety scene can not be ensured. Therefore, a high robustness monitoring scheme capable of resisting earthquake from a physical underlayer, improving temperature drift of immune environment and having a cross-modal signal self-healing capability is needed. Disclosure of Invention In order to achieve the above purpose, the present invention provides the following technical solutions: The invention provides an infrared induction head physiological monitoring system of an enclosed space, which comprises a multi-mode detection acquisition module, a detection module and a detection module, wherein the multi-mode detection acquisition module is used for synchronously acquiring a three-dimensional depth displacement vector, a near infrared photoelectric pulse wave signal and a far infrared thermal radiation pattern sequence of a target face in the enclosed space; The optical path dynamic compensation module is connected with the multi-mode detection acquisition module and is used for dynamically adjusting the exposure parameters or digital gain of near infrared detection by taking the three-dimensional depth displacement vector as a compensation reference so as to output a reference pulse wave signal for eliminating motion artifacts; The thermal inertia separation processing module is used for constructing a time sequence dual-channel filter aiming at the far infrared thermal radiation diagram sequence based on the physical thermal inertia difference between the cabin environment background and the target physiological activity, subtracting and stripping the low-frequency environment thermal noise baseline from the total thermal radiation diagram, and extracting a pure target