CN114216564A - Intelligent infant body temperature detection method based on head multi-zone positioning - Google Patents
Intelligent infant body temperature detection method based on head multi-zone positioning Download PDFInfo
- Publication number
- CN114216564A CN114216564A CN202111422493.4A CN202111422493A CN114216564A CN 114216564 A CN114216564 A CN 114216564A CN 202111422493 A CN202111422493 A CN 202111422493A CN 114216564 A CN114216564 A CN 114216564A
- Authority
- CN
- China
- Prior art keywords
- infant
- temperature
- area
- temperature measurement
- head
- 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.)
- Pending
Links
- 230000036760 body temperature Effects 0.000 title claims abstract description 44
- 238000001514 detection method Methods 0.000 title claims abstract description 34
- 238000009529 body temperature measurement Methods 0.000 claims abstract description 96
- 230000036544 posture Effects 0.000 claims abstract description 51
- 238000004364 calculation method Methods 0.000 claims abstract description 11
- 210000003128 head Anatomy 0.000 claims description 63
- 210000001061 forehead Anatomy 0.000 claims description 55
- 238000000034 method Methods 0.000 claims description 26
- 210000004709 eyebrow Anatomy 0.000 claims description 24
- 238000004458 analytical method Methods 0.000 claims description 16
- 210000000624 ear auricle Anatomy 0.000 claims description 14
- 210000001331 nose Anatomy 0.000 claims description 12
- 210000001508 eye Anatomy 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 9
- 230000002650 habitual effect Effects 0.000 claims description 9
- 230000002123 temporal effect Effects 0.000 claims description 9
- 238000005259 measurement Methods 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 claims description 3
- 230000002035 prolonged effect Effects 0.000 claims description 3
- 238000011897 real-time detection Methods 0.000 abstract 1
- 210000001994 temporal artery Anatomy 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000035900 sweating Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 206010019196 Head injury Diseases 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000017531 blood circulation Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 210000005069 ears Anatomy 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002631 hypothermal effect Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 230000006996 mental state Effects 0.000 description 1
- QSHDDOUJBYECFT-UHFFFAOYSA-N mercury Chemical compound [Hg] QSHDDOUJBYECFT-UHFFFAOYSA-N 0.000 description 1
- 229910052753 mercury Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000024883 vasodilation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J5/0022—Radiation pyrometry, e.g. infrared or optical thermometry for sensing the radiation of moving bodies
- G01J5/0025—Living bodies
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/22—Status alarms responsive to presence or absence of persons
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/61—Control of cameras or camera modules based on recognised objects
- H04N23/611—Control of cameras or camera modules based on recognised objects where the recognised objects include parts of the human body
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/69—Control of means for changing angle of the field of view, e.g. optical zoom objectives or electronic zooming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to an infant intelligent body temperature detection method based on head multi-region positioning, which fully considers the changes of infant sleeping postures, firstly divides the infant sleeping postures into three conditions of supine, side lying and prone, and divides the infant head into three effective temperature measurement regions, and then adopts different temperature calculation methods for different sleeping postures, thereby accurately measuring the infant body temperature and meeting the requirements of guardians on real-time detection of the infant body temperature.
Description
Technical Field
The invention relates to the technical field of human body temperature measurement, in particular to an infant intelligent body temperature detection method based on head multi-region positioning.
Background
In recent years, intensive care and intensive care of children have become common knowledge of parents, so people pay more attention to the physical health of infants and children, and are willing to enhance the monitoring of the health condition of children by means of science and technology, wherein the most common way to pay attention to the body temperature of children is to know the body temperature of children. The most common previous body temperature detection methods are that the forehead is touched by hands and the body temperature is monitored by a thermometer, and particularly, parents frequently detect the body temperature of children when the mental state of the children is poor.
At present, the methods for measuring the temperature of individuals mainly comprise mercury thermometers, electronic thermometers, ear thermometer type thermometers, infrared thermometers and the like. In recent years, a non-contact rapid and accurate temperature measurement method is increasingly popular, and particularly for children, a forehead thermometer and a temporal artery thermometer without safety risk and discomfort are popular. A forehead thermometer, namely a handheld infrared induction thermometer, is a thermometer commonly used in hospitals and is used for measuring the temperature of the forehead of a human body. The temporal artery thermometer uses an infrared scanner to measure the temperature of the temporal artery, which is as accurate as the rectal temperature of a healthy individual. The temporal artery is located on the superficial surface of the two sides of the forehead, is tightly attached to the skin and penetrates through the forehead, is the most practical way for measuring the body temperature of a child under the age of 3 years old, is easy to measure even a fat person, and is convenient to operate, especially when the child falls asleep. The infrared temperature measurement has the advantages of fast reading, body temperature display within 1 second, no need of direct contact with human body, measurement for the baby during sleep, and no contact infection, and has the disadvantages that the measurement result is easily interfered by external factors, such as outdoor temperature, light, radiation and the like, so the measurement result can be more accurate only by correction, and in addition, the price is higher.
The thermometers mentioned above all require the parent to operate, which is feasible during the day and difficult during night sleep. The current childcare philosophy began to advocate children and parents to sleep in separate beds and became increasingly accepted by young parents, so that many infants sleep in beds different from the parent's bed, or even have their own separate bedroom. Therefore, the requirement of continuously monitoring the body temperature of the infant is generated, namely, the infant temperature measuring part needs to be positioned and the body temperature needs to be detected at intervals. Considering that the position and posture of an infant frequently change during sleeping, at present, infrared temperature measuring equipment is combined with a holder, the face and forehead of the infant are positioned based on face recognition, and then the angle of the temperature measuring equipment is adjusted to detect the temperature. However, because the temperature measuring device can only be fixed at a certain position beside the bed, and the infant often turns over and rolls when sleeping, and the posture is changeable, the forehead part for measuring the temperature is often not in the range of infrared temperature measurement sensing. As shown in fig. 3, in the posture of lying on the back, b lying on the front and c lying on the side, the temperature measuring device is usually fixed above the infant or on the side, and the forehead can be accurately positioned and measured only in the posture of lying on the back, a. Therefore, the existing cloud platform type body temperature detection equipment is used for measuring the forehead by default, and the problem of how to accurately measure the temperature of the infant under the condition that the posture of the infant is changed frequently is not fully considered.
Besides the forehead used for conventional non-contact temperature measurement, the temples on the forehead and behind the ears of infants are accurate temperature measurement parts. It has been mentioned that the temples are more accurate than the forehead, and the back of the ear is another interesting part. Measuring the temperature behind the ear eliminates the possibility of false hypothermia due to evaporative cooling of the skin surface due to sweating, which is not noticed by the forehead for many times. Perspiration causes the temperature in the area of the temperature measurement to decrease, while the area directly behind the earlobe has a high blood flow required for arterial measurement and is the least perspiration prone part of the body and at the same time easy to measure. Under the condition of forehead sweating and neck blood vessel complete relaxation, the back of the ear can be used as the only temperature measuring point. The back of the ear cannot be the only temperature measurement point without sweating, head trauma, or uncertainty in the neck whether there is vasodilation. If the forehead area is completely covered by the bandage, with burns or tears, an accurate temperature measurement can be obtained behind the earlobe.
In view of the above, there is a need for an improved and optimized method for detecting body temperature of infants.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide an infant intelligent body temperature detection method based on head multi-region positioning.
The technical scheme adopted by the invention for solving the problems is as follows: an infant intelligent body temperature detection method based on head multi-region positioning is implemented by adopting an infant intelligent body temperature detection system, the infant intelligent body temperature detection system comprises a receiving end, an intelligent analysis module, an infrared temperature measurement module, a camera module, a holder and a support frame, the holder is installed on the support frame, the intelligent analysis module, the infrared temperature measurement module and the camera module are integrally installed on the holder, the camera module is used for shooting infant videos, the intelligent analysis module is used for analyzing the infant videos and controlling the infrared temperature measurement module to measure the temperature according to analysis results, and the receiving end is used for receiving the infant videos and temperature measurement results.
The intelligent infant body temperature detection method based on head multi-zone positioning comprises the following steps:
initialization: the intelligent infant body temperature detection system is arranged above or laterally above an infant bed through a support frame, and the mounting position is ensured to identify and detect the effective temperature measurement part of an infant; setting an infant sleeping habitual area according to the size and the position of the crib, namely the position where parents are accustomed to putting an infant, and inputting the coordinate range into the infant intelligent body temperature detection system;
step S1: detecting the infant sleeping habitual area and confirming the position of the infant;
step S2: identifying an effective infant temperature measurement area and carrying out multi-area positioning:
the method comprises the following specific steps:
step S2.1: estimating the head posture of the infant through a Yolov3 model, and judging the sleeping posture of the infant, wherein the sleeping posture is set to have three conditions, namely supine, lateral lying and prone lying;
step S2.2: identifying and defining a forehead A area, a temple B area and a behind-the-ear C area based on the infant head characteristic points, wherein the forehead A area, the temple B area and the behind-the-ear C area are effective infant temperature measurement areas;
step S3: and carrying out multipoint temperature measurement calculation on the positioned temperature measurement areas: setting: for the supine posture, only the forehead area A can be completely identified, and temperature can be measured through the forehead area A and the temple area B; for the lateral lying posture, only the temple region B can be completely identified, and the temperature can be measured through the forehead region A, the temple region B and the retroauricular region C; for the prone posture, only the area C behind the ear can be completely identified, and the temperature can be measured through the area B of the temporal part and the area C behind the ear; in particular, if two temple B areas exist in the identified area at the same time, the temple B area with the larger area is selected as the thermometric temple B area.
The specific steps of carrying out multipoint temperature measurement calculation in a plurality of temperature measurement areas are as follows:
step S3.1: respectively carrying out five-point temperature measurement on the obtained forehead A area, the temple B area and the retroauricular C area, wherein the five points are respectively a region central point and temperature measurement points at positions half of the distance from the upper edge, the lower edge, the left edge and the right edge, and the temperature measurement obtained temperature TA, TB and TC are respectively the five-point temperature average values of the forehead A area, the temple B area and the retroauricular C area;
step S3.2: and calculating the temperature T0 by adopting different temperature calculation methods according to different sleeping postures:
(1) if the sleeping posture is judged to be supine, determining the temperature according to TA and TB, if | TA-TB | <0.3 ℃, T0 ═ TA + TB)/2, otherwise, T0 is equal to the higher temperature of TA and TB;
(2) if the sleeping posture is judged to be lying on side, measuring the temperature according to TA, TB and TC, if the temperature is | TA-TB | <0.3 ℃, T0 is TB, otherwise, T0 is equal to the higher temperature of the TA, the TB and the TC;
(3) if the sleeping posture is prone, measuring the temperature according to TB and TC, and if the absolute TB-TC is less than 0.3 ℃, determining (TB + TC)/2; otherwise T0 equals the higher of both TB, TC;
step S4: and (3) carrying out environmental compensation correction on the temperature measurement result: measuring an ambient temperature T1, and performing infrared temperature measurement compensation by adopting a compensation algorithm based on a BP neural network to obtain a final temperature measurement result T;
step S5: and processing the abnormal temperature.
Preferably, in step S1, the specific method for detecting and confirming the position of the infant is:
step S1.1: the camera module shoots the infant sleeping habitual area, and the intelligent analysis module carries out human body recognition through a Yolov3 model to detect whether an infant is in the area;
step S1.2: the infrared temperature measurement module detects the temperature of the identified human body and confirms whether the highest temperature is within a set temperature range, wherein the set temperature range is 34-42 ℃;
if the two are both in accordance, the infant is judged to be in the area, and the step S2 is carried out;
if the infant is judged not to be in the region, the shooting scanning range of the camera module is expanded to the range of 3 times of the length and the width, the step S1 is repeated, and if the infant is judged not to be in the expansion region again, an alarm is given out.
Preferably, in step S5, the method for processing the abnormal temperature includes:
if the final temperature measurement result T is not within the range of 34-42 ℃, discarding data and re-measuring, and if the re-measurement result is still not within the range of the set temperature, feeding back temperature errors;
if the final temperature measurement result T is within the range of 34-42 ℃ of the set temperature, but exceeds the alarm threshold, the alarm threshold is set to be 37 ℃, the guardian is alarmed, and meanwhile, the temperature measurement time interval is reduced to half of the original temperature measurement time interval.
Preferably, in the step S2.2, the infant head feature points based on which are set are nose tip, canthus, upper eyebrow edge, earlobe, hairline and forehead edge, wherein the feature points are stable feature points except forehead edge and hairline;
for the supine posture, the specific method for identifying and dividing the effective temperature measuring area of the infant comprises the following steps:
(1) identifying two canthi and nose by using opencv library functions, and calculating a head central axis parallel to the nose;
(2) establishing a face region coordinate system, projecting the face into an xy two-dimensional plane coordinate system, wherein the x axis corresponds to a connecting line of two canthi, the y axis corresponds to a central axis of the head, and the intersection point of the two axes is an original point;
(3) dividing the positions of two canthi into x1 and x2, and marking a line parallel to the central axis of the head along the y direction to form AY1 and AY2 sides;
(4) the upper edges of the eyebrows at the two sides form an AX1 edge along the connecting line;
(5) the minimum value of the ordinate of the intersection point of AY1 and the vertex profile, AY2 and the vertex profile, along the three points of the hairline, is y2, and the line drawn parallel to the x axis along y2 is an AX2 side;
(6) the region formed by connecting four sides of AX1, AX2, AY1 and AY2 is a forehead A region;
(7) the head side profile is extended By AX1 and AX2 along the x axis, the extended line part is rotated downwards By 15 degrees, the intersection points of the extended line part and the head side profile are By1 and By2, two adjacent vertexes of the A area and By1 and By2 form a quadrangle, namely the B area of the temporal part. For the side lying posture, the specific method for identifying and dividing the effective infant temperature measurement area comprises the following steps:
(1) eye corners, noses and ear regions can be identified by using the opencv library function, and eye corners, eyebrows and auricle edge lines are determined by using feature identification;
(2) drawing a tangent BX1 from the connecting point of the auricle and the head side to the eyebrow on the same side, drawing a vertical line BY2 at a position which is half of the tangent point, and drawing a vertical line BY1 from the canthus on the same side and perpendicular to the midline of the eyes;
(3) drawing a line BX2 parallel to BX1 at the intersection of BY1 and the top hairline of the head; if the hairline cannot be identified, then line BX2 is drawn parallel to BX1 at a distance from the midpoint of the top of the head on the eyebrow;
(4) the region formed BY intersecting four sides of BX1, BX2, BY1 and BY2 is the temple region B;
(5) on the side of the temporal region B, the BX1 and the BX2 are extended towards the forehead, and a trapezoidal region formed by intersecting with the contour edge of the forehead is the forehead region A;
(6) the depression below the back of the earlobe is the area C behind the ear, the height is half of the size of the auricle, and the width is set to be half of the height. For the prone posture, the specific method for identifying and dividing the effective infant temperature measurement area comprises the following steps:
(1) the ear region can be identified by using an opencv library function, and the positions of the canthus, the eyebrow edge, the auricle and the earlobe are determined by using feature identification;
(2) the sunken department in behind the earlobe below is behind the ear C district, and height is half auricle size, and the width is set for half of height, and behind the ear C district's outward flange is the circular arc line.
(3) Marking a line BX1 from the connecting point of the auricle and the head side to the eyebrow edge on the same side, and drawing a vertical line BY2 at a position which is half of the distance from the eyebrow edge;
(4) the areas surrounded BY the prolonged BX1 and BY2, the outer edges of the head and the hairline are the temples B areas.
Compared with the prior art, the invention has the following advantages and effects: the infant intelligent body temperature detection method based on head multi-region positioning fully considers the change of the infant sleeping posture, divides the infant sleeping posture into three conditions of supine, lateral and prone positions, divides the infant head into three effective temperature measurement regions, adopts different temperature calculation methods for different sleeping postures, can accurately measure the body temperature of the infant, and meets the requirement of a guardian on detecting the body temperature of the infant in real time.
Drawings
In order to illustrate the embodiments of the present invention or the solutions in the prior art more clearly, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without inventive effort.
Fig. 1 is a flowchart of an infant intelligent body temperature detection method based on head multi-zone positioning in this embodiment.
Fig. 2 is a schematic structural diagram of the infant intelligent body temperature detection system in this embodiment.
Fig. 3 shows three sleeping positions of the infant, wherein a is supine, b is prone, and c is lateral.
FIG. 4 illustrates the effective temperature measurement area in the supine sleeping position.
FIG. 5 illustrates the process of locating the effective temperature measurement area in the supine sleeping position.
FIG. 6 shows the effective temperature measurement area in the side sleeping position.
FIG. 7 illustrates the process of locating the effective temperature measurement area when sleeping on one side.
Fig. 8 illustrates the effective temperature measurement area in the prone sleeping position.
FIG. 9 illustrates the process of locating the effective temperature measurement area when sleeping prone.
Fig. 10 shows the effective temperature measurement areas corresponding to the three sleeping postures.
Description of reference numerals: a receiving end 1; an infrared temperature measurement module 2; a camera module 3; a holder 4 and a support frame 5; and an intelligent analysis module 6.
Detailed Description
The present invention will be described in further detail below by way of examples with reference to the accompanying drawings, which are illustrative of the present invention and are not to be construed as limiting the present invention.
Examples are given.
See fig. 1-10.
The embodiment of the invention discloses an infant intelligent body temperature detection method based on head multi-region positioning, which is implemented by adopting an infant intelligent body temperature detection system, wherein the infant intelligent body temperature detection system comprises a receiving end 1, an intelligent analysis module 6, an infrared temperature measurement module 2, a camera module 3, a cradle head 4 and a support frame 5, the cradle head 4 is installed on the support frame 5, the intelligent analysis module 6, the infrared temperature measurement module 2 and the camera module are integrally installed on the cradle head 4, and the cradle head 4 can adjust the angles of a camera lens and the temperature measurement lens, and the detection method specifically refers to the prior art. The camera module 3 is used for shooting infant videos. The intelligent analysis module 6 is integrated with an intelligent algorithm related to the detection method, plays a brain role of the detection system, and is used for analyzing the infant video and controlling the infrared temperature measurement module 2 to measure the temperature according to an analysis result, in addition, the cradle head 4 is also provided with an alarm module and a wireless transmission module, alarm information can be transmitted to a receiving end through the wireless transmission module, and the receiving end 1 can be equipment such as a guardian mobile phone and the like and is used for receiving the infant video, the temperature measurement result, the alarm information and the like.
In this embodiment, initialization is performed before temperature measurement: the intelligent infant body temperature detection system is arranged above or laterally above an infant bed through a support frame 5, the mounting position is ensured to be capable of identifying and detecting an effective temperature measurement part of an infant, and in order to ensure the accuracy of a temperature measurement method and the identifiability of a human face, the distance from an infrared temperature measurement module 2 to the head of the infant is controlled within the range of 10cm to 100 cm; setting an infant sleeping habitual area according to the size and the position of the crib, namely the position where parents are used to place the infant, and inputting the coordinate range into the infant intelligent body temperature detection system.
The intelligent infant body temperature detection method based on head multi-zone positioning comprises the following steps:
step S1: the method for detecting the infant sleeping habitual area and confirming the infant position specifically comprises the following steps:
step S1.1: the camera module shoots the infant sleeping habitual area, and the intelligent analysis module carries out human body recognition through a Yolov3 model to detect whether an infant is in the area;
step S1.2: the infrared temperature measurement module detects the temperature of the identified human body, and confirms whether the highest temperature is within a set temperature range, wherein the set temperature range is 34-42 ℃, and the set temperature range can avoid mistakenly identifying the doll;
if the two are both in accordance, the infant is judged to be in the area, and the step S2 is carried out;
if the infant is judged not to be in the region, the shooting scanning range of the camera module is expanded to the range of 3 times of the length and the width, the step S1 is repeated, and if the infant is judged not to be in the expansion region again, an alarm is given out.
Step S2: identifying an effective infant temperature measurement area and carrying out multi-area positioning:
the method comprises the following specific steps:
step S2.1: estimating the head posture of the infant through a Yolov3 model, judging the sleeping posture of the infant, wherein the sleeping posture is set to be in three conditions, namely supine, lateral and prone (see figure 3), wherein the recognition training of the head posture can be based on a front face and a side face atlas provided by a FERET, LFW and other public databases, and the picture recognized as the head is the back side or the back side if the picture is not the front side and the side;
step S2.2: identifying and defining a forehead A area, a temple B area and a behind-the-ear C area based on infant head feature points, wherein the forehead A area, the temple B area and the behind-the-ear C area are effective infant temperature measurement areas, and setting the infant head feature points to be stable feature points except for forehead edges and hairline, wherein the nose tip, eye corners, upper eyebrow edges, earlobes, hairline and forehead edges are based on the infant head feature points;
referring to fig. 4 and 5, for the supine posture, the specific method for identifying and dividing the effective temperature measurement area of the infant is as follows:
(1) identifying two canthi and nose by using opencv library functions, and calculating a head central axis parallel to the nose;
(2) establishing a face region coordinate system, projecting the face into an xy two-dimensional plane coordinate system, wherein the x axis corresponds to a connecting line of two canthi, the y axis corresponds to a central axis of the head, and the intersection point of the two axes is an original point;
(3) dividing the positions of two canthi into x1 and x2, and marking a line parallel to the central axis of the head along the y direction to form AY1 and AY2 sides;
(4) the upper edges of the eyebrows at the two sides form an AX1 edge along the connecting line;
(5) the minimum value of the ordinate of the intersection point of AY1 and the vertex profile, AY2 and the vertex profile, along the three points of the hairline, is y2, and the line drawn parallel to the x axis along y2 is an AX2 side;
(6) the region formed by connecting four sides of AX1, AX2, AY1 and AY2 is a forehead A region;
(7) extending AX1 and AX2 along the side profile of the head in the x-axis direction, rotating the extension line part downwards By 15 degrees, wherein the intersection points of the extension line part and the side profile of the head are By1 and By2, and two adjacent vertexes of the area A, By1 and By2 form a quadrangle, namely the area B of the temporal part;
referring to fig. 6 and 7, for the lateral position, the specific method for identifying and dividing the effective infant temperature measurement area is as follows:
(1) eye corners, noses and ear regions can be identified by using the opencv library function, and eye corners, eyebrows and auricle edge lines are determined by using feature identification;
(2) drawing a tangent BX1 from the connecting point of the auricle and the head side to the eyebrow on the same side, drawing a vertical line BY2 at a position which is half of the tangent point, and drawing a vertical line BY1 from the canthus on the same side and perpendicular to the midline of the eyes;
(3) drawing a line BX2 parallel to BX1 at the intersection of BY1 and the top hairline of the head; if the hairline cannot be identified, then line BX2 is drawn parallel to BX1 at a distance from the midpoint of the top of the head on the eyebrow;
(4) the region formed BY intersecting four sides of BX1, BX2, BY1 and BY2 is the temple region B;
(5) on the side of the temporal region B, the BX1 and the BX2 are extended towards the forehead, and a trapezoidal region formed by intersecting with the contour edge of the forehead is the forehead region A;
(6) the depression behind the earlobe is a region C behind the ear, the height is half of the size of the auricle, and the width is half of the height;
referring to fig. 8 and 9, for the prone posture, the specific method for identifying and dividing the effective infant temperature measurement area is as follows:
(1) the ear region can be identified by using an opencv library function, and the positions of the canthus, the eyebrow edge, the auricle and the earlobe are determined by using feature identification;
(2) the sunken department in behind the earlobe below is behind the ear C district, and height is half auricle size, and the width is set for half of height, and behind the ear C district's outward flange is the circular arc line.
(3) Marking a line BX1 from the connecting point of the auricle and the head side to the eyebrow edge on the same side, and drawing a vertical line BY2 at a position which is half of the distance from the eyebrow edge;
(4) the areas surrounded BY the prolonged BX1 and BY2, the outer edges of the head and the hairline are the temples B areas.
Step S3: and carrying out multipoint temperature measurement calculation on the positioned temperature measurement areas: by the dividing method of the three regions described in step S2, it is possible to set: for the supine posture, only the forehead area A can be completely identified, and temperature can be measured through the forehead area A and the temple area B; for the lateral lying posture, only the temple region B can be completely identified, and the temperature can be measured through the forehead region A, the temple region B and the retroauricular region C; for prone posture, only the area C behind the ear can be identified completely, and the temperature can be measured through the area B of the temporal part and the area C behind the ear, see fig. 10;
the specific steps of carrying out multipoint temperature measurement calculation in a plurality of temperature measurement areas are as follows:
step S3.1: respectively carrying out five-point temperature measurement on the obtained forehead A area, the temple B area and the retroauricular C area, wherein the five points are respectively a region central point and temperature measurement points at positions half of the distance from the upper edge, the lower edge, the left edge and the right edge, and the temperature measurement obtained temperature TA, TB and TC are respectively the five-point temperature average values of the forehead A area, the temple B area and the retroauricular C area;
step S3.2: and calculating the temperature T0 by adopting different temperature calculation methods according to different sleeping postures:
(1) if the sleeping posture is judged to be supine, determining the temperature according to TA and TB, if | TA-TB | <0.3 ℃, T0 ═ TA + TB)/2, otherwise, T0 is equal to the higher temperature of TA and TB;
(2) if the sleeping posture is judged to be lying on side, measuring the temperature according to TA, TB and TC, if the temperature is | TA-TB | <0.3 ℃, T0 is TB, otherwise, T0 is equal to the higher temperature of the TA, the TB and the TC;
(3) if the sleeping posture is prone, measuring the temperature according to TB and TC, and if the absolute TB-TC is less than 0.3 ℃, determining (TB + TC)/2; otherwise T0 equals the higher of both TB, TC.
Step S4: and (3) carrying out environmental compensation correction on the temperature measurement result: measuring the environment temperature T1, performing infrared temperature measurement compensation by using a compensation algorithm based on a BP neural network to obtain a final temperature measurement result T, and optimally controlling the environment temperature within the range of 15-31 ℃ to ensure the compensation and correction effect of the environment temperature.
Step S5: the abnormal temperature is processed, and the specific processing method comprises the following steps:
if the final temperature measurement result T is not within the range of the set temperature of 34-42 ℃, data are abandoned for re-measurement, the re-measurement result is still not within the range of the set temperature, and the feedback temperature is wrong.
If the final temperature measurement result T is within the range of 34-42 ℃ of the set temperature, but exceeds the alarm threshold value, the alarm threshold value can be set to 37 ℃ for example, and the guardian is alarmed, and meanwhile, the temperature measurement time interval is reduced to half of the original temperature measurement time interval.
In addition, it should be noted that the specific embodiments described in the present specification may be different in the components, the shapes of the components, the names of the components, and the like, and the above description is only an illustration of the structure of the present invention. All equivalent or simple changes in the structure, characteristics and principles of the invention are included in the protection scope of the patent. Various modifications, additions and substitutions for the specific embodiments described may be made by those skilled in the art without departing from the scope of the invention as defined in the accompanying claims.
Claims (4)
1. An infant intelligent body temperature detection method based on head multi-region positioning is implemented by adopting an infant intelligent body temperature detection system, the infant intelligent body temperature detection system comprises a receiving end, an intelligent analysis module, an infrared temperature measurement module, a camera module, a holder and a support frame, the holder is installed on the support frame, the intelligent analysis module, the infrared temperature measurement module and the camera module are integrally installed on the holder, the camera module is used for shooting infant videos, the intelligent analysis module is used for analyzing the infant videos and controlling the infrared temperature measurement module to measure the temperature according to analysis results, and the receiving end is used for receiving the infant videos and temperature measurement results;
the method is characterized in that: the intelligent infant body temperature detection method based on head multi-zone positioning comprises the following steps:
initialization: the intelligent infant body temperature detection system is arranged above or laterally above an infant bed through a support frame, and the mounting position is ensured to identify and detect the effective temperature measurement part of an infant; setting an infant sleeping habitual area according to the size and the position of the crib, namely the position where parents are accustomed to putting an infant, and inputting the coordinate range into the infant intelligent body temperature detection system;
step S1: detecting the infant sleeping habitual area and confirming the position of the infant;
step S2: identifying an effective infant temperature measurement area and carrying out multi-area positioning:
the method comprises the following specific steps:
step S2.1: estimating the head posture of the infant through a Yolov3 model, and judging the sleeping posture of the infant, wherein the sleeping posture is set to have three conditions, namely supine, lateral lying and prone lying;
step S2.2: identifying and defining a forehead A area, a temple B area and a behind-the-ear C area based on the infant head characteristic points, wherein the forehead A area, the temple B area and the behind-the-ear C area are effective infant temperature measurement areas;
step S3: and carrying out multipoint temperature measurement calculation on the positioned temperature measurement areas:
setting: for the supine posture, only the forehead area A can be completely identified, and temperature can be measured through the forehead area A and the temple area B; for the lateral lying posture, only the temple region B can be completely identified, and the temperature can be measured through the forehead region A, the temple region B and the retroauricular region C; for the prone posture, only the area C behind the ear can be completely identified, and the temperature can be measured through the area B of the temporal part and the area C behind the ear;
the specific steps of carrying out multipoint temperature measurement calculation in a plurality of temperature measurement areas are as follows:
step S3.1: respectively carrying out five-point temperature measurement on the obtained forehead A area, the temple B area and the retroauricular C area, wherein the five points are respectively a region central point and temperature measurement points at positions half of the distance from the upper edge, the lower edge, the left edge and the right edge, and the temperature measurement obtained temperature TA, TB and TC are respectively the five-point temperature average values of the forehead A area, the temple B area and the retroauricular C area;
step S3.2: and calculating the temperature T0 by adopting different temperature calculation methods according to different sleeping postures:
(1) if the sleeping posture is judged to be supine, determining the temperature according to TA and TB, if | TA-TB | <0.3 ℃, T0 ═ TA + TB)/2, otherwise, T0 is equal to the higher temperature of TA and TB;
(2) if the sleeping posture is judged to be lying on side, measuring the temperature according to TA, TB and TC, if the temperature is | TA-TB | <0.3 ℃, T0 is TB, otherwise, T0 is equal to the higher temperature of the TA, the TB and the TC;
(3) if the sleeping posture is prone, measuring the temperature according to TB and TC, and if the absolute TB-TC is less than 0.3 ℃, determining (TB + TC)/2; otherwise T0 equals the higher of both TB, TC;
step S4: and (3) carrying out environmental compensation correction on the temperature measurement result: measuring an ambient temperature T1, and performing infrared temperature measurement compensation by adopting a compensation algorithm based on a BP neural network to obtain a final temperature measurement result T;
step S5: and processing the abnormal temperature.
2. The intelligent infant body temperature detection method based on multi-zone head positioning according to claim 1, characterized in that: in step S1, the specific method for detecting and confirming the position of the infant is:
step S1.1: the camera module shoots the infant sleeping habitual area, and the intelligent analysis module carries out human body recognition through a Yolov3 model to detect whether an infant is in the area;
step S1.2: the infrared temperature measurement module detects the temperature of the identified human body and confirms whether the highest temperature is within a set temperature range, wherein the set temperature range is 34-42 ℃;
if the two are both in accordance, the infant is judged to be in the area, and the step S2 is carried out;
if the infant is judged not to be in the region, the shooting scanning range of the camera module is expanded to the range of 3 times of the length and the width, the step S1 is repeated, and if the infant is judged not to be in the expansion region again, an alarm is given out.
3. The intelligent infant body temperature detection method based on multi-zone head positioning according to claim 2, characterized in that: in step S5, the method for processing the abnormal temperature includes:
if the final temperature measurement result T is not within the range of 34-42 ℃, discarding data and re-measuring, and if the re-measurement result is still not within the range of the set temperature, feeding back temperature errors;
if the final temperature measurement result T is within the range of 34-42 ℃ of the set temperature, but exceeds the alarm threshold, the alarm threshold is set to be 37 ℃, the guardian is alarmed, and meanwhile, the temperature measurement time interval is reduced to half of the original temperature measurement time interval.
4. The intelligent infant body temperature detection method based on multi-zone head positioning according to claim 1, characterized in that: in the step S2.2, the infant head feature points based on which are set are nose tip, canthus, upper eyebrow edge, earlobe, hairline and forehead edge, wherein the feature points are stable feature points except forehead edge and hairline;
for the supine posture, the specific method for identifying and dividing the effective temperature measuring area of the infant comprises the following steps:
(1) identifying two canthi and nose by using opencv library functions, and calculating a head central axis parallel to the nose;
(2) establishing a face region coordinate system, projecting the face into an xy two-dimensional plane coordinate system, wherein the x axis corresponds to a connecting line of two canthi, the y axis corresponds to a central axis of the head, and the intersection point of the two axes is an original point;
(3) dividing the positions of two canthi into x1 and x2, and marking a line parallel to the central axis of the head along the y direction to form AY1 and AY2 sides;
(4) the upper edges of the eyebrows at the two sides form an AX1 edge along the connecting line;
(5) the minimum value of the ordinate of the intersection point of AY1 and the vertex profile, AY2 and the vertex profile, along the three points of the hairline, is y2, and the line drawn parallel to the x axis along y2 is an AX2 side;
(6) the region formed by connecting four sides of AX1, AX2, AY1 and AY2 is a forehead A region;
(7) extending AX1 and AX2 along the side profile of the head in the x-axis direction, rotating the extension line part downwards By 15 degrees, wherein the intersection points of the extension line part and the side profile of the head are By1 and By2, and two adjacent vertexes of the area A, By1 and By2 form a quadrangle, namely the area B of the temporal part; for the side lying posture, the specific method for identifying and dividing the effective infant temperature measurement area comprises the following steps:
(1) eye corners, noses and ear regions can be identified by using the opencv library function, and eye corners, eyebrows and auricle edge lines are determined by using feature identification;
(2) drawing a tangent BX1 from the connecting point of the auricle and the head side to the eyebrow on the same side, drawing a vertical line BY2 at a position which is half of the tangent point, and drawing a vertical line BY1 from the canthus on the same side and perpendicular to the midline of the eyes;
(3) drawing a line BX2 parallel to BX1 at the intersection of BY1 and the top hairline of the head; if the hairline cannot be identified, then line BX2 is drawn parallel to BX1 at a distance from the midpoint of the top of the head on the eyebrow;
(4) the region formed BY intersecting four sides of BX1, BX2, BY1 and BY2 is the temple region B;
(5) on the side of the temporal region B, the BX1 and the BX2 are extended towards the forehead, and a trapezoidal region formed by intersecting with the contour edge of the forehead is the forehead region A;
(6) the depression behind the earlobe is a region C behind the ear, the height is half of the size of the auricle, and the width is half of the height; for the prone posture, the specific method for identifying and dividing the effective infant temperature measurement area comprises the following steps:
(1) the ear region can be identified by using an opencv library function, and the positions of the canthus, the eyebrow edge, the auricle and the earlobe are determined by using feature identification;
(2) the sunken department in behind the earlobe below is behind the ear C district, and height is half auricle size, and the width is set for half of height, and behind the ear C district's outward flange is the circular arc line.
(3) Marking a line BX1 from the connecting point of the auricle and the head side to the eyebrow edge on the same side, and drawing a vertical line BY2 at a position which is half of the distance from the eyebrow edge;
(4) the areas surrounded BY the prolonged BX1 and BY2, the outer edges of the head and the hairline are the temples B areas.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111422493.4A CN114216564A (en) | 2021-11-26 | 2021-11-26 | Intelligent infant body temperature detection method based on head multi-zone positioning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111422493.4A CN114216564A (en) | 2021-11-26 | 2021-11-26 | Intelligent infant body temperature detection method based on head multi-zone positioning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114216564A true CN114216564A (en) | 2022-03-22 |
Family
ID=80698530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111422493.4A Pending CN114216564A (en) | 2021-11-26 | 2021-11-26 | Intelligent infant body temperature detection method based on head multi-zone positioning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114216564A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114699046A (en) * | 2022-04-25 | 2022-07-05 | 深圳市华屹医疗科技有限公司 | Sleep monitoring method, monitor and monitoring system |
| CN117268559A (en) * | 2023-10-25 | 2023-12-22 | 武汉星巡智能科技有限公司 | Multi-mode infant abnormal body temperature detection method, device, equipment and medium |
| CN117690159A (en) * | 2023-12-07 | 2024-03-12 | 武汉星巡智能科技有限公司 | Infant groveling and sleeping monitoring method, device and equipment based on multi-mode data fusion |
-
2021
- 2021-11-26 CN CN202111422493.4A patent/CN114216564A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114699046A (en) * | 2022-04-25 | 2022-07-05 | 深圳市华屹医疗科技有限公司 | Sleep monitoring method, monitor and monitoring system |
| CN117268559A (en) * | 2023-10-25 | 2023-12-22 | 武汉星巡智能科技有限公司 | Multi-mode infant abnormal body temperature detection method, device, equipment and medium |
| CN117268559B (en) * | 2023-10-25 | 2024-05-07 | 武汉星巡智能科技有限公司 | Multi-mode infant abnormal body temperature detection method, device, equipment and medium |
| CN117690159A (en) * | 2023-12-07 | 2024-03-12 | 武汉星巡智能科技有限公司 | Infant groveling and sleeping monitoring method, device and equipment based on multi-mode data fusion |
| CN117690159B (en) * | 2023-12-07 | 2024-06-11 | 武汉星巡智能科技有限公司 | Infant groveling and sleeping monitoring method, device and equipment based on multi-mode data fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114216564A (en) | Intelligent infant body temperature detection method based on head multi-zone positioning | |
| US11986273B2 (en) | Detecting alcohol intoxication from video images | |
| WO2017099527A1 (en) | Technique for controlling equipment based on biometric information | |
| US11141090B2 (en) | Human-state estimating method and human-state estimating system | |
| US10638938B1 (en) | Eyeglasses to detect abnormal medical events including stroke and migraine | |
| CN108992079A (en) | A kind of Infant behavior monitoring method based on emotion recognition and Application on Voiceprint Recognition | |
| GB2598245A (en) | Clip-on device with inward-facing visible-light camera | |
| KR20160068631A (en) | Method for measuring movement of cervical vertebra, device thereof and wearable apparatus | |
| CN108986121A (en) | Baby sleep monitoring device and monitoring method | |
| TW201530480A (en) | Intelligent monitoring system | |
| CN112101080A (en) | Identification system using thermal sensors | |
| CN111214211A (en) | Sleep monitoring method and device and intelligent bed | |
| US10959646B2 (en) | Image detection method and image detection device for determining position of user | |
| EP3393346B1 (en) | Device, method and computer program product for continuous monitoring of vital signs | |
| KR101851099B1 (en) | Method and System for Preventing Software-based Forward Head Posture Using Image Processing and Machine Learning | |
| CN110458010A (en) | A kind of intelligent desk lamp, a kind of sitting posture detecting method and a kind of electronic equipment | |
| CN111248904A (en) | Infant nursing device and method thereof | |
| CN117268559B (en) | Multi-mode infant abnormal body temperature detection method, device, equipment and medium | |
| CN113017581A (en) | Plastic button with health condition monitoring function | |
| CN217283049U (en) | Body temperature detection type intelligent baby monitor | |
| Rajesh et al. | Baby monitoring system using wireless sensor networks | |
| KR102150623B1 (en) | Method and system for analyzing baby behavior | |
| Wipulasundara | A novel concept for remotely monitoring babies | |
| KR20200092063A (en) | Smart safety cradle | |
| CN109188723A (en) | A kind of spectacle-frame and myopia prevention system based on flight time distance detection device |
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 |