Veterinary non-contact infrared body temperature and respiratory rate synchronous monitor

Abstract

The present application relates to the technical field of body temperature and respiratory rate synchronous monitoring, in particular to a non-contact infrared body temperature and respiratory rate synchronous monitor for veterinarians. The present application obtains the respiratory rate and body temperature data of the current monitoring target through the data acquisition module; filters out the abnormal time and constructs the abnormal period through the body temperature analysis module; further obtains the temperature rise contribution degree of each temperature measurement point in the abnormal period; further filters out the temperature rise center and temperature rise area according to the numerical value of the temperature rise contribution degree and the spatial distribution of the corresponding temperature measurement point; analyzes the consistency of the temperature rise process in time and space between the abnormal periods through the respiratory compensation module, and obtains the consistent weight of each temperature rise area; finally, the final compensation module is used to obtain the temperature area to be compensated, improve the body temperature abnormal threshold of the temperature area to be compensated, solve the physiological temperature rise misjudgment problem, and improve the monitoring accuracy and robustness of body temperature and respiratory rate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of synchronous monitoring of body temperature and respiratory frequency, in particular to a non-contact infrared body temperature and respiratory frequency synchronous monitor for veterinarians. BACKGROUND

[0002] In the field of veterinary medicine, synchronous monitoring of body temperature and respiratory frequency, the correlation changes of the two parameters are high specificity indicators for evaluating the health status of animals, which are more valuable than single parameter monitoring, and can help veterinarians more accurately analyze the physiological state of animals.

[0003] For some animals, there are local heat source interference in the body temperature monitoring scene, such as muscle heat production after horse exercise and rumen temperature rise after ruminant feeding, which may cause the collected body temperature to be too high, and when the temperature in the animal body rises, the animal often speeds up the respiratory frequency to help dissipate heat. In the prior art, the normal range of body temperature and respiratory frequency (i.e. setting upper and lower thresholds) is often set to monitor whether the relevant data of the animal is abnormal by the value, but for the above-mentioned situation, it is easy to cause misjudgment. SUMMARY

[0004] In order to solve the technical problem of inaccurate abnormal identification of temperature in a specific area of an animal, the purpose of the present application is to provide a non-contact infrared body temperature and respiratory frequency synchronous monitor for veterinarians, and the technical solution adopted is as follows:

[0005] The data acquisition module acquires the respiratory frequency of the current monitoring target and the body temperature of each temperature measuring point.

[0006] The body temperature analysis module filters out abnormal time according to the temperature difference of each temperature measuring point at the same time; acquires endpoints by cutting the nearest preset number of continuous non-abnormal time on both sides of the time domain at the abnormal time, and constructs an abnormal period; acquires the temperature rise contribution degree of each temperature measuring point according to the rising degree of the temperature of each temperature measuring point at all abnormal times in the abnormal period; and filters out the temperature rise center and the temperature rise area according to the value of the temperature rise contribution degree and the spatial distribution of the corresponding temperature measuring point.

[0007] The respiratory compensation module analyzes the consistency of the temperature rise process between the abnormal periods in time and space, and acquires the consistent weight of each temperature rise area.

[0008] The final compensation module acquires the temperature compensation area according to the temperature rise contribution degree of each temperature measuring point in all abnormal periods and the consistent weight, and increases the body temperature abnormal threshold of the temperature compensation area.

[0009] Further, the method for acquiring the abnormal time comprises:

[0010] According to the variance of the body temperature of each temperature measuring point and the high amount of the body temperature higher than the average body temperature, a temperature difference coefficient of each time point is obtained;

[0011] An abnormal time point of the body temperature is screened out based on the temperature difference coefficient.

[0012] Further, the method for screening out the abnormal time point of the body temperature based on the temperature difference coefficient comprises:

[0013] A time point with the temperature difference coefficient greater than a preset temperature difference threshold value is marked as an abnormal time point.

[0014] Further, the method for obtaining the temperature rise contribution degree comprises:

[0015] In the abnormal time period, an average body temperature of a preset number of non-abnormal time points with the smallest time sequence of all temperature measuring points is obtained as a reference body temperature; and a temperature rise contribution degree of each temperature measuring point is obtained according to the extent of the body temperature of each temperature measuring point at all abnormal time points being greater than the reference body temperature.

[0016] Further, the method for screening out the temperature rise center and the temperature rise region comprises:

[0017] A temperature measuring point with the temperature rise contribution degree greater than a preset first temperature rise threshold value and greater than the temperature rise contribution degrees of all adjacent temperature measuring points is marked as a temperature rise center; a temperature measuring point with the temperature rise contribution degree greater than a preset second temperature rise threshold value is marked as a candidate temperature measuring point; adjacent candidate temperature measuring points are included in an initial temperature rise region with the temperature rise center as a starting point; adjacent candidate temperature measuring points of edge temperature measuring points of the initial temperature rise region are included; the initial temperature rise region is updated; the candidate temperature measuring points are repeatedly included until all edge temperature measuring points of the latest initial temperature rise region do not have adjacent candidate temperature measuring points that are not included; the latest initial temperature rise region is taken as a temperature rise region; the preset first temperature rise threshold value is greater than the preset second temperature rise threshold value.

[0018] Further, the method for obtaining the consistent weight comprises:

[0019] A temperature rise region with the largest number of overlapping temperature measuring points in other abnormal time periods is recorded as a corresponding region of the temperature rise region in the other abnormal time periods;

[0020] In each abnormal time period in which the temperature rise region and the corresponding region are located, a time consistency coefficient is obtained by analyzing the consistency of the change lag of the respiratory frequency compared with the change of the body temperature of the temperature rise center;

[0021] A space consistency coefficient is obtained according to the distribution discrete feature of the temperature rise center of each temperature rise region and the corresponding region;

[0022] According to the overlapping degree between each temperature rising region and its corresponding region, the presence rate of all abnormal time periods, the time consistency coefficient and the space consistency coefficient, a consistency weight of each temperature rising region is obtained.

[0023] Further, the method for obtaining the time consistency coefficient comprises:

[0024] In each abnormal time period to which any temperature rising region and its corresponding region belong, the time difference of three groups of key points corresponding to the respiratory frequency and the body temperature of the temperature rising center before rising, at the maximum point and after falling is obtained; according to the consistency of the time difference of each key point in all abnormal time periods, a time consistency coefficient is obtained.

[0025] Further, the method for obtaining the space consistency coefficient comprises:

[0026] The reciprocal of the minimum circumscribed circle area of the temperature rising center of each temperature rising region and its corresponding region is taken as the space consistency coefficient.

[0027] Further, the method for obtaining the temperature region to be compensated comprises:

[0028] The temperature rising contribution degree and the consistency weight of each temperature measuring point in all abnormal time periods are fused to obtain a compensation factor of each temperature measuring point; a region composed of temperature measuring points corresponding to all compensation factors greater than a preset compensation threshold is taken as a temperature region to be compensated.

[0029] Further, the method for increasing the body temperature abnormal threshold of the temperature region to be compensated comprises:

[0030] 1.05 times of the historical highest body temperature of the temperature region to be compensated is taken as the corresponding body temperature abnormal threshold.

[0031] The present application has the following beneficial effects:

[0032] The application firstly acquires the respiratory frequency and body temperature data of the current monitoring target, provides an analysis basis, screens out abnormal moments and constructs abnormal time periods through the body temperature analysis module, screens out moments and time periods that may exist physiological metabolism and other special temperature rising conditions, further acquires the temperature rising contribution degree of each temperature measuring point in the abnormal time period, represents the temperature rising highlight degree of the temperature measuring point, and is convenient for subsequent positioning and extraction of the temperature compensation area, further utilizes the continuity of the local body temperature rising of the animal in space, analyzes the value of the temperature rising contribution degree and the spatial distribution of the corresponding temperature measuring point, screens out the temperature rising center and temperature rising area, and lays a foundation for subsequent positioning of the temperature compensation area, further analyzes the consistency of the temperature rising process between the abnormal time periods in time and space, analyzes the physiological temperature rising rule of the animal, acquires the consistent weight of each temperature rising area, and is convenient for subsequent more accurate positioning of the temperature compensation area, finally acquires the temperature compensation area, improves the body temperature abnormal threshold of the temperature compensation area, and avoids that the normal temperature rising of the physiological temperature rising area of the animal is misjudged as the body temperature abnormality. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, below, a brief introduction will be given to the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained from these drawings without any creative effort.

[0034] Figure 1 A system block diagram of a non-contact infrared body temperature and respiratory frequency synchronous monitoring instrument for veterinary use provided by an embodiment of the present application;

[0035] Figure 2 A flowchart of a consistent weight acquisition method provided by an embodiment of the present application;

[0036] Figure 3 A comparison schematic diagram of the body temperature change curve and the respiratory frequency curve of the temperature rising center provided by an embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined inventive purpose, the specific implementation, structure, features and effects of the non-contact infrared body temperature and respiratory frequency synchronous monitoring instrument for veterinary use according to the present application are described in detail as follows in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0039] The specific scheme of the non-contact infrared body temperature and respiratory frequency synchronous monitoring instrument for veterinary use provided by the present application is specifically described below in combination with the drawings.

[0040] Please refer to Figure 1 , which shows the system block diagram of the non-contact infrared body temperature and respiratory frequency synchronous monitoring instrument for veterinary use provided by one embodiment of the present application, specifically comprising: a data acquisition module 101, a body temperature analysis module 102, a respiratory compensation module 103 and a final compensation module 104.

[0041] The data acquisition module 101 acquires the respiratory frequency of the current monitoring target and the body temperature of each temperature measurement point.

[0042] In one embodiment of the present application, the visible light camera is used to collect and recognize the animal position and key anatomical points (for example, to recognize the position of the animal in the image and locate the position of the animal's nostrils), and output the animal positioning; the thermal imaging camera is used to collect the infrared radiation intensity of the animal's body surface, and output the animal's body temperature data; the thermal imaging camera is used to capture the thermal fluctuation of the animal's nostrils, and output the animal's respiratory frequency; the hardware synchronous clock is used to unify the timestamps of each component in the instrument to realize signal alignment, and the communication module is connected to transmit data to other modules.

[0043] The data acquisition frequency is set to 1 time per second, and 100 temperature measurement points are uniformly arranged on the surface of the animal's body to acquire the body temperature data; the infrared thermal imaging camera is used to track and detect the nostril area of the target to record video in real time. Since the hot air in the animal's body is expelled through the nostrils when the animal exhales, it will cause the temperature of the nostril area to rise. The temperature peak curve of the nostril area at each moment reflects the respiration of the target, and the interval of the extreme points in the curve is used to estimate the respiratory frequency of the current monitoring target at each moment. The historical data of the monitoring target for 7 consecutive days and in a healthy state are acquired.

[0044] It should be noted that collecting the nostril area can require higher equipment precision than the body surface, and the implementer can also obtain the respiratory frequency by collecting the abdominal micro-movement of the animal, adjust the data collection frequency and the time domain range of data analysis, and adjust the number of temperature measurement points.

[0045] The body temperature analysis module 102: screening out abnormal time according to the temperature difference of each temperature measurement point at the same time; taking the nearest preset number of continuous non-abnormal time endpoints on both sides of the time domain of the abnormal time to construct an abnormal period; in the abnormal period, according to the rising degree of the temperature of all abnormal times of each temperature measurement point, obtaining the temperature rising contribution degree of each temperature measurement point; according to the value of the temperature rising contribution degree and the spatial distribution of the corresponding temperature measurement point, screening out the temperature rising center and the temperature rising area.

[0046] When collecting the body temperature of the animal by infrared temperature measurement technology, due to the animal hair or blood vessel distribution and other reasons, there can be certain temperature difference in different parts of the animal; but for some animal breeds, for example, after the ruminant eats, the abdomen will appear a larger local temperature difference, which can cause misjudgment of the body temperature rise of the animal. The high temperature in this case is caused by animal metabolism, although the temperature rise condition appears, it is not abnormal pathological temperature rise, and measures such as injecting fever-reducing drugs do not need to be taken.

[0047] Therefore, first, analyze the local body temperature difference condition of the current monitoring target. Screening out abnormal time according to the temperature difference of each temperature measurement point at the same time, screening out the time when there may be physiological metabolism and other special temperature rise conditions, preparing for subsequent construction of abnormal period to identify the temperature rise area.

[0048] Preferably, in an embodiment of the present application, considering that the larger the variance of the body temperature at the same time, the greater the temperature difference; at the same time, the greater the high amount of the body temperature higher than the average body temperature, the more serious the local high temperature condition, the more likely there is physiological metabolism and other special temperature rise conditions, so according to the variance of the body temperature of each temperature measurement point and the high amount of the body temperature higher than the average body temperature, obtain the temperature difference coefficient of each time;

[0049] Screening out the abnormal time of the body temperature based on the temperature difference coefficient.

[0050] As an example, taking the difference between the body temperature of each temperature measurement point at the current analysis time and the average body temperature as the independent variable, mapping it through the ReLU function, taking the sum of the mapping values of all temperatures as the high amount, and then multiplying the high amount by the variance of all temperatures, after linear normalization in the corresponding data dimension, the normalization result is taken as the temperature difference coefficient of the current analysis time;

[0051] The ReLU function expression is Since the body temperature output less than or equal to the average value of the body temperature is 0, the body temperature higher than the average value of the body temperature is automatically screened by means of the ReLU function, and the high amount is obtained; the temperature difference of each temperature measuring point at the same time is shown by the variance and the high amount.

[0052] The preset temperature difference threshold is set to 0.7, the time when the temperature difference coefficient is greater than the preset temperature difference threshold is marked as an abnormal time, and the time less than or equal to the preset temperature difference threshold, or other times outside the abnormal time, is a non-abnormal time.

[0053] Each time is analyzed and screened, and the screening process is consistent and will not be repeated.

[0054] The abnormal time indicates that the target may have a local temperature that affects the monitoring accuracy, and the temperature difference may be caused by the physiological characteristics of the animal itself (for example, the rumen of a ruminant), or the hair of the animal may affect the emissivity of the infrared radiation, resulting in errors in the data and causing a large temperature difference. Therefore, the nearest preset number of continuous non-abnormal times are obtained on both sides of the time domain of the abnormal time to obtain the endpoints, an abnormal time period is constructed, and a local area with physiological temperature rise is extracted to exclude noise data interference.

[0055] In an embodiment of the present application, the preset number is 30, and the abnormal time not divided into the abnormal time period is taken as the center, and the endpoints are searched from near to far on both sides of the time domain. Only when there are 30 continuous non-abnormal times, the farthest non-abnormal time is marked as the endpoint.

[0056] If there are not enough 30 non-abnormal times, for example, there are 15 continuous non-abnormal times on the left side (negative direction) of the time domain, but the 16th is an abnormal time, then 30 continuous non-abnormal times are searched on the left side of the 16th abnormal time, and the time period between the left and right endpoints is finally taken as the abnormal time period. When there is no data on one side that meets the endpoint condition or can be searched, all the data on one side is included in the abnormal time period.

[0057] It should be noted that in other embodiments of the present application, the implementer can set other preset numbers; the interval time between the abnormal time and the adjacent abnormal time period can also be limited to be not less than the preset number, that is, the different abnormal time periods are not intersected, and when the limit condition is not met, the intersected abnormal time periods are combined.

[0058] Each abnormal time not divided into the abnormal time period is analyzed, and the process of constructing the abnormal time period is consistent and will not be repeated.

[0059] The abnormal period can occur physiological temperature rise of local area of the animal, reflecting the monitoring target temperature rise and recovery process; and in order to locate the animal in the abnormal period of significant temperature rise, facilitate subsequent positioning of the compensation temperature zone, in the abnormal period, according to the temperature rise degree of all abnormal time of each temperature measuring point, the temperature rise contribution degree of each temperature measuring point is obtained, which represents the temperature rise prominence degree of the temperature measuring point.

[0060] Preferably, in one embodiment of the present application, in the abnormal period, considering that the non abnormal time at the left end of the period reflects the normal state of the animal before the abnormal time, the average temperature of the preset number of non abnormal time of the time sequence minimum of all temperature measuring points is obtained as the reference temperature, representing the basal body temperature level before the body temperature abnormality;

[0061] Considering that the larger the body temperature at the abnormal time compared with the reference temperature, the greater the temperature rise degree at the corresponding temperature measuring point, and the higher the temperature rise contribution degree, the temperature rise contribution degree of each temperature measuring point is obtained according to the overlarge degree of the body temperature at all abnormal times of each temperature measuring point compared with the reference temperature.

[0062] As an example, in the current analysis of the abnormal period, each temperature measuring point is selected as a target temperature measuring point, the difference between the highest temperature of all temperature measuring points in the whole abnormal period and the reference temperature is taken as the denominator, the difference between the body temperature of the target temperature measuring point at each abnormal time and the reference temperature is taken as the numerator, and the fractional ratio is taken as the overlarge degree of the body temperature at each abnormal time. The average value of the overlarge degree of the body temperature at all abnormal times is taken as the temperature rise contribution degree of the target temperature measuring point in the current analysis of the abnormal period.

[0063] Among them, the difference between the highest temperature of all temperature measuring points in the whole abnormal period and the reference temperature is the maximum temperature rise amplitude, which shows the temperature rise degree of the target temperature measuring point at each abnormal time, and finally the average value is obtained, which shows the temperature rise degree of the body temperature at all abnormal times of the temperature measuring point.

[0064] The analysis process of each temperature measuring point in each abnormal period is consistent, and will not be repeated.

[0065] In the case of local temperature rise of the animal, the temperature rise amplitude (i.e. temperature rise contribution degree) is often most significant in some specific areas, which are usually the starting point or center area of temperature rise. Although the temperature rise degree of the temperature measuring point near the temperature rise center is slightly low, it shows continuity or aggregation in space. Therefore, according to the numerical value of the temperature rise contribution degree and the spatial distribution of the corresponding temperature measuring point, the temperature rise center and the temperature rise area are screened out, which conforms to the actual animal physiological structure and heat diffusion mechanism, and lays a foundation for subsequent positioning of the compensation temperature zone.

[0066] Preferably, in one embodiment of the present application, the temperature rise degree of each temperature measuring point on the surface of the monitoring target in a temperature rise and fall process is reflected, and the local peak point with a large temperature rise degree is screened out from the temperature measuring points, which represents a temperature rise center of a local temperature rise area, and then the adjacent area of the temperature rise center according to the temperature diffusion can obtain several temperature rise areas in the marking section.

[0067] Based on this, the temperature measuring point with a temperature rise degree greater than a preset first temperature rise threshold and greater than the temperature rise degree of all adjacent temperature measuring points is marked as a temperature rise center; the temperature measuring point with a temperature rise degree greater than a preset second temperature rise threshold is marked as a selected temperature measuring point, the adjacent selected temperature measuring points are included in the initial temperature rise area with the temperature rise center as the starting point, the adjacent selected temperature measuring points of the edge temperature measuring points of the initial temperature rise area are included, the initial temperature rise area is updated, the selected temperature measuring points are repeatedly included, and the latest initial temperature rise area is taken as the temperature rise area until all edge temperature measuring points of the latest initial temperature rise area do not have adjacent selected temperature measuring points not included. The preset first temperature rise threshold is greater than the preset second temperature rise threshold.

[0068] As an example, the preset first temperature rise threshold is 0.6, and the preset second temperature rise threshold is 0.2; the repeated inclusion of the selected temperature measuring point is that: the adjacent selected temperature measuring points of the edge temperature measuring points of the latest initial temperature rise area are included, the initial temperature rise area is updated, then in the updated latest initial temperature rise area, the adjacent selected temperature measuring points of the edge temperature measuring points are included, and this process is repeated, which is similar to region growing, until all edge temperature measuring points of the latest initial temperature rise area do not have adjacent selected temperature measuring points not included.

[0069] It should be noted that the acquisition process of the temperature rise center and the temperature rise area of each abnormal period is consistent, and only one example is described here, and the description will not be repeated. There can be multiple temperature rise centers and temperature rise areas under one abnormal period. When the temperature rise areas of multiple temperature rise centers overlap under one abnormal period, region merging is performed, and the average value of the parameters used by multiple temperature rise centers is used to replace the parameters in subsequent analysis.

[0070] The respiratory compensation module 103: analyzes the consistency of the temperature rise process between the abnormal periods in time and space, and obtains the consistent weight of each temperature rise area.

[0071] For physiological temperature rise of animals, although physiological local temperature rise can cause large temperature differences between some points and other points, it still conforms to its physiological law. When the body temperature of an animal rises, it will speed up the breathing frequency to dissipate heat to regulate the body temperature, so after the body temperature of an animal rises, the breathing frequency is often accelerated, and because the breathing frequency is accelerated, the heat dissipation capacity of the animal is enhanced, so the body temperature will fall after the breathing frequency increases for a period of time.

[0072] Since the physiological temperature rise of the animal is related to physiological behaviors of the animal, such as exercise, eating, etc., and such physiological behaviors occur multiple times, the multiple physiological temperature rises have similar performances in time, and since the organs performing the physiological temperature rise do not change, the multiple physiological temperature rises also have concentrated performances in space. By analyzing the consistency of the temperature rise in time and space between the abnormal periods, the consistent weight of each temperature rise region is obtained, and for noise data, the performance of the data in the overall time sequence often does not have regularity and is obviously different from the change rule of other points. Obtaining the consistent weight also facilitates more accurate positioning of the temperature zone to be compensated in the subsequent process, avoiding the interference of noise data or data of other regions.

[0073] Preferably, in an embodiment of the present application, please refer to Figure 2 which shows a flowchart of a method for obtaining a consistent weight provided by an embodiment of the present application, and specifically comprises:

[0074] Step S301: The temperature rise region with the largest number of overlapping temperature measurement points in other abnormal periods is recorded as the corresponding region in other abnormal periods.

[0075] In order to analyze the consistency of the physiological temperature rise of the animal in time and space, first, the temperature rise regions in different abnormal periods are matched. It is considered that the more the number of overlapping temperature measurement points in different periods, the more likely it is that the same body region of the animal, so the temperature rise region with the largest number of overlapping temperature measurement points in other abnormal periods is recorded as the corresponding region in other abnormal periods.

[0076] As an example, the region at the first appearance of each temperature rise region is taken as the reference, and each temperature rise region is matched with the temperature rise region in other abnormal periods; wherein, when the temperature measurement points of a certain temperature rise region do not coincide with the temperature measurement points of the historically appeared temperature rise region, or when it appears in the first abnormal period in time sequence, it is determined as the first appearance.

[0077] In other embodiments of the present application, the implementer can also set the number of overlapping temperature measurement points of the temperature rise region and the historically appeared temperature rise region to be less than a certain threshold, such as less than 2, and also determine it as the first appearance.

[0078] Step S302: In each temperature rise region and the corresponding region, the consistency of the change lag of the respiratory frequency compared with the body temperature of the temperature center is analyzed in the abnormal period, and a time consistency coefficient is obtained.

[0079] Considering that the respiratory frequency change caused by the physiological normal temperature rise of the animal has a hysteresis, the stronger the consistency of the change hysteresis, the more similar the performance of multiple physiological temperature rises from the time perspective, the higher the consistency weight, and the temperature size of the temperature center of the temperature rise region represents the temperature change of the whole temperature rise region, the consistency analysis is performed.

[0080] In an embodiment of the present application, considering that each data corresponds to the temperature rise initial, physiological response peak and adjustment recovery stage before rising, reaching the maximum point and after falling respectively, the three stages can more comprehensively reflect the response relationship between the temperature change and the respiratory frequency of the animal in the abnormal period, therefore, in any temperature rise region and each abnormal period to which the region belongs, the time difference of three groups of key points corresponding to the respiratory frequency and the temperature of the temperature center before rising, reaching the maximum point and after falling is obtained.

[0081] Please refer to Figure 3 , which shows a comparison diagram of the temperature change curve and the respiratory frequency curve of a temperature center provided by an embodiment of the present application; Figure 3 The horizontal axis is the time axis, and the vertical axis is the data axis, which is realized as a temperature change curve, the dashed line is a respiratory frequency curve, the hollow circle (the leftmost of the two curves, which is the minimum point on the left side of the maximum point of the curve) represents the key point corresponding to the pre-rising, the circle located at the highest point of the two curves represents the key point corresponding to the maximum point, and the maximum value is the local maximum value, which is the maximum point in the whole time domain; the solid circle (the rightmost of the two curves, which is the minimum point on the right side of the maximum point of the curve) represents the key point corresponding to the post-falling, and there are three groups of key points in total.

[0082] Then, the time consistency coefficient is obtained according to the consistency of the time difference of each key point in all abnormal periods.

[0083] As an example, in the abnormal period corresponding to each temperature rise region, the existing least square method is used to perform curve fitting on the respiratory frequency and the temperature of the temperature center of the abnormal period respectively, the maximum point, the minimum point on the left side of the maximum point and the minimum point on the right side of the maximum point on each curve are marked, and three groups of key points corresponding to the maximum point, the pre-rising and the post-falling are obtained.

[0084] The time difference between the respiratory frequency corresponding time of each group of key points and the temperature corresponding time is taken as the hysteresis, which represents the time difference of a group of key points; all hysteresis of each temperature rise region and the corresponding region are divided into 3 sets according to the types of key points, the average value of the absolute value of the difference of all element pairs (any combination of two elements) in each set is taken as the difference coefficient of each key point, and the inverse of the average value of the difference coefficients of the three key points is taken as the time consistency coefficient.

[0085] Wherein, the difference degree of time difference is expressed by the absolute difference value, and the consistency degree of time difference is expressed by the negative correlation mapping of the reciprocal of the difference coefficient of the three key points, to obtain the time consistency coefficient.

[0086] It should be noted that, considering the hysteresis of the respiratory frequency, when comparing two data curves, the key point corresponding to the decline of the respiratory frequency curve is captured as the target, and the time domain range can be appropriately relaxed.

[0087] Step S303: According to the distribution dispersion characteristics of the temperature rise center of each temperature rise area and its corresponding area, the spatial consistency coefficient is obtained.

[0088] Considering that the more dispersed the distribution of the temperature rise center of the temperature rise area and its corresponding area is, the more unlikely it is to correspond to the same animal body area, and the more unlikely it is to be a special area of physiological temperature rise of the animal, the spatial consistency coefficient is obtained.

[0089] As an example, considering that the smaller the circumscribed circle of all temperature rise centers is, the more concentrated the temperature rise center distribution is, the reciprocal of the minimum circumscribed circle area of the temperature rise center of each temperature rise area and its corresponding area is taken as the spatial consistency coefficient.

[0090] Step S304: According to the overlap degree between each temperature rise area and its corresponding area, and the existence rate in all abnormal periods, the time consistency coefficient and the spatial consistency coefficient are combined to obtain the consistency weight of each temperature rise area.

[0091] It is also considered that the higher the overlap degree between the temperature rise area and its corresponding area is, the stronger the consistency of the corresponding temperature rise area is, and the higher the existence rate in the abnormal period is, which means that the recurrence rate of the temperature rise area is higher, and it is more likely to be a physiological normal temperature rise area. Therefore, the overlap degree, the existence rate, the time consistency coefficient and the spatial consistency coefficient are combined to obtain the consistency weight of each temperature rise area.

[0092] As an example, each temperature rise area and its corresponding area are sorted in time sequence, and for each pair of time sequence adjacent temperature rise areas, the sum of the number of temperature measurement points of the two temperature rise areas is taken as the denominator, the 2 times of the number of overlapping temperature measurement points is taken as the numerator, and the fractional ratio is taken as the overlap sub-coefficient of the pair of adjacent temperature rise areas. The average value of all overlap sub-coefficients is taken as the overlap coefficient, representing the overlap degree between the temperature rise area and its corresponding area.

[0093] The proportion value of the total number of abnormal periods of each temperature rise area and its corresponding area in all abnormal periods is taken as the existence rate.

[0094] The product of the overlap coefficient, the existence rate, the time consistency coefficient and the space consistency coefficient is finally normalized after linear normalization in the corresponding data dimension, and the normalized result is used as the consistency weight of each temperature rising region.

[0095] It should be noted that the consistency weight of each temperature rising region in one temperature rising region and its corresponding region is the same. The analysis process for each temperature rising region and its corresponding region is consistent, and will not be repeated; in other embodiments of the present application, the implementer can also fuse the overlap coefficient, the existence rate, the time consistency coefficient and the space consistency coefficient by adding or weighted summation, and will not be repeated.

[0096] Finally, the compensation module 104: according to the temperature rising contribution degree and the consistency weight of each temperature measurement point in all abnormal time periods, obtains the temperature zone to be compensated, and improves the body temperature abnormal threshold of the temperature zone to be compensated.

[0097] Since the temperature measurement points constituting the temperature rising region in each abnormal time period are different, the temperature rising contribution degree of each temperature measurement point in each abnormal time period is analyzed in detail, and the consistency weight of the corresponding temperature rising region is analyzed, the special evaluation of each temperature measurement point is obtained, the temperature zone to be compensated is obtained, and the body temperature abnormal threshold of the temperature zone to be compensated is improved, so as to avoid that the normal temperature rising of the physiological temperature rising region of the animal is misjudged as the body temperature abnormality.

[0098] Preferably, in an embodiment of the present application, considering that the higher the temperature rising contribution degree of the temperature measurement point is, the more prominent the point is in the temperature rising process, and the more likely it is the center or active area of the physiological heat source of the animal, and the higher the consistency weight is, indicating that the temperature rising region where the point is located has strong time and space consistency in multiple abnormal time periods, and is more likely to be the region of the non-pathological temperature rising phenomenon caused by the stable physiological structure of the animal, so the temperature rising contribution degree and the consistency weight of each temperature measurement point in all abnormal time periods are fused to obtain the temperature measurement point to be compensated.

[0099] The region composed of the temperature measurement points corresponding to all the temperature measurement factors greater than the preset compensation threshold is recorded as the temperature zone to be compensated.

[0100] As an example, the preset compensation threshold is 0.75, for each temperature measurement point, the product of the temperature rising contribution degree of the temperature measurement point in each abnormal time period and the consistency weight of the temperature rising region where the temperature measurement point is located is used as the special sub-factor of each abnormal time period; the sum of all special sub-factors is normalized in the corresponding data dimension to obtain the temperature measurement point to be compensated.

[0101] When the temperature measurement point does not belong to any temperature rising region in a certain abnormal time period, the corresponding consistency weight is zero, and the special sub-factor is zero; when the temperature measurement point does not input any temperature rising region in all abnormal time periods, the temperature measurement factor to be compensated is zero.

[0102] 1.05 times of the highest body temperature (in healthy time) of the history of the temperature zone to be compensated is set as the corresponding body temperature abnormal threshold.

[0103] It should be noted that in other embodiments of the present application, the implementer can adjust the increase rate or amount of the temperature zone to be compensated, or adjust the preset compensation threshold; the analysis process for each temperature measurement point is consistent and will not be repeated. The implementer can also set the update frequency, such as updating once every 7 days, updating the setting of the temperature zone to be compensated and the threshold, and when updating the body temperature abnormal threshold, only referring to the body temperature within the latest historical body temperature abnormal threshold.

[0104] To sum up, in view of the technical problem of inaccurate identification of temperature abnormalities in specific areas of animals, the present application provides a non-contact infrared body temperature and respiratory frequency synchronous monitor for veterinarians. The present application obtains the respiratory frequency and body temperature data of the current monitoring target through the data acquisition module; the abnormal time is screened out and the abnormal period is constructed through the body temperature analysis module; further, the temperature rise contribution degree of each temperature measurement point in the abnormal period is obtained; further, the temperature rise center and the temperature rise area are screened out according to the numerical value of the temperature rise contribution degree and the spatial distribution of the corresponding temperature measurement point; the consistency of the temperature rise process in time and space between the abnormal periods is analyzed through the respiratory compensation module, and the consistent weight of each temperature rise area is obtained; finally, the temperature zone to be compensated is obtained through the final compensation module, and the body temperature abnormal threshold of the temperature zone to be compensated is increased. The present application obtains the body temperature and respiratory frequency data, identifies the body temperature abnormal period and the temperature rise area, analyzes the time and spatial consistency combined with the respiratory frequency change, extracts the temperature zone to be compensated and dynamically adjusts the body temperature abnormal threshold, solves the physiological temperature rise misjudgment problem, and improves the accuracy and robustness of the body temperature and respiratory frequency monitoring.

[0105] It should be noted that the above-mentioned embodiments of the present application are in the order of description only, and do not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous.

[0106] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment mainly describes the difference from other embodiments.

Claims

1. A non-contact infrared body temperature and respiratory rate synchronous monitor for veterinary use, characterized in that, The monitor comprises: a data acquisition module: acquiring the respiratory rate of the current monitoring target and the body temperature of each temperature measuring point; a body temperature analysis module: screening out abnormal time points according to the body temperature differences of each temperature measuring point at the same time; acquiring endpoints by intercepting the nearest preset number of continuous non-abnormal time points on both sides of the time domain at the abnormal time points, and constructing an abnormal time period; in the abnormal time period, acquiring the temperature rise contribution degree of each temperature measuring point according to the rising degree of the body temperature of each temperature measuring point at all the abnormal time points; screening out a temperature rise center and a temperature rise region according to the numerical value of the temperature rise contribution degree and the spatial distribution of the corresponding temperature measuring point; a respiratory compensation module: analyzing the consistency of the temperature rise process in time and space between the abnormal time periods, and acquiring a consistent weight of each temperature rise region; a final compensation module: acquiring a temperature zone to be compensated according to the temperature rise contribution degree and the consistent weight of each temperature measuring point in all the abnormal time periods, and increasing the body temperature abnormal threshold of the temperature zone to be compensated; the method for acquiring the consistent weight comprises: the temperature rise region with the largest number of overlapping temperature measuring points in other abnormal time periods is recorded as the corresponding region of the temperature rise region in other abnormal time periods; in each abnormal time period in which the temperature rise region and the corresponding region are located, the consistency of the change lag of the respiratory rate compared with the body temperature of the temperature rise center is analyzed to acquire a time consistency coefficient; the distribution discrete features of the temperature rise center of each temperature rise region and the corresponding region are analyzed to acquire a space consistency coefficient; the consistent weight of each temperature rise region is acquired according to the overlapping degree between each temperature rise region and the corresponding region, the existence rate in all abnormal time periods, and the time consistency coefficient and the space consistency coefficient.

2. The non-contact infrared body temperature and respiration rate synchronous monitor for veterinary use according to claim 1, characterized in that, the method for acquiring the abnormal time points comprises: a temperature difference coefficient of each time point is acquired according to the variance of the body temperature of each temperature measuring point and the high amount of the body temperature higher than the average body temperature; abnormal time points of the body temperature are screened out based on the temperature difference coefficient.

3. The non-contact infrared body temperature and respiration rate synchronous monitor for veterinary use according to claim 2, characterized in that, the method for screening out abnormal time points of the body temperature based on the temperature difference coefficient comprises: the time points with the temperature difference coefficient greater than a preset temperature difference threshold are marked as abnormal time points.

4. The non-contact infrared body temperature and respiration rate synchronous monitor for veterinary use according to claim 1, characterized in that, the method for acquiring the temperature rise contribution degree comprises: in the abnormal time period, the average body temperature of the preset number of non-abnormal time points with the smallest time sequence of all temperature measuring points is acquired as a reference body temperature; the temperature rise contribution degree of each temperature measuring point is acquired according to the large degree of the body temperature of each temperature measuring point at all the abnormal time points compared with the reference body temperature.

5. The non-contact infrared body temperature and respiration rate synchronous monitor for veterinary use according to claim 1, characterized in that, the method for screening out the temperature rise center and the temperature rise region comprises: The temperature measurement point with the temperature rise contribution degree greater than a preset first temperature rise threshold and greater than the temperature rise contribution degrees of all adjacent temperature measurement points is marked as a temperature rise center; the temperature measurement point with the temperature rise contribution degree greater than a preset second temperature rise threshold is marked as a candidate temperature measurement point; the adjacent candidate temperature measurement points are included in an initial temperature rise region with the temperature rise center as a starting point; the adjacent candidate temperature measurement points of the edge temperature measurement points of the initial temperature rise region are included, the initial temperature rise region is updated, the candidate temperature measurement points are repeatedly included until there is no adjacent candidate temperature measurement point not included in all edge temperature measurement points of the latest initial temperature rise region, and the latest initial temperature rise region is taken as a temperature rise region; the preset first temperature rise threshold is greater than the preset second temperature rise threshold.

6. The non-contact infrared body temperature and respiration rate synchronous monitor for veterinary use according to claim 1, characterized in that, The method for obtaining the time consistency coefficient comprises: In each of the abnormal time periods of the temperature rise region and the corresponding region, the time difference of three groups of key points corresponding to the respiratory frequency and the body temperature of the temperature rise center before rising, at the maximum point, and after falling is obtained; and the time consistency coefficient is obtained according to the consistency of the time difference of each key point in all the abnormal time periods.

7. The non-contact infrared thermometer for veterinary use as claimed in claim 1, wherein, The method for obtaining the space consistency coefficient comprises: The reciprocal of the minimum circumscribed circle area of the temperature rise center of each temperature rise region and the corresponding region is taken as the space consistency coefficient.

8. The non-contact infrared body temperature and respiration rate synchronous monitor for veterinary use according to claim 1, characterized in that, The method for obtaining the temperature compensation region comprises: The temperature rise contribution degree and the consistency weight of each temperature measurement point in all the abnormal time periods are fused to obtain a temperature compensation factor of each temperature measurement point; and a region formed by temperature measurement points corresponding to all temperature compensation factors greater than a preset temperature compensation threshold is recorded as a temperature compensation region.

9. The non-contact infrared body temperature and respiration rate synchronous monitor for veterinary use according to claim 1, characterized in that, The method for increasing the body temperature abnormal threshold of the temperature compensation region comprises: 1.05 times of the historical maximum body temperature of the temperature compensation region is set as the corresponding body temperature abnormal threshold.

Images