TWI626926B - System and method for monitoring patient with abnormal body temperature - Google Patents

Abstract

The invention provides a body temperature abnormality detecting system and method for detecting an individual with abnormal body temperature in an area. The system includes a connected sensing module and a processing module. The sensing module continuously generates a thermal image of the area, divides the thermal image into a plurality of mesh images, and determines whether there is a mesh image whose temperature exceeds the threshold temperature, and finally determines the target heat according to the mesh image exceeding the threshold temperature. image. The processing module generates a temperature versus time graph of the target thermal image, and determines whether there is an individual with abnormal body temperature in the region, and if so, according to the temperature of the target thermal image and each mesh image in the target thermal image The coordinates send a warning message to report the location of the individual whose body temperature is abnormal in this area and its body temperature.

Description

Body temperature abnormality detecting system and method

The present invention relates to a detection system and method, and more particularly to a body temperature abnormality detecting system and method capable of detecting whether an individual having an abnormal body temperature in a region can report the position of the individual in the region and The image of the individual.

As far as the current common body temperature detection system is concerned, it is mostly the main purpose of screening for fever epidemic. Such body temperature detection systems are more suitable for areas where people flow more frequently and densely, such as airport halls, hospital halls, office buildings, department stores, and the like. However, such a body temperature detection system cannot accurately grasp the position of an individual with abnormal body temperature in the detected area. In addition, since such a body temperature detecting system uses thermal images alone as a basis for screening, even if an individual with abnormal body temperature is found in the detected area, the appearance and even the identity of the individual cannot be known.

The invention provides a body temperature abnormality detecting system for detecting whether an abnormal body temperature exists in an area. The body temperature abnormality detecting system comprises a sensing module and a processing module, wherein the processing module is connected to the sensing module. The sensing module includes an image capturing module, an image deconstructing module and an error eliminating module. The image deconstruction module is connected to the image capture module, and the error elimination module is connected to the image deconstruction module.

The image capture module is configured to continuously generate thermal images for the area. The image deconstruction module is configured to divide the thermal image of the region into a plurality of mesh images according to a preset coordinate system, and respectively determine whether there is a mesh image in the thermal image whose temperature exceeds the threshold temperature, wherein each of the thermal images The grid image has coordinates that are positioned by the coordinate system. The error elimination module determines the target thermal image based on the mesh image that exceeds the threshold temperature.

The processing module includes an image analysis module and a determination module, wherein the determination module is connected to the image analysis module. The image analysis module receives the target thermal image to obtain a temperature versus time graph of each of the mesh images in the target thermal image. Next, the determining module determines whether there is an individual with abnormal body temperature in the area according to the relationship diagram, wherein the relationship diagram shows whether the temperature of the target thermal image rises with time. If the image analysis module determines that there is an individual whose body temperature is abnormal in the area, the determination module sends a warning message according to the coordinates of each grid image in the target thermal image to report the position of the individual whose body temperature is abnormal in the area.

In an embodiment of the body temperature abnormality detecting system provided by the present invention, the image capturing module continuously generates an actual image for the area, each actual image corresponds to each thermal image, and the image deconstructing module is configured according to The pre-set coordinate system divides the actual image of the area into a plurality of grid images, wherein each grid image of the actual image has coordinates that are positioned by the coordinate system. In addition, the processing module further includes a feature recognition module. The feature recognition module finds the corresponding mesh images in the actual image according to the time of the target thermal image generation and the coordinates of each mesh image in the target thermal image, and performs feature recognition on the mesh images.

In an embodiment of the body temperature abnormality detecting system provided by the present invention, after the feature recognition module finds the corresponding mesh images in the actual image, the feature recognition module restores the mesh images into three dimensions. The image is characterized by facial features or cranial features to identify different individuals.

The embodiment of the invention further provides a method for detecting a fever individual, which is suitable for a fever individual detection system for detecting whether a body has a fever in an area. The applicable individual detection system for fever includes a connected sensing module and a processing module.

The sensing module includes an image capturing module, an image deconstructing module and an error eliminating module, the image deconstructing module is connected to the image capturing module, and the error eliminating module is connected to the image deconstructing module. The processing module includes an image analysis module and a determination module, and the determination module is connected to the image analysis module.

The method for detecting abnormal body temperature includes: generating a thermal image for the region through the image capturing module; and dividing the thermal image of the region into a plurality of meshes according to a preset coordinate system through the image deconstruction module; a grid image, wherein each grid image of the thermal image has a coordinate positioned by the coordinate system; the error elimination module is used to determine the target thermal image according to the grid image exceeding a threshold temperature; and the image is received through the image analysis module Thermal image to obtain the temperature versus time relationship of each grid image in the target thermal image; through the judgment module, determine whether there is an individual with abnormal body temperature in the region according to the relationship diagram, and if there is an abnormal body temperature in the region A warning message is sent according to the coordinates of each grid image in the target thermal image to report the position of the individual whose body temperature is abnormal in the area.

In an embodiment of the method for detecting a body temperature abnormality according to the present invention, the method for detecting abnormal body temperature includes: generating an actual image for the region through the image capturing module, wherein each actual image corresponds to Each of the thermal images is divided into a plurality of mesh images according to a preset coordinate system, wherein each of the mesh images of the actual image has a coordinate positioned by the coordinate system. In addition, in this embodiment, the body temperature abnormality detecting system applied to the body temperature abnormality detecting method further includes a feature recognition module, and the body temperature abnormality individual detecting method further comprises: transmitting the feature recognition module according to the target heat image. The generated time and the coordinates of each mesh image in the target thermal image, find the corresponding mesh images in the actual image, and perform feature recognition on the mesh images.

In an embodiment of the method for detecting a body temperature abnormality according to the present invention, after the corresponding mesh images in the actual image are found, the feature recognition module restores the mesh images into a three-dimensional image, The three-dimensional image is characterized for facial features or skull features.

In summary, the body temperature abnormality detecting system and method provided by the present invention can automatically detect an individual with abnormal temperature symptoms in an area, and know where the individual is currently located in the area, and Obtain the actual image of the individual. Symptoms of abnormal body temperature are a warning symptom for many diseases, and if found in time, it will be conducive to the diagnosis and treatment of the disease. Therefore, through the body temperature abnormality detecting system and method provided by the invention, the medical staff can firstly know whether any of the people in the waiting room have symptoms of abnormal body temperature; in addition, the medical staff can further know that there is The location of a person with symptoms and identity. In the first place, it will help medical staff to take immediate care of this person.

The detailed description of the present invention and the accompanying drawings are to be understood by the claims The scope is subject to any restrictions.

10‧‧‧Sensing module

12‧‧‧Image capture module

14‧‧‧Image Deconstruction Module

16‧‧‧Error elimination module

20‧‧‧Processing module

22‧‧‧Image Analysis Module

24‧‧‧Judgement module

26‧‧‧Character Identification Module

27‧‧‧Displacement Search Module

28‧‧‧Image restoration module

T‧‧‧Temperature regression line

A~E‧‧‧ Grid Image

S601~S607‧‧‧Steps

S701~S703, S704A~S704B, S705A~S705B, S706A~S706B, S707~S708‧‧‧ steps

1 is a block diagram of a body temperature abnormality detecting system according to an exemplary embodiment of the present invention.

FIG. 2 is a schematic diagram of dividing a thermal image into a plurality of mesh images.

Figure 3 is a graph of temperature vs. time for each grid image in the target thermal image.

FIG. 4 is a block diagram of a body temperature abnormality detecting system according to another exemplary embodiment of the present invention.

5A and 5B respectively show thermal images and actual images taken at the same time for a region.

FIG. 6 is a flowchart of a method for detecting an abnormal body temperature according to an exemplary embodiment of the present invention.

FIG. 7 is a flowchart of a method for detecting an abnormal body temperature according to another exemplary embodiment of the present invention.

Various illustrative embodiments are described more fully hereinafter with reference to the accompanying drawings. However, the inventive concept may be embodied in many different forms and should not be construed as being limited to the illustrative embodiments set forth herein.

Rather, these exemplary embodiments are provided so that this invention will be in the

In the figures, like numerals are used to indicate like elements.

The body temperature abnormality detecting system and method provided by the invention can automatically detect whether an individual has symptoms of abnormal body temperature in an area and transmit the data to the device of the medical staff through the Internet of Things.

The area in which the body temperature abnormality detecting system and the method provided by the present invention are applied is a small area. For example, the body temperature abnormality detecting system and the method thereof provided by the present invention are applicable to a waiting room of a hospital, and the like. In the following description, a plurality of embodiments will be specifically described to describe a body temperature abnormality detecting system and a method thereof provided by the present invention.

[An embodiment of a body temperature abnormality detecting system]

Please refer to FIG. 1. FIG. 1 is a block diagram of a body temperature abnormality detecting system according to an exemplary embodiment of the present invention. As described above, the area to which the body temperature abnormality detecting system provided in the present embodiment is applied is a small area, such as a waiting room of a hospital, but the invention is not limited thereto.

As shown in FIG. 1 , the body temperature abnormality detecting system provided in this embodiment includes a sensing module 10 and a processing module 20 . As shown in FIG. 1 , the sensing module 10 is connected to the processing module 20 . The sensing module 10 mainly includes an image capturing module 12 and an image deconstructing module 14 The error elimination module 16 and the processing module 20 mainly include an image analysis module 22 and a determination module 24.

The image deconstruction module 14 is connected to the image capture module 12 , and the error elimination module 16 is connected to the image deconstruction module 14 . In addition, the image analysis module 22 is connected to the determination module 24. It should be noted that, in this embodiment, the image capturing module 12 is mainly used as a sensor hardware for performing image capturing (eg, a thermal imager, a camera, etc.), and the image deconstructing module 14 is The error elimination module 16 is mainly implemented by a firmware for image deconstruction and error elimination (eg, computer, etc.), and the image analysis module 22 and the determination module 24 are mainly provided with a servo set in the cloud. Device implementation (eg, cloud server).

Further, the image capturing module 12 is configured to continuously generate thermal images for an area. In this embodiment, the image capturing module 12 includes at least one thermal imager to continuously capture thermal images of the detected area. In other embodiments, it is considered that a single camera may not be able to capture the entire detected area. Therefore, the image capturing module 12 may also include multiple cameras to perform complete detection on the entire detected area. Measurement.

The image deconstruction module 14 is configured to divide a thermal image of an area into a plurality of grid images according to a preset coordinate system, so that the error elimination module 16 successively determines whether there is a temperature exceeding one threshold in each thermal image. A grid image of temperature. In this embodiment, the image deconstruction module 14 divides the thermal image captured by the image capturing module 12 into a plurality of mesh images by a custom coordinate system, and each grid in the thermal image The image is positioned using the coordinates of the coordinate system. It should be noted that, by means of space coordinate conversion, the coordinates of each mesh image in the coordinate system can be converted into position information in real space, and the details about coordinate conversion should be in the technical field of the present invention. It is familiar to those with ordinary knowledge, so there is not much description here. It is worth noting that the threshold temperature is 37.5 degrees Celsius. According to the clinical literature of most hospitals, 37.5 degrees Celsius is the average body temperature of the human body in a healthy state.

The error elimination module 16 determines a target thermal image based on a grid image that exceeds the threshold temperature. Please refer to FIG. 2. FIG. 2 is a schematic diagram of dividing a thermal image into a plurality of mesh images. As shown in FIG. 2, the grid image displays different temperatures according to the depth of the color, and the grid image with a lighter color represents a lower temperature. Conversely, the grid image with a darker color represents a higher temperature. In addition, the darkest grid image is a grid image that exceeds the threshold temperature. That is to say, the darkest grid image may be determined by the error elimination module 16 as the target thermal image.

Next, the error elimination module 16 is further described to determine the details of the book target thermal image from the plurality of grid images. When the error elimination module 16 determines the target thermal image based on the mesh image exceeding the threshold temperature, the error elimination module 16 associates a plurality of adjacent mesh images exceeding the threshold temperature (as shown at A in FIG. 2). And judge A is the target thermal image.

However, in a thermal image divided into multiple grid images, there may be a case where only a single grid image whose temperature exceeds the threshold temperature (as in B of FIG. 2), or several sporadic grid images may appear. The temperature exceeds the threshold temperature (as in Figure 2, C). In terms of the area of the individual with abnormal body temperature in the thermal image, the above two cases cannot be individuals with abnormal body temperature, that is, the above two cases cannot be judged as the target thermal image.

Therefore, in the present embodiment, in order to exclude hot spots displayed by individuals who are not abnormal in body temperature in the image, the number of adjacent grid images exceeding the threshold temperature must be greater than or equal to the number of thresholds.

Next, after the error elimination module 16 determines the target thermal image, the thermal image is received by the image analysis module 22 in the processing module 20. According to the received thermal image, the image analysis module 22 can obtain a temperature versus time relationship of each mesh image in the target thermal image, and according to the relationship diagram, the determination module 24 can determine whether the region exists. Individuals with abnormal body temperature. For example, please refer to FIG. 3. FIG. 3 is a relationship diagram of temperature versus time of each mesh image in the target thermal image.

Further, when the judging module 24 determines whether there is an individual with abnormal body temperature in the region according to the temperature versus time graph of each grid image in the target thermal image, the judging module 24 performs regression analysis on the graph. Obtain a temperature regression line (temperature regression line T as shown in Figure 3). Then, based on the obtained temperature regression line, the determination module 24 can determine whether the temperature of the target thermal image is greater than or equal to a warning temperature. It is worth noting that this warning temperature is 38.5 degrees Celsius. According to the clinical literature of most hospitals, when the body temperature is between 37.5 degrees Celsius and 38.5 degrees Celsius, it can still be regarded as the human body in a healthy state; however, when the body temperature exceeds 38.5 degrees Celsius, it means that the human body may be in the body. In an abnormal state.

Therefore, if the determining module 24 determines that the temperature of the target thermal image is greater than or equal to 38.5 degrees Celsius, it can be determined that there is an individual with abnormal body temperature in the detected area. At this time, the judging module 24 sends a warning message according to the temperature of the target thermal image and the coordinates of each grid image in the target thermal image to report the position of the individual of the fever in this area. That is to say, the warning message may include coordinates in the actual space converted by the target thermal image in the preset coordinate system, and the temperature of the target thermal image (ie, the body temperature of the individual whose body temperature is abnormal).

It should be noted that, as shown in FIG. 3, the temperature regression line T is a time segment according to the temperature versus time of each mesh image in the target thermal image (ie, t-Δt~t+ in FIG. 3). The temperature data recorded in Δt) was obtained.

Therefore, if the temperature of the target thermal image is not greater than or equal to 38.5 degrees Celsius according to the temperature regression line T, it means that the individual has a higher time zone (t-Δt~t+Δt). Body temperature, but the human body is not in an abnormal state. Therefore, the judging module 24 does not make a judgment on the target thermal image that "there is an individual with abnormal temperature in the detected area".

For different kinds of diseases, the abnormal temperature of the body has its characteristics, that is to say, from the change of temperature with time, the symptoms of abnormal body temperature caused by different kinds of diseases are slightly different. Therefore, in this embodiment, the determining module 24 can further select a relationship between the temperature of each mesh image in the target thermal image and time. A plurality of reference maps stored in advance are compared. It should be noted that different reference pictures show changes in body temperature over time when different types of diseases cause abnormal body temperature.

In short, the judging module 24 can determine whether there is an individual with abnormal body temperature in an area according to the temperature versus time map of each grid image in the target thermal image, and the judging module 24 can initially determine the possibility. What kind of disease causes the body temperature abnormality of the individual.

According to the above, if the temperature versus time graph of each grid image in the target thermal image matches one of the plurality of reference maps, the judging module 24 judges that there is an individual whose body temperature is abnormal in the region, and according to the temperature of the target thermal image. The coordinates of each grid image in the target thermal image and the matching reference maps send a warning message. It should be noted that the warning message may include the coordinates in the actual space converted by the target thermal image in the preset coordinate system, the temperature of the target thermal image (ie, the body temperature of the individual whose body temperature is abnormal), and the individual whose body temperature is abnormal. Possible diseases.

In this way, the abnormal body temperature detection system provided by the embodiment can assist the medical staff to know whether there is an abnormal temperature in the post-diagnosis room during the waiting period, and can also initially understand the diseases that the individual may have.

In order to more specifically describe the body temperature abnormality detecting system of the present invention, an embodiment will be further described below. In the following embodiments, portions different from the embodiment illustrated in FIG. 1 above will be described, and the remaining omitted portions are the same as the embodiment illustrated in FIG. 1 described above. In addition, for the sake of convenience, like reference numerals or numerals indicate similar elements.

[Another embodiment of a body temperature abnormality detecting system]

Please refer to FIG. 4. FIG. 4 is a block diagram of a body temperature abnormality detecting system according to another exemplary embodiment of the present invention. As in the foregoing embodiment, the area applicable to the body temperature abnormality detecting system provided by the embodiment is also a small area, such as a waiting room of a hospital, but the invention is not limited thereto.

The body temperature abnormality detecting system provided by the embodiment has similar system architecture and similar working mechanism as the body temperature abnormality detecting system provided by the foregoing embodiment. However, in the system architecture, one of the differences between the body temperature abnormality detecting system provided in this embodiment and the body temperature abnormality detecting system provided by the foregoing embodiment is that, in this embodiment, the processing module 20 further includes The feature recognition module 26, the displacement search module 27 and the image restoration module 28 are provided. As shown in FIG. 4, the feature recognition module 26, the displacement search module 27, and the image restoration module 28 are both connected to the determination module 24. It should be noted that the image capturing module 12 is mainly used for sensor hardware (such as a thermal imager, a camera, etc.) for image capturing, and the image deconstructing module 14 and the error eliminating module 16 are mainly It is implemented by firmware for image deconstruction and error elimination (eg, computer, etc.). In addition, image analysis module 22, determination module 24 feature recognition module 26, displacement retrieval module 27 and image restoration mode Group 28 is mainly implemented by a server device installed in the cloud (for example, a cloud server).

In addition, in the working mechanism, the body temperature abnormality detecting system provided by the embodiment is different from the body temperature abnormality detecting system provided by the foregoing embodiment in that, in the present embodiment, in the present embodiment, The image capture module 12 also continuously generates an actual image for an area. That is, in the embodiment, the image capturing module 12 includes at least a thermal imager and a general imaging device for continuously capturing thermal images and actual images on the detected area, wherein the imaging device captures Each actual image corresponds to each thermal image captured by the camera.

In the embodiment, the image deconstruction module 14 is configured to respectively divide a thermal image of an area and an actual image into a plurality of mesh images according to a preset coordinate system. Similarly, in the embodiment, the image deconstruction module 14 divides the thermal image and the actual image captured by the image capturing module 12 into a plurality of mesh images by using a customized coordinate system, and The thermal image and each of the grid images in the actual image are positioned with coordinates of the coordinate system. It should be noted that, by means of spatial coordinate conversion, the coordinates of the thermal image and each of the mesh images in the actual image can be converted into position information in real space, where The details of the coordinate conversion should be familiar to those of ordinary skill in the art to which the present invention pertains, and thus will not be described herein.

The working mechanism of the feature recognition module 26 and the image restoration module 28 is another difference in the working mechanism of the body temperature abnormality detecting system provided by the embodiment and the body temperature abnormality detecting system provided by the foregoing embodiment.

Referring to FIG. 5A and FIG. 5B, FIG. 5A and FIG. 5B respectively show thermal images and actual images taken at the same time for a region. In this embodiment, when the error elimination module 16 determines the target thermal image (as shown in FIG. 5A), the target thermal image is received by the image analysis module 22 in the processing module 20; Corresponding to the time when the thermal image of the target thermal image is captured and the coordinates of each mesh image in the target thermal image, the feature recognition module 26 finds the corresponding mesh images in the actual image captured at the same time (as shown in the figure). Feature E at 5B) to characterize the mesh images.

In this embodiment, since the feature recognition module 26 recognizes the feature of the mesh image as the feature recognition of the character, the feature recognition module 26 first restores the mesh image to a three-dimensional image to This 3D image is characterized. For example, the feature recognition module 26 preferentially identifies the three-dimensional image for the facial features. However, if the feature recognition module 26 does not recognize the facial features in the three-dimensional image, the feature recognition module 26 will then target The skull features to characterize this three-dimensional image. However, the manner in which the feature recognition module 26 performs feature recognition on the three-dimensional image is merely illustrated herein, and is not intended to limit the present invention. Based on the identified biometric features (e.g., facial features or skull features), the decision module 24 can compare the identity of the individual from a database (not shown). It should be noted that the database stores a plurality of identity information and corresponding biometrics, and may be a built-in database built in the body temperature abnormality detecting system of the embodiment, or may be an external database. For example, the body temperature abnormality detecting system of the present embodiment can be connected to the medical record database of each medical institution.

Furthermore, considering the situation in which the body temperature abnormality occurs due to physical discomfort, it is reasonable that the individual cannot move too large a distance in a short time, and understandably, the individual's moving path has continuity. . Therefore, in the embodiment, the displacement retrieval module 27 records the displacement distance of the target thermal image in the region after a predetermined time by using the coordinate system set in advance. Next, the determining module 24 determines whether there is a target thermal image having a character feature determined by the feature recognition module 26 within the displacement distance. Further, if the displacement of the target thermal image in the region after the preset time exceeds the preset distance, the determining module 24 determines that the target thermal image is not an image of an individual whose body temperature is abnormal. Therefore, the processing module 20 stops processing the related data and data of the target thermal image.

In addition, in order to be able to find out the current appearance of the individual with abnormal body temperature, so that the medical staff can quickly find it in the waiting room, in this embodiment, the image restoration module 28 will find the feature recognition module 26 The mesh images in the actual image are subjected to image restoration to generate an image of the individual whose body temperature is abnormal. That is to say, in this embodiment, the warning message may include the coordinates in the actual space converted by the target thermal image in the preset coordinate system, the temperature of the target thermal image (ie, the body temperature of the abnormal body temperature), and In addition to the diseases that individuals with abnormal body temperature may have, the warning message may also include the identity of the individual with abnormal body temperature and its current appearance.

In the following description, the body temperature abnormality detecting method which can be executed in the body temperature abnormality detecting system which can be illustrated in FIG. 1 and FIG. 4 will be described in the following description, so please refer to FIG. 4 is understood in conjunction with Figures 5A-5B.

[An embodiment of a method for detecting abnormal body temperature]

Please refer to FIG. 6. FIG. 6 is a flowchart of a method for detecting an abnormal body temperature according to an exemplary embodiment of the present invention.

The body temperature abnormality detecting method provided in this embodiment can be applied to the body temperature abnormality detecting system shown in FIG. 1 . Referring to FIG. 1 , the body temperature abnormality detecting system illustrated in FIG. 1 includes the sensing module 10 and the processing module 20 . As shown in FIG. 1 , the sensing module 10 is connected to the processing module 20 . The sensing module 10 mainly includes a shadow The image capturing module 12 and the image determining module 24 and the determining module 24 are mainly included in the image capturing module 22 and the error removing module 16 . The image deconstruction module 14 is connected to the image capture module 12 , and the error elimination module 16 is connected to the image deconstruction module 14 . In addition, the image analysis module 22 is connected to the determination module 24. The body temperature abnormality detecting method provided in this embodiment can describe the following steps, as shown in FIG. 6.

In step S601, the image capturing module 12 continuously generates a thermal image for an area. In step S602, the image deconstruction module 14 divides the thermal image of the region into a plurality of mesh images according to a predetermined coordinate system, wherein each of the mesh images of the thermal image has a positioning by the coordinate system. A standard. In step S603, the error rejection module 16 is used to determine a target thermal image based on the grid image exceeding a threshold temperature. In step S604, the image analysis module 22 receives the target thermal image, and generates a relationship between temperature and time of each mesh image in the target thermal image. In step S605, the transmission determination module 24 determines whether there is an individual with abnormal body temperature in the region based on the relationship map. If the determination module 24 does not determine that there is an individual whose body temperature is abnormal in the area, then the process returns to step S603 to continue to determine the target thermal image; if the determination module 24 determines that the temperature of the target thermal image increases with time, then proceeds to step S606. . In step S606, the determining module 24 compares the temperature versus time relationship of each mesh image in the target thermal image with a plurality of pre-stored reference images to find the temperature of each mesh image in the target thermal image. A reference diagram that corresponds to the time diagram. Finally, proceeding to step S607, in step S607, the determining module 24 determines that there is an individual whose body temperature is abnormal in the area, and then according to the coordinates of each grid image in the target thermal image, the determining module 24 sends a warning message to report The individual whose body temperature is abnormal is located in the area.

It should be noted that the relevant details of each step S601-S607 in the method for detecting body temperature abnormality provided by this embodiment have been described in the foregoing related descriptions of FIG. 1 to FIG. 3, and the details are no longer detailed. Said.

[Another embodiment of the method for detecting abnormal body temperature]

Please refer to FIG. 7. FIG. 7 is a flowchart of a method for detecting an abnormal body temperature according to another exemplary embodiment of the present invention.

The body temperature abnormality detecting method provided in this embodiment can be applied to the body temperature abnormality detecting system shown in FIG. 4 described above. Referring to FIG. 4 , the body temperature abnormality detecting system illustrated in FIG. 4 includes the sensing module 10 and the processing module 20 . As shown in FIG. 4, the sensing module 10 is connected to the processing module 20. The sensing module 10 mainly includes an image capturing module 12, an image deconstructing module 14 and an error eliminating module 16, and the processing module 20 mainly includes an image analyzing module 22, a determining module 24, and a feature identifying module 26, The displacement search module 27 and the image restoration module 28. The image deconstruction module 14 is connected to the image capture module 12 , and the error elimination module 16 is connected to the image deconstruction module 14 . In addition, the image analysis module 22, the feature recognition module 26, the displacement search module 27, and the image restoration module 28 are all connected to the determination module 24. The body temperature abnormality detecting method provided in this embodiment can describe the following steps, as shown in FIG. 7.

In step S701, the image capturing module 12 continuously generates a thermal image and an actual image for an area. In step S702, the image deconstruction module 14 divides the thermal image and the actual image of the region into a plurality of mesh images according to a preset coordinate system, wherein the thermal image and each of the actual images are divided into a plurality of mesh images. The grid images each have a coordinate positioned by the coordinate system. In step S703, the error elimination module 16 determines a target thermal image based on the mesh image exceeding a threshold temperature.

The maximum difference from the body temperature abnormality detecting method provided by the foregoing embodiment is that after the target thermal image is determined in step S703, the process proceeds to step S704A and step S704B to perform feature recognition on the target thermal image, thereby obtaining Identity information of individuals with abnormal body temperature.

In step S704A, the image analysis module 22 finds the corresponding mesh images in the actual image according to the time when the target thermal image is captured and the coordinates of each of the mesh images in the target thermal image. Next, in step S704B, the image analysis module 22 performs the The mesh image is restored to a three-dimensional image, and the feature recognition module 26 performs feature recognition based on the three-dimensional image.

Next, in step S705A, according to the coordinate system set in advance, the displacement search module 27 records the displacement of the target thermal image after a predetermined time in the region, and then in step S705B, the displacement retrieval module 2724 determines the target. Whether the displacement of the thermal image in the region after the preset time exceeds a predetermined distance. If the displacement search module 27 determines in step S705B that the displacement of the target thermal image after the preset time has elapsed in the region exceeds the preset distance, the process returns to step S703 to continue to determine the target thermal image. On the other hand, if the displacement search module 27 determines in step S705B that the displacement of the target thermal image after the predetermined time has elapsed in the region does not exceed the preset distance, the process proceeds to step S706A.

In step S706A, the image analysis module 22 receives the target thermal image and generates a relationship between temperature and time of each mesh image in the target thermal image. Next, in step S706B, the determining module 24 performs regression analysis on the relationship graph and obtains a temperature regression line, and determines whether the temperature of the target thermal image is greater than or equal to a warning temperature according to the temperature regression line. If it is determined in step S706B that the temperature of the target thermal image is less than the warning temperature based on the temperature regression line, the process returns to step S703 to continue to determine the target thermal image. On the other hand, if it is determined in step S706B that the temperature of the target thermal image is greater than or equal to the alert temperature based on the temperature regression line, the process proceeds to step S707.

In step S707, the determining module 24 compares the temperature versus time relationship of each mesh image in the target thermal image with a plurality of pre-stored reference images to compare the mesh images with the target thermal image. A diagram of the temperature versus time graph is shown in the reference figure. Finally, in step S708, the determining module 24 determines that there is an individual whose body temperature is abnormal in the region, and issues a warning message according to the temperature of the target thermal image and the coordinates of each grid image in the target thermal image to report the individual whose body temperature is abnormal. The location within the area and its abnormal body temperature.

It should be noted that the relevant details of each step S701~S703, S704A~S704B, S705A~S705B, S706A~S706B, S707~S708 in the body temperature abnormality detecting method provided by the embodiment are described in the foregoing. 2-4 and 5A-5B are not described in detail herein.

[Possible effects of the examples]

In summary, the body temperature abnormality detecting system and method provided by the present invention can automatically detect an individual with abnormal body temperature in an area, and know where the individual is currently located in the area, and obtain The actual image of the individual.

That is to say, through the body temperature abnormality detecting system and method provided by the present invention, in the waiting period, the medical staff can firstly know whether any of the persons currently in the waiting room have symptoms of abnormal body temperature; in addition, the medical staff It is also possible to further know the location of the person with abnormal temperature in the waiting room and the identity and appearance of the person. In this way, it will help medical staff to take immediate care of this person.

In addition, in the body temperature abnormality detecting system and method provided by the present invention, through the error eliminating module and the displacement searching module, the thermal image of the individual who is not abnormal in body temperature can be effectively excluded, so that the present invention provides The body temperature detection system and method for abnormal body temperature are more accurate and reliable.

The above description is only an embodiment of the present invention, and is not intended to limit the scope of the invention.

Claims (10)

A body temperature abnormality detecting system for detecting whether a body temperature abnormality exists in an area includes: a sensing module comprising: an image capturing module for continuously generating a heat for the area An image decoupling module is coupled to the image capturing module for dividing the thermal image of the region into a plurality of mesh images according to a preset coordinate system, wherein each mesh of the thermal image The image has a coordinate positioned by the coordinate system; and an error elimination module is connected to the image deconstruction module to determine a target thermal image according to the mesh image exceeding a threshold temperature; and a processing module, connecting The sensing module includes: an image analysis module, configured to receive the target thermal image to generate a relationship between temperature and time of each mesh image in the target thermal image; a determining module connected to the The image analysis module is configured to determine, according to the relationship diagram, whether the individual has abnormal body temperature in the region, and if the individual has abnormal body temperature in the region, the determining module is configured according to the target The coordinate of each grid image in the thermal image sends a warning message to report the position of the individual whose body temperature is abnormal in the area; a displacement retrieval module, according to the coordinate system preset, the displacement retrieval module Recording a displacement of the target thermal image after a predetermined time in the area, and determining whether the displacement of the target thermal image in the area after the preset time exceeds a preset distance; and a feature recognition module, The image analysis module finds the corresponding mesh images in the actual image according to the time of the target thermal image generation and the coordinates of each of the mesh images in the target thermal image, and the mesh images are captured. The image is reduced to a three-dimensional image, and the feature recognition module performs feature recognition on the three-dimensional image for the facial feature or the skull feature. The body temperature abnormality detecting system according to claim 1, wherein the determining module determines the temperature of each mesh image in the target thermal image when the determining module determines whether the individual has abnormal body temperature in the region. The relationship map of time is compared with a plurality of pre-stored reference maps to compare one of the reference maps that match the temperature versus time of each of the grid images in the target thermal image. The body temperature abnormality detecting system according to claim 1, wherein when the error eliminating module determines the target thermal image according to the mesh image exceeding the threshold temperature, the error eliminating module will exceed the threshold. The grid images adjacent to the temperature are combined to determine the target thermal image, wherein the number of adjacent grid images exceeding the threshold temperature is greater than or equal to a threshold number. The body temperature abnormality detecting system of claim 1, wherein the image capturing module continuously generates an actual image for the area, each of the actual images corresponding to each of the thermal images, and the image deconstructing The module divides the actual image of the area into a plurality of grid images according to the coordinate system preset, wherein each grid image of the actual image has a coordinate positioned by the coordinate system; wherein the processing module Further comprising an image restoration module, when the determining module determines that the temperature of the target thermal image is greater than or equal to the warning temperature, according to the corresponding grid images in the actual image found by the feature recognition module, The image restoration module performs image restoration on the mesh images to obtain an image of the individual whose body temperature is abnormal. The body temperature abnormality detecting system of claim 4, wherein the determining module determines whether the body temperature abnormality exists in the area according to the relationship between the temperature of each mesh image in the target thermal image and time In the body, the determining module performs regression analysis on the relationship graph and obtains a temperature regression line to determine whether the temperature of the target thermal image is greater than or equal to a warning temperature according to the temperature regression line. An individual detection method for abnormal body temperature is applicable to a body temperature abnormality detecting system for detecting whether a body temperature abnormality exists in a region, and the body temperature abnormality detecting system comprises one sensing module connected with a processing module, the sensing module includes an image capturing module, an image deconstructing module and an error eliminating module, the image deconstructing module is connected to the image capturing module and the error eliminating module is connected In the image deconstruction module, the processing module includes an image analysis module, a determination module, a displacement retrieval module and a feature recognition module, and the determination module is coupled to the image analysis module, and the abnormal body temperature The detecting method includes: continuously generating a thermal image for the region through the image capturing module; and dividing the thermal image of the region into a plurality of meshes according to a preset coordinate system through the image deconstructing module An image, wherein each of the grid images of the thermal image has a coordinate positioned by the coordinate system; and the error exclusion module is used to determine the grid image according to a temperature exceeding a threshold And generating a target thermal image; and recording, by the displacement retrieval module, a displacement of the target thermal image in the region after a predetermined time according to the preset coordinate system; and determining the target heat through the displacement retrieval module Whether the displacement of the image in the region after the preset time exceeds a predetermined distance; and the image analysis module and the feature recognition module, according to the time of the target thermal image generation and each network in the target thermal image The coordinates of the image are used to find the corresponding mesh images in the actual image, and the mesh images are restored into a three-dimensional image to identify the three-dimensional image for facial features or skull features; An analysis module that receives the target thermal image to obtain the target a relationship between temperature and time of each grid image in the thermal image; and, through the determination module, determining whether the individual has an abnormal temperature in the region according to the relationship graph, and if there is an abnormal body temperature in the region And sending a warning message according to the coordinate of each grid image in the target thermal image to report the location of the individual whose body temperature is abnormal in the area. The method for detecting a body temperature abnormality according to claim 6 further includes: using the determining module, the relationship between temperature and time of each mesh image in the target thermal image and a plurality of pre-stored reference images Comparing, the comparison map is matched with the relationship between the temperature and the time of each grid image in the target thermal image; wherein the reference maps are based on the lesion scale of the plurality of specific temperature abnormal symptoms The resulting body temperature versus time graph. The method for detecting an abnormal body temperature according to claim 6, wherein in the step of determining the target thermal image based on the mesh image exceeding the threshold temperature, the error elimination module is configured to have a plurality of phases exceeding the threshold temperature. The neighboring grid image is combined to determine the target thermal image, wherein the number of adjacent grid images exceeding the threshold temperature is greater than or equal to a threshold number; wherein, the area is determined according to the relationship diagram In the step of whether the individual has abnormal body temperature, the determining module performs regression analysis on the relationship graph and obtains a temperature regression line to determine whether the temperature of the target thermal image is greater than or equal to a warning temperature according to the temperature regression line. The method for detecting a body temperature abnormality according to claim 6, further comprising: continuously generating an actual image for the area through the image capturing module, wherein each of the actual images corresponds to each of the thermal images; The image deconstruction module divides the actual image of the area into a plurality of grid images according to the coordinate system preset, wherein each grid image of the actual image has a coordinate positioned by the coordinate system. The method for detecting a body temperature abnormality according to claim 8, wherein the body temperature abnormality detecting system further comprises an image reducing module, and the method for detecting the body temperature abnormality further comprises: finding the module according to the feature identifying module The corresponding mesh images in the actual image are subjected to image restoration by the image restoration module to obtain an image of the individual whose body temperature is abnormal.