An AI monitoring method and system for hospital epidemic investigation scenarios based on image recognition

Abstract

The present application relates to an AI monitoring system for hospital flow survey scenes based on image recognition, which includes: using a camera and an infrared imaging device to respectively obtain a frame of high-definition image and a frame of infrared imaging at the same time; processing the high-definition image and the infrared imaging separately to form a first image and a second image with overlapping images; recognizing the first image and dividing it into multiple areas with a single face; using multiple areas to crop the second image to obtain multiple infrared blocks; converting multiple infrared blocks into grayscale images, and obtaining the grayscale value corresponding to each block; establishing a function of grayscale level and temperature value, and determining the temperature value of each infrared block according to the grayscale value corresponding to each block. It crops the second image according to the area where the first image is located, obtains an infrared imaging picture reflecting the face, and obtains the corresponding temperature in conjunction with the grayscale value, so as to cope with the problem of large flow of people in the hospital that is difficult to distinguish, and achieves real-time and accurate monitoring.

Description

Technical Field

[0001] The present application relates to the field of remote sensing detection technology, and in particular to an AI monitoring method and system for hospital epidemic investigation scenarios based on image recognition. Background Art

[0002] Gathering of people in public places is an important place for the spread of epidemic diseases, so temperature measurement in public places plays an important role in the prevention and control of epidemic diseases. How to use more stringent prevention and control measures on a large scale in public places to ensure accurate temperature measurement while improving temperature measurement efficiency and avoiding cross infection will be an effective prevention and control preparation for the doubling of population mobility pressure.

[0003] In some related technologies, infrared thermal imaging is a technology that converts infrared radiation of an object into a visible image. That is, the infrared radiation of the object is focused on the detector through the lens and converted into an electrical signal, which is then amplified and digitized and transmitted to the signal processor, and then converted into an image that can be recognized on the display. Generally, the temperature monitoring system with infrared thermal imaging cannot accurately and quickly identify the specific and accurate temperature of each object in the process of monitoring the temperature of the human body in the complex flow of people in the hospital, resulting in the monitoring being unable to accurately reflect the actual situation. This is a problem that needs to be solved urgently. Summary of the invention

[0004] The embodiments of the present application provide an AI monitoring method and system for hospital epidemic investigation scenarios based on image recognition to solve the problem that general thermal imaging methods in related technologies cannot accurately identify the body temperature of each object.

[0005] In the first aspect, an AI monitoring method for hospital epidemiological investigation scenarios based on image recognition includes the following steps:

[0006] A camera and an infrared imaging device are used to respectively obtain a frame of high-definition image and a frame of infrared imaging at the same time;

[0007] Processing the high-definition image and the infrared image separately to form a first image and a second image with overlapping images;

[0008] Recognizing the first image and dividing it into a plurality of regions each having a single human face;

[0009] Using the plurality of the regions to crop the second image to obtain a plurality of infrared blocks;

[0010] Convert the plurality of infrared blocks into grayscale images, and obtain the grayscale value corresponding to each block;

[0011] A function of gray level and temperature value is established, and the temperature value of each infrared block is determined according to the gray value corresponding to each block.

[0012] Preferably, the method further comprises the following steps:

[0013] Obtaining high-definition images and infrared images of corresponding frames at multiple adjacent moments taken by a camera and an infrared imaging device;

[0014] Repeat the steps of processing the one frame of high-definition image and the one frame of infrared imaging to form a first image and a second image with overlapping images; identifying the first image and dividing it into a plurality of regions with a single face; cropping the second image using the plurality of regions to obtain a plurality of infrared blocks; converting the plurality of infrared blocks into grayscale images and obtaining the grayscale value corresponding to each block; establishing a function of grayscale level and temperature value, and determining the temperature value of each infrared block according to the grayscale value corresponding to each block, to obtain the body temperature of each person in the high-definition image at different times;

[0015] Calculate the deviation between the body temperature of each person in the HD image at different times and the standard temperature;

[0016] Take the deviation value at the earliest moment and the deviation value at the last moment, and make the difference;

[0017] The obtained value is summed with the standard temperature to obtain the body temperature of each person in the final high-definition image.

[0018] Preferably, in the step of calculating the deviation between the body temperature of each person in the high-definition image at different times and the standard temperature, the deviation value needs to be multiplied by the ambient temperature compensation coefficient.

[0019] Preferably, the ambient temperature compensation coefficient is the ratio of the ambient temperature to the standard temperature, and the standard temperature is 36.5°C.

[0020] Preferably, establishing a function of gray level and temperature value specifically comprises the following steps:

[0021] Under the same ambient temperature, infrared imaging is performed on experimental subjects with different body temperatures to obtain multiple images, which are then converted into grayscale images;

[0022] Select a target area image corresponding to a human face in each grayscale image, and record the corresponding temperature of the human face;

[0023] Obtain the grayscale value of each target area image at the same ambient temperature;

[0024] Then change the ambient temperature and repeat the above steps to obtain the grayscale value of each target area image at different ambient temperatures;

[0025] Taking the maximum grayscale value and the minimum grayscale value corresponding to each target area image to form a grayscale level corresponding to the temperature of the target area image;

[0026] A function is calculated according to the obtained multiple gray levels and the corresponding temperatures.

[0027] In the second aspect, an AI monitoring system for hospital epidemiological investigation scenarios based on image recognition is provided, which includes:

[0028] The first module uses a camera and an infrared imaging device to respectively obtain a frame of high-definition image and a frame of infrared imaging at the same time;

[0029] The second module processes the one frame of high-definition image and the one frame of infrared image respectively to form a first image and a second image with overlapping images;

[0030] A third module recognizes the first image and divides it into a plurality of regions having a single human face;

[0031] A fourth module, using the plurality of the regions to crop the second image to obtain a plurality of infrared blocks;

[0032] The fifth module converts multiple infrared blocks into grayscale images and obtains the grayscale value corresponding to each block;

[0033] The sixth module establishes a function of gray level and temperature value, and determines the temperature value of each infrared block according to the gray value corresponding to each block.

[0034] Preferably, it also includes a display module, a storage and operation module and a power supply module; the storage and operation module is used to control the first module, the second module, the second module, the third module, the fourth module, the fifth module and the sixth module, and store the shooting monitoring data of the camera and the infrared imaging device;

[0035] The display module is connected with the storage and operation module and the power module, and displays the shooting and monitoring data.

[0036] Preferably, the storage and operation module is also connected to an alarm module. When a person's body temperature is higher than a set threshold, the storage and operation module drives the alarm module to issue an alarm and / or output an alarm prompt to a display screen.

[0037] In the third aspect, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, an AI monitoring method for hospital epidemiological investigation scenarios based on image recognition is implemented.

[0038] In a fourth aspect, a computer-readable storage medium is also provided, which stores an AI monitoring method for hospital epidemiological investigation scenarios based on image recognition.

[0039] The beneficial effects of the technical solution provided by this application include:

[0040] An embodiment of the present application provides an AI monitoring method for a hospital epidemic investigation scenario based on image recognition, which obtains a frame of high-definition image and infrared image at the same time through a camera and an infrared imaging device, and then processes the high-definition image and the infrared image respectively to form a first image and a second image with overlapping images; the first image is recognized and divided into multiple areas with a single face; the second image is cropped using the multiple areas to obtain multiple infrared blocks, and the second image is cropped according to the area where the face position of the corresponding person in the high-definition image layer is located, so that an infrared imaging picture reflecting the face can be accurately obtained to avoid the problem of difficulty in distinguishing due to too many people; and in a subsequent step, the multiple infrared blocks are converted into grayscale images, and the grayscale value corresponding to each block is obtained; a function of grayscale level and temperature value is established, and the temperature value of each infrared block can be determined in the function according to the grayscale value corresponding to each block, so as to accurately identify the body temperature of each object to cope with the problem of large flow of people in the hospital and achieve real-time and accurate monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 A flow chart of the AI ​​monitoring method for hospital epidemiological investigation scenarios based on image recognition provided in Example 1 of the present application. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. Example 1

[0044] The embodiments of the present application provide an AI monitoring method and system for hospital epidemic investigation scenarios based on image recognition to solve the problem that general thermal imaging methods in related technologies cannot accurately identify the body temperature of each object.

[0045] See also Figure 1 , an AI monitoring method for hospital epidemiological investigation scenarios based on image recognition, comprising the following steps:

[0046] S1, using a camera and an infrared imaging device to respectively obtain a frame of high-definition image and a frame of infrared imaging at the same time;

[0047] S2, processing a frame of high-definition image and a frame of infrared image respectively to form a first image and a second image with overlapping images;

[0048] S3, recognizing the first image and dividing it into a plurality of regions having a single human face;

[0049] S4, using the multiple regions to crop the second image to obtain multiple infrared blocks;

[0050] S5, converting the multiple infrared blocks into grayscale images, and obtaining the grayscale value corresponding to each block;

[0051] S6. Establish a function of gray level and temperature value, and determine the temperature value of each infrared block according to the gray value corresponding to each block.

[0052] A frame of high-definition image and infrared image at the same moment is obtained through a camera and an infrared imaging device, and then the high-definition image and the infrared image are processed separately to form a first image and a second image with overlapping images; the first image is recognized and divided into multiple areas with a single face; the second image is cropped using the multiple areas to obtain multiple infrared blocks, and the second image is cropped according to the area where the face position of the corresponding person is located in the high-definition image layer, so that an infrared imaging picture reflecting the face can be accurately obtained to avoid the problem of difficulty in distinguishing due to too many people; and in the subsequent steps, the multiple infrared blocks are converted into grayscale images, and the grayscale value corresponding to each block is obtained; a function of grayscale level and temperature value is established, and the temperature value of each infrared block can be determined in the function according to the grayscale value corresponding to each block, so as to accurately identify the body temperature of each object to cope with the problem of large flow of people in the hospital and achieve real-time and accurate monitoring.

[0053] In some preferred embodiments, the following steps are also included:

[0054] Obtaining high-definition images and infrared images of corresponding frames at multiple adjacent moments taken by a camera and an infrared imaging device;

[0055] Repeatedly process a frame of high-definition image and a frame of infrared imaging to form a first image and a second image with overlapping images; recognize the first image and divide it into multiple regions with a single face; use the multiple regions to crop the second image to obtain multiple infrared blocks; convert the multiple infrared blocks into grayscale images, and obtain the grayscale value corresponding to each block; establish a function of grayscale level and temperature value, and determine the temperature value of each infrared block according to the grayscale value corresponding to each block, so as to obtain the body temperature of each person in the high-definition image at different times;

[0056] Calculate the deviation between the body temperature of each person in the HD image at different times and the standard temperature;

[0057] Take the deviation value at the earliest moment and the deviation value at the last moment, and make the difference;

[0058] The obtained value is summed with the standard temperature to obtain the body temperature of each person in the final high-definition image.

[0059] In this embodiment, the time factor is added to better conform to the temperature changes of the human body at different times, making the detection more accurate.

[0060] In some preferred embodiments, in the step of calculating the deviation between the body temperature of each person in the high-definition image at different times and the standard temperature, the deviation value needs to be multiplied by the ambient temperature compensation coefficient.

[0061] The ambient temperature compensation coefficient is the ratio of the ambient temperature to the standard temperature, and the standard temperature is 36.5°C.

[0062] In some preferred embodiments, establishing a function of gray level and temperature value specifically includes the following steps:

[0063] Under the same ambient temperature, infrared imaging is performed on experimental subjects with different body temperatures to obtain multiple images, which are then converted into grayscale images;

[0064] Select a target area image corresponding to a human face in each grayscale image, and record the corresponding temperature of the human face;

[0065] Obtain the grayscale value of each target area image at the same ambient temperature;

[0066] Then change the ambient temperature and repeat the above steps to obtain the grayscale value of each target area image at different ambient temperatures;

[0067] Taking the maximum grayscale value and the minimum grayscale value corresponding to each target area image to form a grayscale level corresponding to the temperature of the target area image;

[0068] A function is calculated according to the obtained multiple gray levels and the corresponding temperatures.

[0069] It utilizes the corresponding relationship between the thermal imaging image and the corresponding temperature value, and uses the grayscale value for judgment, which can eliminate the interference of other colors and is more convenient for comparison and judgment.

[0070] This application also proposes an AI monitoring system for hospital epidemiological investigation scenarios based on image recognition, which includes:

[0071] The first module uses a camera and an infrared imaging device to respectively obtain a frame of high-definition image and a frame of infrared imaging at the same time;

[0072] The second module processes a frame of high-definition image and a frame of infrared image respectively to form a first image and a second image with overlapping images;

[0073] The third module recognizes the first image and divides it into a plurality of regions having a single human face;

[0074] A fourth module, using the multiple regions to crop the second image to obtain multiple infrared blocks;

[0075] The fifth module converts multiple infrared blocks into grayscale images and obtains the grayscale value corresponding to each block;

[0076] The sixth module establishes a function of gray level and temperature value, and determines the temperature value of each infrared block according to the gray value corresponding to each block.

[0077] In some preferred embodiments, it also includes a display module, a storage and operation module and a power module; the storage and operation module is used to control the first module, the second module, the second module, the third module, the fourth module, the fifth module and the sixth module, and store the shooting monitoring data of the camera and the infrared imaging device;

[0078] The display module is connected with the storage and operation module and the power module, and displays the shooting and monitoring data.

[0079] In some preferred embodiments, the storage and operation module is also connected to an alarm module. When a person's body temperature is higher than a set threshold, the storage and operation module drives the alarm module to issue an alarm and / or output an alarm prompt to a display screen.

[0080] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, an AI monitoring method for hospital epidemiological investigation scenarios based on image recognition is implemented.

[0081] A computer-readable storage medium stores an AI monitoring method for performing hospital flow survey scenarios based on image recognition. Specifically, computer-readable storage media include permanent and non-permanent, removable and non-removable media that can implement information storage by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer-readable storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable storage media do not include temporary computer-readable media (transitory media), such as modulated data signals and carriers. Example 2

[0082] Compared with the embodiment 1, the only difference is that it further includes: S7, identifying the shape edge of each infrared block, if there is an extended part in the edge area, it is determined that the object is not fully protected at this moment and is a susceptible person;

[0083] S8 obtains scene face feature values ​​for faces in the area of ​​a single face according to a face recognition algorithm, after receiving the information of the infected person, identifies the face of the infected person and obtains the face feature value of the source of infection by using the face recognition algorithm, and then queries whether there is a face feature value of the source of infection in the scene face feature value, if not, then the process ends; if so, then identify in the stored video data whether there is an object of the face feature value of the source of infection with an edge of the shape of the infrared block with an outward extension, if so, then the line connecting the object and the outward extension is converted into the first high-risk area; in the monitoring scene, a streamer that can float with the airflow is added, and the direction of the airflow can be known according to the inclination of the streamer, and after the object of the face feature value of the source of infection has an outward extension, the extension part in the direction of the airflow is marked as the second high-risk area;

[0084] S9: If there are susceptible persons in the first high-risk area, the susceptible persons are marked as the first key targets; if there are susceptible persons in the second high-risk area, the susceptible persons are marked as the second key targets.

[0085] The above method can characterize whether the protection of personnel is complete through the edge shape of the infrared block in the monitoring scene. If there is a situation such as adjusting the mask or sneezing, the edge shape will inevitably change, mainly because the area of ​​the infrared block becomes larger and has irregular extensions (non-perspective relationship). Some highly contagious diseases may even be spread through aerosols on the basis of droplet transmission. In this embodiment, through the situation of the infected person that has been known, the first key object and the second key object can be quickly determined according to the transmission ability of the virus, and the focus of the flow survey can be quickly determined. The performance of the airflow in the image of this embodiment can achieve more accurate control for highly contagious viruses, especially viruses that can be transmitted through aerosols.

[0086] In the description of the present application, it should be noted that the terms "upper", "lower", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0087] It should be noted that, in this application, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0088] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features applied for herein.

Claims

1. An AI monitoring method for hospital epidemiological investigation scenarios based on image recognition, characterized in that: The following steps are involved: Obtaining high-definition images and infrared images of corresponding frames at multiple adjacent moments taken by a camera and an infrared imaging device; Calculate the body temperature of each person in the HD image at different times; Wherein, obtaining the body temperature of each person at the same time includes the following steps: A camera and an infrared imaging device are used to respectively obtain a frame of high-definition image and a frame of infrared imaging at the same time; Processing the one frame of high-definition image and the one frame of infrared image respectively to form a first image and a second image with overlapping images; Recognizing the first image and dividing it into a plurality of regions each having a single human face; Using the plurality of the regions to crop the second image to obtain a plurality of infrared blocks; Convert the plurality of infrared blocks into grayscale images, and obtain the grayscale value corresponding to each block; Establishing a function of gray level and temperature value, and determining the temperature value of each infrared block according to the gray value corresponding to each of the blocks; Identify the shape edge of each infrared block. If there is an outward extension in the edge area, it is judged that the object in the infrared block is not fully protected at the moment and is a susceptible person; For the faces in the area of ​​a single face, a scene face feature value is obtained according to a face recognition algorithm. After receiving the information of the infected person, the face of the infected person is identified and the face feature value of the source of infection is obtained by using a face recognition algorithm. Then, it is queried whether there is a face feature value of the source of infection in the scene face feature value. If not, the process ends. If so, it is identified in the stored video data whether the object of the face feature value of the source of infection has an edge of the shape of the infrared block with an outward extension. If so, the line connecting the object and the outward extension is designated as the first high-risk area. A streamer that can float with the airflow is added to the monitoring scene. The direction of the airflow can be known according to the inclination of the streamer. After the object of the face feature value of the source of infection has an outward extension, the extension part in the direction of the airflow is marked as the second high-risk area. If there are susceptible persons in the first high-risk area, the susceptible persons are marked as the first key targets; if there are susceptible persons in the second high-risk area, the susceptible persons are marked as the second key targets.

2. The AI ​​monitoring method for hospital epidemiological investigation scenarios based on image recognition according to claim 1, characterized in that: Establishing the function of gray level and temperature value includes the following steps: Under the same ambient temperature, infrared imaging is performed on experimental subjects with different body temperatures to obtain multiple images, which are then converted into grayscale images; Select a target area image corresponding to a human face in each grayscale image, and record the corresponding temperature of the human face; Obtain the grayscale value of each target area image at the same ambient temperature; Then change the ambient temperature and repeat the above steps to obtain the grayscale value of each target area image at different ambient temperatures; Taking the maximum grayscale value and the minimum grayscale value corresponding to each target area image to form a grayscale level corresponding to the temperature of the target area image; A function is calculated according to the obtained multiple gray levels and the corresponding temperatures.

3. An AI monitoring system for hospital epidemiological investigation scenarios based on image recognition, characterized in that: It includes: The first module uses a camera and an infrared imaging device to respectively obtain a frame of high-definition image and a frame of infrared imaging at the same moment, as well as high-definition images and infrared imaging frames corresponding to multiple adjacent moments; The second module processes the one frame of high-definition image and the one frame of infrared image respectively to form a first image and a second image with overlapping images; A third module recognizes the first image and divides it into a plurality of regions having a single human face; A fourth module, using the plurality of the regions to crop the second image to obtain a plurality of infrared blocks; The fifth module converts multiple infrared blocks into grayscale images and obtains the grayscale value corresponding to each block; The sixth module establishes a function of gray level and temperature value, and determines the temperature value of each infrared block according to the gray value corresponding to each block; Calculate the body temperature of each person in the high-definition image at different times; identify the shape edge of each infrared block. If there is an outward extension in the edge area, it is judged that the object in the infrared block is not fully protected at the moment and is a susceptible person; For the faces in the single face area, the scene face feature value is obtained according to the face recognition algorithm. After receiving the infected person information, the face of the infected person is identified and the face feature value of the source of infection is obtained by using the face recognition algorithm. Then, it is queried whether there is a face feature value of the source of infection in the scene face feature value. If not, the process ends. If so, it is identified in the stored video data whether the object of the face feature value of the source of infection has an outwardly extending portion of the edge of the shape of the infrared block. If so, the line connecting the object and the outwardly extending portion is designated as the first high-risk area. Add a ribbon that can float with the airflow in the monitoring scene. The direction of the airflow can be known according to the inclination of the ribbon. After the object with the facial feature value of the infectious source has an outward extension, the extension part in the direction of the airflow is marked as the second highest risk area; If there are susceptible persons in the first high-risk area, the susceptible persons are marked as the first key targets; if there are susceptible persons in the second high-risk area, the susceptible persons are marked as the second key targets.

4. The AI ​​monitoring system for hospital epidemiological investigation scenarios based on image recognition as claimed in claim 3, characterized in that: It also includes a display module, a storage and operation module and a power module; the storage and operation module is used to control the first module, the second module, the third module, the fourth module, the fifth module, and the sixth module, and store the shooting monitoring data of the camera and the infrared imaging device; The display module is connected with the storage and operation module and the power module, and displays the shooting and monitoring data.

5. The AI ​​monitoring system for hospital epidemiological investigation scenarios based on image recognition as claimed in claim 4, characterized in that: The storage and operation module is also connected to an alarm module. When a person's body temperature is higher than a set threshold, the storage and operation module drives the alarm module to issue an alarm and / or output an alarm prompt to a display screen.

6. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements the AI ​​monitoring method for hospital epidemic investigation scenarios based on image recognition as described in any one of claims 1 to 2.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an AI monitoring method for executing a hospital epidemic investigation scenario based on image recognition as described in any one of claims 1 to 2.

Images