KR101754154B1 - Fever Patient Monitering Syatem at Public Place by Using Multiple Band Camera and Statistical Sampling and Method thereof - Google Patents

Abstract

The present invention relates to a multi-band camera and a system for detecting a fever in a public place using statistical sampling, comprising a wide area sensing camera for photographing a large area of a public place, generating visible light images and infrared images of the captured images, And a zoom unit for zooming the lens to transmit a visible light image and an infrared image of the photographed image by photographing a small area of a public place, and a camera unit for photographing the photographed image based on the visible light image and the infrared image of a small area received from the camera unit. The feature points constituting the face are determined to detect the face region, the wearing accessories are detected, the age and gender of the subject are extracted to generate a constituent pixel set of the face region, and the actual resolution is calculated on the basis of the face constituent pixel set The infrared region of the calculated face region Dimensional distribution of pixel values on the basis of the calculated set of facial region constituting pixel sets, facial feature points and reference templates, and generates a two-dimensional distribution of pixel values according to the facial type of the object, Transforms the face image into a face image on a two-dimensional plane, calculates an estimated temperature based on the meta information of the transformed face area, and if the calculated estimated temperature exceeds the body temperature, A face detector for detecting a face by using a face detector based on a wide visible light image and an infrared image received from the camera to specify a plurality of subjects and assigning a unique ID to each of the plurality of subjects, They are created and deleted when entering and exiting a physically fixed frame, The distance between the individual subject and the camera and the distance between the camera image and the camera image are set to be the input values of the detected body area image size and the detected face size in the screen frame, Estimating an angle with respect to a plane, estimating and calibrating a disturbance factor such as an angle between the individual subject and a camera, an angle with respect to a camera image plane, and an ambient temperature, and correcting the actual body temperature to generate samples. The estimated temperature is generated to form an estimated temperature distribution, a statistical representative value (median value, average, etc.) is obtained, a queue of FIFO type consisting of the obtained statistical representative values is maintained, and a statistical representative value of a new subject is inputted Remove the oldest statistical representative value, By inserting ever representative value as the sample value in the queue proceed to update continuously calculates the distribution of the statistical representative value, and if the distribution is not less than the trash hold it, characterized in that the management server is configured to control so as to sound the alarm.

Description

TECHNICAL FIELD [0001] The present invention relates to a multi-band camera and a public place fever patient detection system using statistical sampling, and a fever patient detection method using the same. [0002]

The present invention relates to the detection of a fever patient using a wide area camera and a wide area wide area camera at a location where multiple are gathered. Generally, a fever patient can be detected by detecting the temperature of a subject by using an infrared camera that focuses on each individual. In this case, the CPU receives the image from the infrared camera, extracts the face region of the subject in the image, detects the temperature of the face region, and issues an alarm when the temperature is above the body temperature.

      A conventional technique related to the present invention is disclosed in Korean Patent No. 10-1355947 (published on Apr. 29, 2014). 1 is a block diagram of a portable temperature measuring device using the conventional infrared array sensor. 1, the portable temperature measuring device using the conventional infrared array sensor may include a camera module, an infrared array sensor module, and an electronic circuit module. For example, the camera module may include a lens unit 40, a CCD 42, and the like. The infrared array sensor module includes a thermopile array sensor 53, an amplifier 54, a low-pass filter 55, a first A / D converter 56, a second A / D converter 57, A signal processing unit 58, and the like. The electronic circuit module includes an automatic gain control unit 42, a digital signal processing unit 43, a CCD driving unit 44, a controller 45, a detecting unit 46, a superimposing unit 47, A switching unit 49, a display 50, a storage 51, a TAS driving unit 52, and the like. Here, the display 50 may be a small-sized LCD suitable for a portable terminal. In particular, in consideration of simplification of a key button and user's convenience, an LCD having a touch screen function Can be used. In this display 50, camera image information photographed by a CCD (Charge Coupled Device) 41 or thermal image information measured by the thermopile array sensor 53 may be selectively displayed, Alternatively, the image information and the thermal image information may be overlapped and displayed. Furthermore, an indicator generated by the image generation unit 48, which will be described later, may be superimposed on the display 50 together with the image information and / or the thermal image information. For this purpose, the controller 45 controls the operation of an OSD (On Screen Display) generating unit 48 according to a user key input or the like to display various menu and OSD images Is displayed on the display 50

Lock. For example, when the user selects only the forehead region of a person (subject) and wants to measure a thermal image, the controller 45 controls the image generation unit 48 so that the forehead corresponding to the human forehead shape And an indicator having an edge line is displayed on the display unit 50. On the other hand, the indicator has a profile corresponding to the entire or local shape of the subject. Here, the area in which the indicator is displayed defines a target area as a target area (entire or local area of the subject) where the user intends to measure an accurate temperature. The specific shape of the indicator and the information on the display area are determined by using the thermo-file array sensor 53, for example, in order to obtain a measurement distance and a measurement angle most suitable for measuring a forehead portion of a person, And is stored and managed in the storage 51 in advance. The controller 45 controls the display 45 so that the image information of the camera photographed by the CCD 41 or the thermal image information acquired by the thermopile array sensor is superimposed on the display 50 in a state where the indicator is displayed, And controls the switching unit 49. For example, the user appropriately adjusts the measurement distance and the measurement angle so that the forehead portion of the person displayed by the camera image is displayed according to the outline of the indicator. If the user pushes the thermal image measurement key button while the forehead portion of the person is properly matched to the indicator through the above process, the controller 45 controls the operation of the TAS driver 52 , Thereby enabling the thermopile array sensor 53 to be enabled. The electric signal outputted from the thermopile array sensor 53 passes through the amplifier 54, the low-pass filter 55, the first A / D converter 56 and the digital signal processor 58, And processed as information. This thermal image information primarily refers to the temperature value acquired from each of the plurality of thermopile sensors, and the controller 45 calculates the temperature of the thermopile sensor based on the actual sensor data measured by each thermopile sensor Value is calculated and converted to a temperature value. In addition, the controller 45 may display on the display 50 the thermal image with a different color according to the temperature value measured in pixel units. In this case, in order that each pixel value of the thermopile array sensor has a meaningful temperature value, the temperature inside the sensor is measured through a separate temperature sensor, and the temperature value measured by each sensor is compensated through the measured value. That is, it is assumed that the temperature inside the sensor is assumed to be the same as the ambient temperature, and the temperature value measured by the temperature sensor inside the sensor is regarded as the ambient temperature so that the temperature of each pixel (i.e., the corresponding infrared sensor) Correct the measured value. Conventionally, since a thermistor sensor having a nonlinear logarithmic function is used as the temperature sensor, the accuracy of temperature measurement of the sensor itself is low. However, in this case, the accuracy of temperature measurement of the sensor itself can be improved by using a PTAT (Propotional to Absolute Temperature) sensor that exhibits a linear characteristic of a linear function form. Using the PTAT sensor, the process of calibrating the temperature value is simpler. For example, a total of four PTAT sensors are mounted on each corner of the thermopile array sensor, and the measured values are corrected by using the average temperature value of the PTAT sensors as the temperature of the sensor itself. Finally, the average value of the four PTAT sensors is taken, but before the mean value is taken, the range of temperature of each sensor is measured. After all four sensors are within ± 0.2 ° C, it is determined that the sensors are stable and the average value is finally obtained. Value. On the other hand, in a state where thermal image information (i.e., thermal image of a subject) or image information (i.e., a camera image of a subject) is displayed on the display 50 together with an indicator on the display 50, Adjust the target area of the image of the subject to superimpose on the indicator. Thereafter, the temperature of the target area of the subject is measured.

The portable temperature measuring apparatus using the conventional infrared array sensor configured as described above can accurately measure the temperature of a specific part such as the forehead, the hand, and the ear of a subject. However, in a public place where a large number of people gather, There is a problem that it can not be detected. It is therefore an object of the present invention to photograph people in a public place where a large number of people are gathered, to detect a fever patient among the people photographed, and to issue an alarm. It is also an object of the present invention to detect the presence of a fever patient in a multiple genome using a statistical sampling method.

A multi-band camera of the present invention having the above-described object and a public-area heat sensor system using statistical sampling is composed of a wide-area sensing camera for capturing a large area of a public place and generating and transmitting visible light images and infrared images of the photographed images, A camera unit for photographing a small area of a public place by performing vertical and horizontal rotation and zooming of the lens in the main body, for transmitting a visible light image and an infrared image of the photographed image, a visible light image of a small area received from the camera unit, A feature point constituting a face of a subject is determined based on an image to detect a face region, a wearable accessory is detected, an age and a gender of a subject are extracted to generate a constituent pixel set composed of pixels of the face region, Based on set infrared resolution And generating a two-dimensional pixel value distribution based on the calculated facial feature set pixel set, facial feature feature point, and reference template when the calculated infrared real image resolution of the facial region is equal to or greater than the threshold, Dimensional distribution of a pixel value according to a shape of a reference face is projected on a frontal two-dimensional plane after performing morphological conversion into a reference face, and the meta information of the morphologically transformed face region is received to calculate an estimated temperature, And detects a face by detecting a face by applying a face detector based on a wide visible light image and an infrared image received from the camera and detects a plurality of subjects Give a unique ID to the subject, and the subject enters and exits the physically fixed screen frame The output values for the individual subjects are maintained by the tracking unit which is continuously operated when moving within the screen frame, and the sizes of the detected body region images and the detected face sizes The distance between the individual subject and the camera and the angle with respect to the camera image plane are estimated and the perturbation factors such as the angle between the individual subject and the camera and the angle with respect to the camera image plane and the ambient temperature are estimated and corrected, And generates an estimated temperature distribution of the samples among the plurality of discrete subjects in a manner of estimating and generating the estimated temperature distribution to obtain a statistical representative value (median value, average, etc.) and a FIFO type queue as the statistical representative value of the new subject is input, The oldest statistical representative value is removed and the most recent statistical representative value is inserted into the queue as a sample value. The update is continuously performed. The distribution chart for the statistical representative value is calculated. When the distribution is above the threshold value, And a management server.

The public-area heat-generating patient detecting system using statistical sampling and the heat-generating patient detecting method using the multi-band camera according to the present invention configured as described above has the effect of detecting a fever patient in a place where the public is gathered. Further, another effect of the present invention is that the individual fever patient belonging to multiple genus can be detected and detected through the wide area sensing camera and the narrow range sensing camera.

1 is a configuration diagram of a portable temperature measuring device using a conventional infrared array sensor,
FIG. 2 is a block diagram of an entire system for detecting a fever in a public place using statistical sampling with a multi-
3 is a detailed configuration diagram of a management server according to the present invention,
FIG. 4 is a control flowchart of a method for detecting a fever patient in the narrow-range detection camera according to the present invention.
FIG. 5 is a control flow diagram of a method for detecting a patient fever in a public place using statistical sampling with a multi-band camera of the present invention.
FIG. 6 is a view showing an individual subject infrared image photograph of a narrow-range detection camera applied to the present invention,
FIG. 7 is a view showing a picture of a screen image of a wide area sensing camera applied to the present invention,
8 is a graph of statistical distribution values for N samples applied to the present invention.

2 to 8, a multi-zone camera according to the present invention having the above-described objects, a public-area heat-generating patient detection system using statistical sampling, and a heat-generating patient detection method using the same will be described.

FIG. 2 is a block diagram of an entire system for detecting a fever in a public place using statistical sampling with a multi-band camera of the present invention. 2, the multi-band camera of the present invention and a public place fever patient detection system using statistical sampling are provided with a wide area sensing camera for photographing a large area of a public place and generating and transmitting a visible light image and an infrared image of the captured image, A camera unit 100 for transmitting a visible light image and an infrared image of a photographed image by photographing a small area of a public place by performing a vertical and a horizontal rotation and a zooming of the lens and a visible light image of a small area received from the camera unit Based on the infrared image, the feature points constituting the face of the subject are determined to detect the face region, the wearing accessories are detected, the age and sex of the subject are extracted to generate a constituent pixel set of the face region, The infrared actual color resolution is calculated on the basis of the calculated infrared real color resolution, Dimensional distribution of pixel values on the basis of the calculated set of facial area constituting pixels, the facial constituting minutiae, and the reference template when the infrared real color resolution of the area is equal to or greater than the threshold, Transforms the distribution into a face shape, projects it on a frontal two-dimensional plane, receives meta information of the transformed face area to calculate an estimated temperature, compares the calculated estimated temperature with a body temperature And detects a face by detecting a face by applying a face detector on the basis of a wide visible light image and an infrared image received from the camera unit to identify a plurality of objects, And when the subject enters and leaves a physically fixed screen frame, respectively When the moving object is moved within the screen frame, the output value of the individual object is maintained by the continuously operating tracking unit, and the body area image size and the detected face size detected in the screen frame are inputted, Camera distance and camera image plane, estimating and correcting the perturbation factors such as the distance between the individual subject and the camera, the angle with respect to the camera image plane, and the ambient temperature, and estimating and correcting the actual body surface temperature, And generates an estimated temperature distribution to obtain a statistical representative value (median value, average, etc.), maintains a FIFO type queue composed of the obtained statistical representative values, As the statistical representative value of the subject is input, the oldest statistical representative value And the most recent statistical representative value is inserted into the queue as a sample value to continue the update. The distribution of the statistical representative value is calculated to calculate the standard deviation value, and when the standard deviation exceeds the threshold value, the alarm is issued And an Internet network or a wireless communication network 300 for connecting the camera unit and the management server via a network. In the above description, the camera is composed of a wide area sensing camera and a narrow range sensing camera. However, the wide area sensing camera or the narrow area sensing camera may be composed of a visible light camera and an infrared camera.

3 is a detailed configuration diagram of a management server applied to the present invention. 3, the management server according to the present invention includes a transmission / reception module 210 for receiving a visible light image and an infrared image from a camera unit, and a transmission / reception module 210 for receiving a visible light image and an infrared image, The feature points constituting the face are determined to detect the face region, the wearing accessories are detected, the age and gender of the subject are extracted to generate a constituent pixel set of the face region, and the infrared actual resolution is calculated based on the face constituent pixel set Generates a two-dimensional pixel value distribution based on the calculated facial feature set pixel set, the facial feature feature point, and the reference template when the infrared real image resolution of the calculated face area is equal to or greater than the threshold, The two-dimensional distribution of pixel values according to the face type of the face Dimensional plane, and calculates the estimated temperature based on the meta information (such as temperature distribution, site-specific error characteristics, ambient temperature, subject age, etc.) of the morphologically transformed face region, (220) for providing an alarm through an alarm unit when the estimated temperature is higher than the body temperature, and a face detector for applying a face detector on the basis of the visible light image of a large area and the infrared image received from the camera unit Is generated and deleted when a plurality of subjects are identified and a unique ID is given to each of a plurality of subjects and a subject enters and exits a physically fixed frame, And the output value for the individual subject is maintained by the tracking unit operating as the subject A distance angle estimator 240 for estimating the distance between the individual subject and the camera and the angle with respect to the camera image plane based on the body area image size and the detected face size detected by the screen frame, A perturbation factor compensating unit 250 for estimating and compensating for perturbation factors such as the distance between the individual subject and the camera, the angle with respect to the camera image plane, and the ambient temperature; An individual object statistical processing unit 260 for generating an estimated temperature of each individual sample among a plurality of individual subjects in a manner of generating an estimated temperature distribution and obtaining statistical representative values (median, average, etc.) A first-in first-out (FIFO) queue of statistical representative values of individual subjects is maintained and a statistical representative value of a new subject is input A total object statistic processing unit 270 for continuously updating the statistical representative value by inserting the most recent statistical representative value into the queue as a sample value and calculating a standard deviation value for the statistical representative value, An alarm unit 280 for issuing an alarm when the standard deviation value of the statistical representative value is equal to or greater than the threshold or the estimated temperature of the individual subject is equal to or more than the body temperature, a display unit 285 for displaying visible light, infrared image, An object subject heat generating unit 220, a subject specifying unit 230, a distance angle estimating unit 240, a perturbation factor compensating unit 250, an individual subject statistic processing unit 260, And a main control unit 290 for controlling the statistical processing unit 270, the display unit 285, and the alarm unit 280.

4 is a control flowchart of a method for detecting a fever patient in a narrow-range detection camera according to the present invention. 4, the narrow-range camera of the narrow-range detection camera according to the present invention detects an individual subject gathered in a multi-set location, generates a visible light image and an infrared image of the photographed subject, and transmits the infrared light image to the management server A step S11 of detecting the face area, detecting the wearable accessory, and extracting the age and sex of the subject by determining the subject photographing angle and the feature point of the face on the basis of the received visible light image information and the infrared image information (S12), the management server creating (S13) a face area constituting pixel set composed of pixels based on the extracted face area, wearing accessory, age and gender information, (S14) of calculating the infrared realistic resolution of the face region based on the face realm (S15) judging whether or not the infrared real image resolution is equal to or greater than the threshold, and if the infrared real image resolution is equal to or greater than the threshold, determining whether the infrared real image resolution is greater than or equal to the threshold A step S16 of generating a two-dimensional distribution of pixel values by the management server, and a step of projecting the two-dimensional distribution of the pixel values according to the face type of the object to the front two-dimensional plane (S18) of calculating the estimated temperature on the basis of the meta information such as the face region temperature distribution, the error characteristic, the ambient temperature, the age of the subject, and the like, (S19) comparing the estimated temperature and the temperature of the body temperature with each other and providing an alarm when the estimated temperature and the body temperature are exceeded.

FIG. 5 is a control flowchart for a method for detecting a public place fever patient using statistical sampling with a multi-band camera of the present invention. 5, a method for detecting a heat source in a public place using statistical sampling with a multi-band camera according to the present invention comprises the steps of (S21) capturing a large area of a camera section and transmitting the generated visible light image and infrared image to a management server, A step S22 of detecting a face by detecting a face by applying a face detector on the basis of the received large-area visible light image and the infrared image, a step S22 of assigning a unique ID to each of the plurality of specified objects A step S23 of maintaining a temperature output value for an individual subject by a tracking unit that is continuously generated even when the subject moves into and out of a physically fixed screen frame and moves in the screen frame, A size of the body region image detected in the screen frame, and a detected face size, (S24) estimating and calculating an angle between the individual subject and the camera and an angle with respect to the camera image plane; estimating a perturbation factor such as the distance between the individual subject and the camera, the angle with respect to the camera image plane, (S25) of generating a corrected temperature by estimating an actual body temperature and generating an estimated temperature of the sample among a plurality of individual subjects in a manner of estimating the corrected body temperature to generate an estimated temperature distribution, and calculating a statistical representative value (Step S26) of obtaining the statistical representative value of the object, and maintaining the FIFO queue of the obtained statistical representative value and removing the oldest statistical representative value as the statistical representative value of the new object is inputted, The statistical representative value is inserted into the queue as a sample value, and the update is continuously performed. The distribution of the statistical representative value is calculated (S27) calculating a standard deviation value, and issuing an alarm when the standard deviation value is equal to or greater than the threshold value (S28).

6 is a photograph of an individual subject infrared image of the narrow-range sensing camera applied to the present invention. 6, an individual subject infrared image of the narrow-range sensing camera is an image captured by an infrared camera configured in a narrow-sense camera, and detects a face region in a subject of the infrared image and estimates a temperature of the face region It is possible to do so.

7 is a photograph of a screen image of a wide area sensing camera according to the present invention. In FIG. 7, a screen frame is generated for an individual subject in a place where the multiple is collected, and the screen frame is formed by continuously moving by the tracking unit. The face region is detected with respect to the individual subject, And the standard deviation value can be calculated from the distribution map by calculating the distribution map for all the individual subjects.

8 is a graph showing an example of a distribution chart for statistical representative values for N samples applied to the present invention. 8, the distribution of the statistical representative values of N samples applied to the present invention indicates that when the estimated temperatures of the individual subjects are the same, the standard deviation value becomes 0. When the estimated temperatures of the individual subjects are the same, the standard deviation value becomes 0, It is.

100: camera unit, 200: management server,
210: a sending / receiving module, 220: an individual subject heat processing section,
230: Subject identifying unit, 240: Distance angle estimating unit,
250: Perturbation factor compensating unit, 270: Overall object statistical processing unit

Claims (10)

delete A multi-zone camera and a public location fever patient detection system using statistical sampling,
Wide-area sensing camera that takes a wide area of public places, creates visible light image and infrared image of the photographed image and transmits it, and zooms the lens by zooming in and out. A camera unit 100 including a narrow-range detection camera for transmitting a visible light image and an infrared image of the image;
A transmission / reception module 210 for receiving a visible light image and an infrared image from the camera unit, and a feature point determining unit 220 for determining a feature point constituting a face of the subject based on the visible light image and the infrared image of small area received from the camera unit, And detecting wearable accessories, extracting the age and sex of the subject to generate a constituent pixel set of the facial region, calculating a pseudo resolution based on the set of facial region constituting pixels, Dimensional distribution of pixel values on the basis of the calculated set of facial region constituting pixel sets, facial feature points and reference templates, and generates a two-dimensional distribution of pixel values according to the facial type of the object, After the transformation is performed, it is projected onto the frontal two-dimensional plane, (220) for calculating an estimated temperature based on the meta information of the subject and providing an alarm through the alarm unit when the calculated estimated temperature exceeds the body temperature, A face detector is applied based on the visible light image of the area and the infrared image to detect a face to identify a plurality of subjects and to give a unique ID to each of the plurality of subjects and to enter the screen frame in which the subject is physically fixed A body specifying unit 230 for storing an output value for an individual subject by a tracking unit that is continuously operated when the body frame is moved within a frame, And the detected face size as input, the distance between the individual subject and the camera and A distance angle estimating unit 240 for estimating an angle with respect to the camera image plane, a perturbation factor compensating unit 250 for estimating and compensating for perturbation factors such as the distance between the individual subject and the camera, And an estimating unit configured to estimate the temperature of the individual sample based on the corrected temperature of the sample and to generate an estimated temperature distribution and to obtain a statistical representative value, The object statistic processing unit 260 maintains a first-in first-out (FIFO) queue having statistical representative values of the acquired individual subjects, and removes the oldest statistical representative value as a statistical representative value of a new subject is input And inserts the most recent statistical representative value into the queue as a sample value to continuously update the statistical representative An alarm unit 280 for issuing an alarm when the standard deviation value for the statistical representative value is equal to or greater than the threshold value or when the individual subject estimated temperature is equal to or higher than the body temperature, A display unit 285 for displaying a visible light image, an infrared image, data, and the like, and the transmission / reception module 210, the individual subject heat processing unit 220, the subject specifying unit 230, the distance angle estimating unit 240, A management server 200 including a compensation unit 250, an individual object statistic processing unit 260, an entire object statistic processing unit 270, a display unit 285, and a main control unit 290 for controlling the alarm unit 280;
And an Internet network or a wireless communication network (300) for connecting the camera unit and the management server via a network.
3. The method of claim 2,
Wherein the wide area sensing camera and the narrow area sensing camera are arranged in a manner that,
A visible light camera, and an infrared camera, and a system for detecting a public place fever patient using statistical sampling.
delete A multi-zone camera and a public location fever patient detection system using statistical sampling,
A camera unit 100 composed of a wide area sensing camera for photographing a large area of a public place and generating and transmitting a visible light image and an infrared image of the photographed image;
A transmission / reception module 210 for receiving a visible light image and an infrared image from the camera unit, and a face detector for applying a face detector based on the visible light image and the infrared image of a large area received from the camera unit to detect a plurality of objects A unique ID is assigned to an individual object among a plurality of objects, and when a subject enters and exits a physically fixed screen frame, it is created and deleted, respectively. In case of moving within a screen frame, An object specifying unit 230 for maintaining an output value for a subject, an image size of the body region detected by the screen frame, and a detected face size, The distance-angle estimating unit 240, A perturbation factor compensating unit 250 for estimating and compensating for a perturbation factor such as an angle with respect to a camera image plane and an angle with respect to a camera image plane, an ambient temperature, and a method of estimating the temperature corrected by the perturbation factor compensating unit An individual object statistic processing unit 260 for generating an estimated temperature distribution of each individual sample among a plurality of individual subjects to form an estimated temperature distribution and obtaining a statistical representative value, and a FIFO First Out) queue is maintained and the oldest statistical representative value is removed as the statistical representative value of the new subject is input. The most recent statistical representative value is inserted into the queue as the sample value, A total object statistic processing unit 270 for calculating a standard deviation value for the statistical representative value, An alarm unit 280 for issuing an alarm when the sub-deviation value is equal to or greater than the threshold, or when the estimated temperature of the individual subject is equal to or higher than the body temperature, a display unit 285 for displaying visible light, infrared image, data, An object specifying unit 230, a distance angle estimating unit 240, a perturbation factor compensating unit 250, an individual object statistical processing unit 260, an entire object statistic processing unit 270, a display unit 285, and an alarm unit 280 A management server 200 configured as a main control unit 290 for controlling the main control unit 290;
And an Internet network or a wireless communication network (300) for connecting the camera unit and the management server via a network.
6. The method of claim 5,
The wide-
Wherein the system comprises a visible light camera and an infrared camera.
delete delete A multi-zone camera and method for detecting a public area fever patient using statistical sampling,
A step S21 of photographing a large area of the camera section and transmitting the generated visible light image and the infrared image to the management server;
A step (S22) of detecting a plurality of objects by detecting a face by applying a face detector on the basis of a wide-area visible light image and an infrared image received by the management server;
A unique ID is assigned to an individual subject among the specified plurality of subjects, and when the subject enters and exits a physically fixed screen frame, the tracking unit is continuously generated and deleted even when the subject moves within the screen frame (S23) the temperature output value for the individual subject is maintained;
(S24) estimating and calculating an angle with respect to a camera image plane and a distance between an individual subject and a camera using the body area image size and the detected face size detected in the screen frame as inputs;
(S25) correcting the temperature by estimating a perturbation factor such as the distance between the individual subject and the camera, the angle with respect to the camera image plane, and the ambient temperature;
A step (S26) of generating an estimated temperature distribution of samples from among a plurality of discrete objects in a manner that estimates the corrected temperature as an actual body surface temperature to form an estimated temperature distribution and obtaining a statistical representative value;
The FIFO type queue composed of the obtained statistical representative values is maintained, the oldest statistical representative value is removed as the statistical representative value of the new object is input, and the most recent statistical representative value is inserted into the queue as the sample value to update Calculating a standard deviation value by calculating a distribution diagram for the statistical representative value continuously (S27);
And issuing an alarm when the standard deviation value is equal to or greater than the threshold value (S28);
A step S11 of photographing an individual subject gathered in a multi-set place, generating a visible light image and an infrared image of the photographed subject, and transmitting the visible light image and the infrared image to the management server, and a step S11 of receiving the visible light image information and the infrared image information (S12) of detecting a face area, detecting a wearing accessory and extracting the age and sex of a subject by determining a subject photographing angle and a feature point of the face on the basis of the detected face area, (S13) of generating a face area constituting pixel set composed of pixels on the basis of gender information and a step (S14) of calculating an infrared practical resolution of a face area based on the face area constituting pixel set by the management server A step (S15) of judging whether or not the face area infrared real image resolution is equal to or greater than the threshold, (S16) of generating a two-dimensional pixel value distribution based on a face region constituting pixel set, a face region feature point and a reference template when the infrared server resolution is equal to or greater than the threshold, (S17) of performing a morphological conversion of a two-dimensional distribution of pixel values according to a face type into a reference face shape and then projecting the two-dimensional distribution of the pixel value on a two-dimensional front face (S17) (S18) of calculating the estimated temperature based on the meta information such as the ambient temperature, the age of the subject, and the like, receives the visible light image and infrared image of a small area received from the camera and estimates the temperature of the individual subject The method comprising the steps of: (a) Detection method.
10. The method of claim 9,
A method for detecting a public fever patient using statistical sampling with the multi-
And a step (S19) of providing an alarm when the management server compares the calculated estimated temperature with the temperature of the body temperature and if the measured temperature exceeds the estimated temperature, providing the alarm.

Images