KR20090098595A - Artificial intelligence apparatus and method for auto identifying and tracing targets - Google Patents

Abstract

An intelligent apparatus for automatically discriminating and tracking a target and a method thereof are provided to detect the appearance of a plurality of targets automatically, grasp the sizes and positions of the detected targets, and then track the movement of the grasped targets. An intelligent apparatus for automatically discriminating and tracking a target comprises the first contactless-type multi-Infrared temperature sensor, the second contactless-type multi-Infrared temperature sensor and a controller(110). The first contactless-type multi-Infrared temperature sensor detects positions of appeared targets by observing a predetermined sensing area as moving according to a pattern. The second contactless-type multi-Infrared temperature sensor detects the sizes and positions of the targets sensed by the first contactless-type multi-Infrared temperature sensor.

Description

Artificial intelligence apparatus and method for auto identifying and tracing targets}

The present invention relates to an apparatus and a method for automatically recognizing, identifying and tracking a target, and in particular, the present invention uses two non-contact multi-infrared temperature sensors (hereinafter referred to as " IR sensors ") respectively corresponding to left and right eyes. An apparatus and method for artificial intelligence capable of automatically identifying and tracking a target in three dimensions.

Recently, due to the development of the electronic and telecommunications industry, companies and general households as well as major areas such as banks, museums, important cultural assets, and national industrial institutions can easily build security and security systems to capture real-time visitors and external intruders. In this case, there is a monitoring environment. Accordingly, various apparatuses and methods have been proposed to alert the operator with a plurality of sensing means for sensing various abnormal conditions of the surveillance area and means for transmitting and receiving a wired / wireless situation detected by the sensing means.

Looking at the monitoring and tracking method according to the prior art briefly, a method of detecting the approach of an object into a certain space using an access sensor, a method of detecting the position and movement of the object by image processing technology, the position of the target by detecting the direction of sound Detection methods are used, and more advanced technologies use radar, laser, and ultrasound to detect the appearance of a target and to detect its position and movement.

In addition, this method of detecting the target is applied to a surveillance robot, etc. to monitor a certain area in accordance with the operation of the user terminal in the area requiring security and crime prevention, if an abnormality occurs, track and capture the corresponding monitoring target and wirelessly output the monitoring result. A robot appeared. In addition, with the development of robot technology, the above-described target detection method is not only in the field of crime prevention but also its application scope such as cleaning, security, animal care, document receiving and sending, etc. is expanding to real life.

However, in the target sensing method using the proximity sensor according to the prior art, there is a problem that it is notified that the object is located in the designated detection zone, and does not provide accurate information on where the detected object is located in the detection zone. .

On the other hand, in the case of the target detection method using the image processing technology, the frame data of the image screen captured by the camera is input, and the motion vector of the object that appears through the comparison with the previously stored frame data is calculated, and the movement The object is tracked based on the motion vector information of the object region calculated by the vector calculation. However, since the target detection and tracking method according to the image processing technology tracks the detection and movement of the appearance of the object by the digital image processing such as the method of detecting the color change in the image and the edge detection, the large capacity Calculation of data is required, and thus there is a problem in that the time from photographing to sensing is very delayed.

In addition, Korean Patent Laid-Open Publication No. 2004-33986 (published on April 28, 2004) discloses a technology related to an "intelligent image security system that provides a distance and a moving direction of a diopter." According to this technology, two integrated infrared light cameras are mounted in parallel to form a stereo vision system in an imaging device having the same conditions as both eyes of a human body. By using an adaptive threshold value, a method of minimizing errors in extraction of a moving object and providing information such as distance, volume, and direction of the moving object using images acquired from two cameras is provided.

Looking at the techniques presented above, a surveillance camera is typically used to detect the position and direction of movement of the monitored object and use the detected result. Surveillance cameras are usually composed of a sensor that can detect a surveillance object and CCTV (Closed-Circuit Television) to shoot the notification of the movement of the surveillance object through the sensor to inform the monitor. In addition, when there is a movement in the monitoring object, a tracking sensor for detecting and tracking the movement of the monitoring object is separately configured or integrated with the detection sensor. In general, an infrared sensor is used. The infrared sensor detects a temperature emitted by a monitored object, compares the detected monitored object temperature with a reference temperature, and determines the appearance of the monitored object according to the comparison result.

The monitoring apparatus according to the prior art is a sensor for monitoring the position and movement of the object to be monitored, a control unit for receiving the signal of the sensor to calculate the position and the movement direction of the monitoring object to generate a control signal, according to the control signal of the controller It is composed of a driving unit for controlling the direction of the CCTV and a memory unit for recording and storing information about the image taken and the detected object through the CCTV.

However, the monitoring apparatus according to the prior art has a large direction of movement of the monitored object because the control unit receiving the detection signal of the first monitored object must determine the moving direction of the monitored object according to the signal of the sensor and transmit the same to the driving unit to control the driving unit. Or if the movement speed is high, it is impossible to track it quickly and accurately.

In addition, despite the fact that CCTV can not accurately capture the three-dimensional shape of the monitoring object, because unnecessary image data is included, there is an inefficient problem in capacity management because the processing capacity and storage capacity of the captured image are very large compared to the data required.

Accordingly, after installing at least a plurality of infrared sensors for detecting the monitored object through a temperature difference in different directions in a predetermined place, the installed infrared sensors identify the infrared sensor that detected the monitored object and monitor the detected detection. An apparatus for detecting an object position direction using an infrared sensor and a method of controlling the same have been proposed for detecting a direction in which an object is located, and setting the detected direction to a direction in which the object is located to perform a control operation accordingly.

In particular, Korean Patent Publication No. 2002-18515 (published on March 8, 2002) identifies an infrared sensor module that detects the object, and detects a direction in which the object is located by reading a light receiving range of the identified infrared sensor module. An apparatus and method are disclosed for measuring a distance to an object by adjusting a radiation position of a laser of a laser telemeter in the direction in which the detected object is located.

In addition, a method comprising a plurality of cameras, one camera to detect the appearance of the object, the other camera to track the movement of the detected object is disclosed in Korean Patent Laid-Open No. 2006-69799 (2006.06.22 published) , "Intelligent Boundary Robot System".

Meanwhile, Korean Unexamined Patent Publication No. 2004-63170 (published Feb. 16, 2006) discloses a technology related to a robot having an intrusion detection function and an intrusion detection method thereof. According to this technique, a robot having an intrusion detection function includes a speed sensor measuring a moving speed of a moving robot and at least first and second infrared sensors measuring and outputting infrared rays generated from an intruder in real time while the robot is moving. A plurality of infrared sensors are mounted, and after receiving distance information between the first and second infrared sensors, the first analog output signal measured by the first infrared sensor and the second measured by the second infrared sensor according to the movement of the robot. It receives the robot's moving speed measured in real time from the analog output signal and the speed sensor. The first and second analog output signals may be compared and detected based on the distance information and the moving speed to detect and notify the intrusion, thereby detecting the intrusion while the robot is moving.

However, since the detection apparatus according to the related art detects only the direction and distance when the object is detected in a flat manner, it detects the presence of the object and only traces the movement of the detected appearance, and thus cannot identify the object. There is this.

In addition, despite the fact that CCTV can not accurately capture the three-dimensional shape of the monitoring object, because unnecessary image data is included, there is an inefficient problem in capacity management because the processing capacity and storage capacity of the captured image are very large compared to the data required.

In addition, CCTV is not only to photograph the object, but also to capture the screen containing the object, it is not possible to provide enough information to accurately identify the object, or because it requires identification by the operator's naked eye Even if any object is detected by the temperature sensor, there is a problem in that the detected object cannot be automatically identified.

In addition, when the sensing and tracking device having the above problems is applied to the robot, not only the incompleteness of the sensing and tracking itself, but also ultimately, the robot may not elicit a natural response to the object to be detected and tracked.

Therefore, by integrating highly developed technologies in each field, such as existing spiritual processing technology, voice recognition technology, and non-contact multi-infrared temperature sensor, it automatically detects the appearance of a large number of objects and identifies them with a natural response. And intelligent devices and methods that can be tracked are desired.

Accordingly, an object of the present invention is to provide an intelligent device and method that can automatically detect the appearance of a plurality of targets, determine the size and location of the detected targets, and track the movement of the identified targets.

In addition, another object of the present invention is to provide an artificial intelligence device and method that can be accompanied by a natural response occurring to the tracker when detecting, identifying and tracking a plurality of targets.

In addition, another object of the present invention is to provide an artificial intelligence device and method capable of realizing the movement of the robot's face and eyes in response to various external changes occurring in the process of detecting, identifying and tracking a plurality of targets. It is.

An artificial intelligence device capable of automatically identifying and tracking a target according to the present invention for achieving the above objects includes: a first non-contact multi-infrared infrared sensor for observing a predetermined sensing area while moving according to a predetermined pattern to detect the position of appearing targets; Temperature sensor The position of the detected targets in accordance with the target detection signal of the second non-contact multi-infrared temperature sensor and the first non-contact multi-infrared temperature sensor for sensing the size and distance of the targets detected by the first non-contact multi-infrared temperature sensor A control unit which calculates the size and distance of the detected targets according to target detection signals of the second non-contact multi-infrared temperature sensor by controlling the second non-contact multi-infrared temperature sensor to move to the calculated positions of the targets. Characterized in that consisting of.

In addition, the artificial intelligence method capable of automatic target identification and tracking according to the present invention for achieving the above objects, by moving the first non-contact multi-infrared temperature sensor according to a predetermined pattern to scan a predetermined detection area, A process of receiving target detection signals generated during scanning of a detection area. A process of calculating the maximum temperature and two-dimensional coordinates of the targets detected from the received target detection signals. Receiving a detection signal from the second non-contact multi-infrared temperature sensor and calculating the size and distance of the targets; if there is a change in the sensing area where the targets are located, recalculating the coordinates of the changed targets; and Coordinates of the calculated targets and coordinates of the recalculated targets And calculating a direction of movement of the targets.

In addition, the artificial intelligence device capable of automatically identifying and tracking a target according to the present invention for achieving the above objects, the first non-contact type to detect the position of the targets appear by observing a predetermined detection area while moving according to a predetermined pattern Left and right ocular sensors each having a multi-infrared temperature sensor and a second non-contact multi-infrared temperature sensor for sensing the size and distance of the targets detected by the first non-contact multi-infrared temperature sensor. Sensor drive unit for driving the rotation of the pattern Device drive unit for driving the rotation of the upper and lower left and right and the circular pattern of the head mounted with the eyeball-shaped sensor to control the sensor driver and the device driver to drive the eyeball-shaped sensor according to a predetermined pattern, Targets received from the eye sensor And a control unit for calculating the size and position of the detected targets according to the detection signal.

In addition, an artificial intelligence device capable of automatically identifying and tracking a target according to the present invention for achieving the above objects is provided with a cylindrical shape on the outer circumferential surface of the video camera for photographing targets, moving in a circular pattern. The contactless multi-infrared temperature sensor for tracking the appearance of the field and the outer peripheral surface of the non-contact multi-infrared temperature sensor for tracking is provided in a cylindrical shape, moving in a circular pattern to detect the size and distance of the targets detected by the tracking non-contact multi-infrared temperature sensor It is characterized by consisting of a non-contact multiple infrared temperature sensor for identification.

As described above, the present invention has an effect of automatically detecting the appearance of a plurality of targets, determining the size and location of the detected targets, and tracking the movement of the identified targets.

In addition, the present invention has the effect of realizing the movement of the face and eyes of the robot actively responding to various external changes that occur in the process of detecting, identifying and tracking a plurality of targets.

In addition, when the present invention detects, identifies, and tracks a plurality of targets, the present invention expresses a natural response to the AI device, thereby providing a bond to the visitor and a threat to the intruder.

DETAILED DESCRIPTION Hereinafter, detailed descriptions of preferred embodiments of the present invention will be described with reference to the accompanying drawings. Specific details are set forth in the following description, which is provided to aid a more general understanding of the present invention. In the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

In the following description, representative embodiments of the present invention for achieving the above-described technical problem will be presented. Also, in the above embodiment, for convenience of description of the present invention, an apparatus and method for monitoring a predetermined space by using multiple IR sensors having a specific arrangement will be described. It does not limit the scope of the present invention having a monitoring device, it is of course applicable to a monitoring system, a humanoid robot, an insect-type robot, etc. having a similar technical background.

1 is an internal structure diagram of a non-contact multi-infrared temperature sensor applied to the present invention, Figure 2 is a view showing the detection range of the left / right eyes implemented by the non-contact multi-infrared temperature sensor of FIG.

As shown in Figures 1 and 2, the non-contact multiple infrared temperature sensor (hereinafter referred to as "IR sensor") applied to the present invention is composed of two IR sensors that serve as both left and right eyes. The IR sensor R, which plays the role of the right eye (or the left eye), has an array of 1 * 8 to detect the position of the target to be detected by observing a wide range of areas. At this time, the arrangement of the IR sensor (R) is not necessarily limited to the arrangement of 1 * 8, of course, may be configured in other arrangements as necessary. The IR sensor L, which plays the role of the left eye (or the right eye), has an array of 4 * 4 in order to further detect the location of the target detected by the IR sensor R. FIG. The arrangement of the IR sensor L is also not limited.

Alphabets (A through P) or Arabic numerals (1 through 8) corresponding to the respective subsensors constituting the IR sensors (L and R) respectively measure the temperature values of their respective sensing regions at a distance. Referring to FIG. 2, the IR sensor R detects the appearance of a target by scanning a wide range of regions up, down, left and right, and sensing a temperature different from a reference temperature. In addition, the IR sensor (L) detects a target by scanning the area of a relatively narrow range up, down, left and right and circularly to detect the area more subdivided. At this time, the vertical movement of the IR sensors (L, R) and the left and right movements are detected by driving the mounting portion (for example, the eyeball of the robot) on which the sensor is mounted in a periodic or aperiodic pattern according to a pre-programmed process. The area can be observed. According to the current state of the art, such an ocular sensor is capable of moving beyond the movement of the human eye. Since the mechanical operation and optical mechanism of such an ocular sensor are well known techniques, the description thereof is omitted.

3 and 4, an artificial intelligence method capable of detecting, identifying, and tracking a target according to the present invention will be described in detail. 3 is a diagram illustrating an AI method capable of detecting, identifying, and tracking a target according to an exemplary embodiment of the present invention, and FIG. 4 is a diagram for describing a method of obtaining two-dimensional and three-dimensional coordinates of a monitored object by a trigonometric function. Drawing.

As shown in FIG. 3, the target detection, identification and tracking method according to the present invention monitors the appearance of a target while scanning a tracking section while the IR sensor R moves up, down, left and right and in a circle. In this case, the movement of the IR sensor (R) may be implemented in a form in which the associated driving device moves according to the scanning pattern when the sensor is manufactured in an eyeball shape or when mounted in a circular driving device. The IR sensor (R) can track a wide range of targets. In addition, when implemented as an eye sensor and mounted on the eyeball of the humanoid robot, the head of the robot moves according to the scanning pattern, so that the sensor can obtain the same movement as the scanning pattern required for monitoring the corresponding tracking section.

FIG. 3 shows an example in which the IR sensor R senses a temperature rise in a sensing area in which the sub-sensors 6 and 7 are in charge while scanning for a wide range of target tracking. When the IR sensor (R) detects a temperature rise in the sensing area, it calculates the highest temperature and position (coordinate) of the sensing area. The position of the target calculated at this time is obtained as two-dimensional coordinates.

When the detection of the target of the IR sensor R and the calculation of the maximum temperature and position are completed, the head or left eye of the robot moves and the IR located in the detection area of the sub-sensors 4 and 5 of the IR sensor R moves. The sensing region of the sensor L moves to the sensing regions of the sub-sensors 6 and 7 of the IR sensor R on which the target is detected. In the case of implementing this in the form of an eyeball of the robot, the movement of the IR sensor L moves the pupil of the robot so that the IR sensor L moves to the detection area of the sub-sensors 6 and 7 of the IR sensor R. The head can then be programmed to move the head to look at the target.

Subsequently, the IR sensor L moving to the area where the target is located calculates the size and distance of the target using an angle formed by the detected target and its own angle, an angle formed by the IR sensor R, and a distance from the target. . The size and position of the target calculated at this time are obtained in three-dimensional coordinates.

4 is for explaining the mathematical basis for calculating the two-dimensional and three-dimensional coordinates of the target, using a trigonometric calculation method. As shown in FIG. 4, the distance (Eye L to Eye R) is a fixed value, and knowing the angles (Eye L and Eye R), the distance to the target A can be calculated. Can be. Since the IR sensors L and R applied to the present invention are array sensors, a plurality of Eye L and Eye R values may be obtained, respectively, and thus a plurality of coordinates of the target A may be obtained.

Meanwhile, in the description of FIGS. 3 and 4, the IR sensor R and the IR sensor L are sequentially moved in order to detect the target and calculate the temperature and position thereof. It should be noted that the above description is based on the characteristics of R), and the scanning operation may be simultaneously performed for the detection of the target.

The target detection, identification and tracking method of FIG. 3 may be applied even when there are a plurality of targets. Briefly, first, when two or more targets are detected by the IR sensor R, the temperature of each target is measured first. Then IR sensor (L) accurately measures the size and temperature of each target, and identifies the target based on the measured size and temperature.

In addition, tracking of the moving direction of the target according to the present invention is easily performed since the IR sensor applied to the present invention is an array sensor. When the identified target is moved, the movement of the target is easily tracked by this sensing information since the area of the subsensor where the target is detected varies over time. In addition, since the current position of the target is located at the center of the IR sensor, it is possible to predict the future direction of movement of the target.

5 and 6 are exemplary diagrams of an intelligent device capable of detecting, identifying, and tracking a target using the artificial intelligence method of FIG. 3. Hereinafter, target detection, identification, and detection according to the present invention will be described with reference to FIGS. 5 and 6. Various embodiments of an intelligent device employing an intelligent tracking method are described.

Figure 5 is an example of applying a single eye of an insect, such as a dragonfly to the artificial intelligence device according to the present invention. As shown in FIG. 5, an artificial intelligence device according to an embodiment of the present invention includes an image camera 1, an IR sensor for tracking 2, an IR sensor for identification 3, and a lens 4 for near-distance adjustment. It is composed of, and the front and side cross-sectional view is also shown. Reference numerals (A), (B) and (C) of FIG. 5 indicate a camera image range, a tracking range of the tracking IR sensor, and an identification range of the identification IR sensor, respectively.

The image camera 1 is provided with its own zoom function and optical zoom function using the near-distance adjustment lens 4. That is, the near-distance adjustment lens 4 or the image camera 1 may move back and forth so that an image of a target may be checked to a required size.

The movement of the IR sensors (2, 3) can adjust the radiation angle of the sensor while moving in a circular pattern, respectively. The primary adjusted radiation angle enables a wide range of tracking and identification through the near-far adjustment lens 4. In this case, the radiation angles of the respective IR cells of the IR sensors 2 and 3 have unique values, and may be variously changed using the lens 4 for rotation and near-distance adjustment.

6 is an embodiment in which the artificial intelligence device according to the present invention is applied to the eyes of an insect tentacle form. As shown in FIG. 6, the identification IR sensor and the tracking IR sensor are provided on both left and right eyes, and a camera is provided at the center, and a voice recognition sensor is provided on both left and right sides so that sound can be sensed in stereo with temperature. Configured. According to the embodiment of FIG. 6, it is possible to support 3D detection of a target including a voice, and the installation position of such an intelligent device may be installed up and down along with a dragonfly's eye shape.

Hereinafter, with reference to FIGS. 7 and 8, an artificial intelligence apparatus and method capable of automatically identifying and tracking a target according to the present invention including the above-described embodiments will be described. 7 is a functional block diagram of an intelligent device capable of detecting, identifying, and tracking a target according to the present invention, and FIG. 8 is a control flowchart of the intelligent method capable of detecting, identifying, and tracking a target according to the present invention.

Referring to FIG. 7, when the intelligent device according to the present invention is configured according to the functional unit, the controller 110 includes a control unit 110, a left IR sensor 121, and a right IR sensor 123 that control the AI device as a whole. The sensor unit 120, the sensor driver 130 for driving the sensor unit 130 under the control of the controller 110, and the apparatus 120 for performing another function such as a robot ( 170, a device driver 140 for driving the device 170 to drive the IR sensors, and a display unit 150 including a display device or a speaker for an operator and a key input device or a microphone. The input unit 160 is configured to. According to the above configuration, the intelligent device according to the present invention performs a target detection, identification and tracking method. In addition, although not shown in FIG. 7, it is preferable to display the detected target as an image to an operator including an image capturing unit as shown in FIGS. 5 and 6 showing embodiments of the present invention.

7 and 8, in the artificial intelligence method according to the present invention, the controller 110 first recognizes a target by receiving a target detection signal through scanning of a detection area of the IR sensors R and 123. This initial target recognition is performed through 201 scanning. The scanning in step 201 is performed according to the driving of the sensor unit 120 of the sensor driver 130 under the control of the controller 110. When the controller 110 detects a specific range of temperature from the IR sensors R and 123 in the specific sub-sensor in step 203, the controller 110 proceeds to step 205 to calculate the maximum temperature and the two-dimensional coordinates of the target.

Subsequently, the controller 110 proceeds to step 207 to control the sensor driver 130 to move the IR sensors L and 121 to the area where the target is detected by the IR sensors R and 123. When the IR sensors (L, 121) move to the area where the target is detected, the controller 110 receives the detection signal from the IR sensors (L, 121) in step 209 to coordinate the size and distance of the target in three-dimensional coordinates. Calculate exactly as

The size, distance, and temperature of the target calculated in step 209 may be provided to an operator through the display unit 150, and the statistical information on the size, distance, and temperature of the target is provided in advance. It can provide predictions about the identity of the target on which it is based. In addition, although not shown, a CCTV, an image camera, etc. may be provided to accurately capture the target based on the calculated information on the target, recognize the voice from the target, and provide the operator with the captured image and the recognized voice. It may be.

Next, in step 213, if there is a change in the sensing cell of the target (the sensing area of the sub sensor), the controller 110 recognizes it from the signals input from the IR sensors, and proceeds to step 215, wherein the sensor driver The control unit 130 may move the IR sensors R and L to the corresponding cell, and predict the next moving direction of the target from the trend of the moving direction of the target.

On the other hand, the control unit 110 controls the device driving unit 140 to move the device equipped with the sensor unit 120, when the IR sensor is mounted on the pupil of the robot, such as moving the head of the robot The control may be performed to obtain the same result as that of the sensor driver 130.

In addition, when the IR sensor according to the present invention is mounted as a pupil on a robot or the like, the artificial intelligence device and method according to the present invention can detect the natural reaction generated when the robot detects, identifies, and tracks a plurality of targets. Allow movement to be represented.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the scope of the following claims, but also by the equivalents of the claims.

1 is an internal structure of a non-contact multi-infrared temperature sensor applied to the present invention,

FIG. 2 is a diagram illustrating a detection range of left and right pupils implemented by the non-contact multi-infrared temperature sensor of FIG. 1;

3 is a diagram illustrating an AI method capable of detecting, identifying, and tracking a target according to an exemplary embodiment of the present invention;

4 is a view for explaining a method for obtaining two-dimensional and three-dimensional coordinates of a monitored object by a trigonometric function;

5 is an exemplary diagram of an intelligent device capable of detecting, identifying, and tracking a target to which the artificial intelligence method of FIG. 3 is applied;

6 is another exemplary diagram of an AI device capable of detecting, identifying, and tracking a target to which the AI method of FIG. 3 is applied;

7 is a functional block diagram of an intelligent device capable of detecting, identifying and tracking a target according to the present invention;

8 is a control flow diagram of an AI method capable of detecting, identifying and tracking a target in accordance with the present invention.

Claims (13)

An intelligent device capable of automatically detecting, identifying and tracking targets, The first non-contact multi-infrared temperature sensor that detects the position of the appearing targets by observing a predetermined sensing area while moving according to a predetermined pattern. A second non-contact multi-infrared temperature sensor for sensing the size and distance of the targets detected by the first non-contact multi-infrared temperature sensor; The position of the detected targets is calculated according to the targets detection signal of the first non-contact multi-infrared temperature sensor, and the second non-contact multi-infrared temperature sensor is controlled to move to the calculated positions of the targets, thereby the second non-contact multiple An intelligent device capable of automatically identifying and tracking targets, comprising a control unit for calculating the size and distance of the detected targets according to the target detection signal of the infrared temperature sensor. The method of claim 1, Image taking unit for shooting the targets according to the size and distance of the targets calculated by the controller A display unit which displays an image photographed by the image photographing unit; And an input unit for inputting a key input signal for controlling the control unit. The method of claim 2, A voice recognition sensor for recognizing the voices of the targets calculated by the controller and converting the voices into audio signals; And a speaker configured to output the audio signal converted by the voice recognition sensor as voice. In the artificial intelligence method that can automatically detect, identify and track the targets, Moving the first non-contact multi-infrared temperature sensor according to a predetermined pattern to scan a predetermined sensing area and receiving a target sensing signal generated during scanning of the sensing area Calculating the maximum temperature and the two-dimensional coordinates of the detected targets from the received target detection signal Moving a second non-contact multi-infrared temperature sensor to the coordinates of the calculated targets Calculating a size and a distance of targets by receiving a detection signal from the second non-contact multi-infrared temperature sensor Recalculating the coordinates of the changed targets if there is a change in the sensing area where the targets are located; and Calculating the direction of movement of the targets from the coordinates of the calculated targets and the coordinates of the recalculated targets. The method of claim 4, wherein Accumulating the resultant values of the calculated moving directions of the targets, and predicting the next moving direction of the targets from the accumulated resultant. . The method of claim 4, wherein And a step of photographing according to the size of the calculated targets based on the coordinates of the calculated targets. The method of claim 4, wherein Accumulating the resultant values of the calculated moving directions of the targets, and predicting the next moving direction of the targets from the accumulated resultant. . The method of claim 4, wherein Recognizing the voices of the calculated targets, and outputting the recognized voices further characterized in that the automatic detection and tracking of the targets. An intelligent device capable of automatically detecting, identifying and tracking targets, A first non-contact multi-infrared temperature sensor for observing a predetermined detection area while moving according to a predetermined pattern to detect positions of appearing targets; Left and right eyeball sensors, each with two non-contact, multiple infrared temperature sensors Sensor driving unit for driving the rotation of the vertical left and right and the circular pattern of the left and right eye sensor Device driver for driving the rotation of the upper and lower left and right and the circular pattern of the head mounted with the eye sensor The sensor driver and the device driver are controlled to drive the eye sensor according to a predetermined pattern to scan a predetermined surveillance area, and to detect the size and position of the detected targets according to target detection signals received from the eye sensor. Artificial intelligence device capable of automatically identifying and tracking the target consisting of a control unit for calculating. The method of claim 9, wherein the control unit, A first non-contact multi-infrared temperature sensor of the eyeball sensor controls the sensor driver to provide a movement according to a predetermined pattern for detecting the appearance of targets by observing the entire surveillance area; When the first non-contact multi-infrared temperature sensor detects targets, the second non-contact multi-infrared temperature sensor of the eyeball sensor controls the sensor driver to move to the position of the targets detected by the first non-contact multi-infrared temperature sensor. , Intelligent device capable of automatically identifying and tracking targets, characterized in that the control unit is controlled to direct the head mounted with the ocular sensor to the sensed targets. An intelligent device capable of automatically detecting, identifying and tracking targets, Video camera to take images of targets It is provided in the outer peripheral surface of the video camera cylindrical, non-contact multi-infrared temperature sensor for tracking to detect the appearance of the target while moving in a circular pattern and A target formed of a cylindrical shape on the outer circumferential surface of the non-contact multi-infrared temperature sensor for tracking and comprising a non-contact multi-infrared temperature sensor for identification to detect the size and distance of the targets detected by the non-contact multi-infrared temperature sensor for tracking while moving in a circular pattern. AI devices with automatic identification and tracking. The method of claim 11, It is provided on the front of the image camera, the non-contact multi-infrared temperature sensor for tracking and the non-contact multi-infrared temperature sensor for identification, and further comprises a near-distance adjustment lens that can be moved back and forth automatically targets, characterized in that tracking This intelligent device is possible. The method of claim 11, It is provided on the front of the image camera, the non-contact multi-infrared temperature sensor for tracking and the non-contact multi-infrared temperature sensor for identification, and further comprises a near-distance adjustment lens that can be moved back and forth, automatic target identification and tracking This intelligent device is possible.

Images