TWM635852U - Smart recognition device - Google Patents

Abstract

An intelligent identification device, which includes a monitoring connection module to capture image information of multiple target areas, an artificial intelligence processing module receives and stores the image information of the target area, and uses AI model to calculate and identify the image information of multiple target areas, including human figure recognition The module detects whether there is a humanoid intrusion in the image of the target area, and sends a humanoid notification. The fainting recognition module recognizes whether the image of the target area is fainted by a humanoid, and sends a fainting notification. The figure raises his hand for help and sends a notification of raising his hand for help. The fire and smoke recognition module recognizes whether there is smoke in the image of the target area and sends out a fire and smoke notification. The movement recognition module recognizes whether there is an object moving in the image of the target area and sends a motion Notification, the user sets the platform for the user to view image information through the wireless network.

Description

Smart identification device

The present invention relates to an intelligent identification device, in particular to an intelligent identification device with multiple identification functions.

In recent years, home security applications often use the combination of mobile phones and cameras to allow users to check in real time whether there is any abnormality inside or outside the house through the Internet, or to know in real time whether their doors and windows have been opened or damaged through reed switches, or through gas The detector detects whether the gas in the house is leaking, etc., so as to improve the safety of personal and property at home.

The traditional camera monitoring system needs to manually check the screen for a long time to avoid missing important information, which not only has low security protection effect, but also lacks practicability. The current technology has achieved that when the camera detects that a person appears within a certain range, it can automatically send out an alarm to notify the remote user. However, the existing cameras cannot specifically detect whether the object is a person, an animal or other movable objects, and the picture will also be affected by changes in ambient light, shadows or the surrounding environment, resulting in misjudgment and low accuracy. Those who are in other places need to go back to the scene to understand the actual situation, which is not practical and convenient, and if the misjudgment rate is too high, it is easier to cause the user to ignore the mentality and lose the meaning of real-time video monitoring. Therefore, the existing video detection There is still room for improvement. Furthermore, the function of automatic detection of abnormalities is to detect abnormal images through full-screen analysis. This method of shooting and detecting full-environment images will consume a lot of computing resources. Therefore, high-end monitoring devices are required, which may increase the construction of security control equipment. The purchase cost is not easily affordable by ordinary consumers.

In addition, in recent years, the fertility rate has gradually declined and the population has entered an aging population, which has increased the proportion of the population living alone. In order to provide appropriate home care, we provide automatic monitoring systems, human-machine interfaces, home security care systems and smart home environments. Application to give immediate first aid in the event of an accident. Therefore, in response to the appeal of a humanized human-machine interface, motion recognition methods have become a popular research technology. Since human actions are formed by a series of continuous postures, and each action includes the movement of many body parts, how to effectively identify it involves high-dimensional and high-complexity calculation analysis, which will increase action recognition. The difficulty of the method.

According to the known technology, it has not disclosed combined technologies such as indoor and outdoor human detection, people fainting, people raising their hands for help, smoke detection, motion detection, vehicle detection, and specific area frame detection at the same time. means, and it is impossible to forward instant messages or images to specific people through social media platforms. It does not have multiple functions, practicability and convenience, and cannot really solve problems such as home safety, living alone care, and gas leakage. Improved to provide users with a smart security environment guard device with AI smart identification.

In view of this, the purpose of the present invention is to provide an intelligent identification device, which includes a monitoring connection module, an artificial intelligence processing module, and a user setting platform.

The monitoring connection module is used to capture image information of a plurality of target areas, and the artificial intelligence processing module is connected with the signal of the monitoring connection module to receive and store the image information of the target area, and use the AI model to calculate and identify The image information of the plurality of target areas includes a human figure recognition module, a fainting recognition module, a hand-raised recognition module, a fire and smoke recognition module, and a movement recognition module. The human figure recognition module recognizes Whether the image of the target area is intruded by a doll, and a doll notification will be issued corresponding to the intruding doll. The fainted figure sends a faint notification, and the raised-hand-for-help recognition module recognizes whether there is a human-figure raising a hand for help in the image of the target area, and sends a raise-for-help notification corresponding to the raised-hand-like figure, and the fire and smoke recognition module identifies the Whether there is smoke in the image of the target area, and a fire smoke notification is issued corresponding to the smoke generation. The movement recognition module identifies whether there is an object moving in the image of the target area, and sends a movement notification corresponding to the moving object. The user sets the platform It is connected with the signal of the artificial intelligence processing module for the user to log in and view the image information through the wireless network.

Preferably, the artificial intelligence processing module further includes a vehicle identification module, which is used to identify whether a vehicle enters the image of the target area, and sends a vehicle notification corresponding to the entering vehicle.

Preferably, the artificial intelligence processing module further includes a person leaving identification module, which is used to identify whether there is a person leaving the image of the target area, and when it is recognized that the person leaving exceeds a preset time, a person leaving notice is issued.

Preferably, the artificial intelligence processing module further includes an electronic fence module for the user to enclose a warning zone in the image frame of the target area as a detection trigger of a specific monitoring position.

Preferably, the artificial intelligence processing module further includes a distress alarm module, which includes a distress bracelet worn on the limb of the user in the target area, and a warning light signal set in the target area , the distress bracelet is connected to the warning light signal, when the user presses the distress bracelet, the warning light emits a red display light, and the alarm module recognizes the red display light, and sends out a corresponding signal An emergency call for help.

Preferably, the distress alarm module further includes a buzzer arranged in the target area, when the user presses the distress bracelet, the buzzer sounds simultaneously, and the distress alarm module recognizes a red display The light signal corresponds to the emergency call for help.

Preferably, the artificial intelligence processing module further includes a detection module, which is used to identify whether the environment of the target area is abnormal, and when it is detected that the object activity in the target area exceeds a preset time or an abnormality occurs A detection reminder notification.

Preferably, the detection module is selected from one or a combination of cameras, anti-theft devices, smoke sensors, motion detection sensors, infrared sensors, smart sockets and the like.

Preferably, the fainting recognition module recognizes whether the image of the target area has a human figure fainted, and if there is no specific activity within a certain period of time, the fainting notification is issued.

Preferably, the smart identification device further includes a human-machine interface device, which includes a social platform or an instant messaging software, the social platform is Instagram, snapchat, FACEBOOK, Twitter, and the instant messaging software is LINE, WeChat, Messenger, QQ, SKYPE, Wechat, Whatapp, and the user setting platform is embedded in the social networking platform or instant messaging software.

Preferably, the intelligent identification device further includes an abnormal notification module connected with the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the One or more notifications for mobile notifications, notifying the user over the wireless network.

Preferably, the intelligent identification device further includes an abnormal notification module connected with the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the The mobile notification, one or more notifications of the vehicle notification, is notified to the user via the wireless network.

Preferably, the intelligent identification device further includes an abnormal notification module connected with the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the A moving notification, one or more notifications of the person leaving the notification, notifies the user over the wireless network.

Preferably, the intelligent identification device further includes an abnormal notification module connected with the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the One or more notifications of the mobile notification and the SOS notification are notified to the user through the wireless network.

Preferably, the intelligent identification device further includes an abnormal notification module connected with the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the The mobile notification, one or more notifications of the detection alert notification, are notified to the user via the wireless network.

The beneficial effect of this new model is that, through AI image recognition, it can detect 7 kinds of recognition functions such as human shape, fainting, raising hand for help, smoke, movement, vehicle, and personnel leaving, so that users do not need to check the image of the target area at any time, or do it by themselves. Go to the scene to confirm, but use the mobile phone to receive LINE, and the real-time pictures and videos sent by the remote computer to inform the user of the notification, so as to achieve the effect of detecting the alarm of the second place, and provide the user with a smart safety protection device in advance. , the smart recognition device can be connected with existing monitoring equipment, so that the camera can be easily upgraded with plug-and-play, and it can be upgraded to an AI smart camera with recognition function, which greatly reduces the installation cost and improves the competitiveness in the market. The smart recognition device at the designated place performs image recognition, and when an abnormality is triggered, it is sent to the mobile phone of the user bound to the device using LINE, which can truly protect the user's security and privacy from being leaked. Further, a single camera simultaneously It has multiple AI recognition functions, and then performs image detection, allowing users to switch on or off the required recognition functions with the icon switch on the mobile phone, which is a rare operation method in the market.

1: Monitoring connection module

2: Artificial intelligence processing module

3: User setting platform

4: Man-machine interface device

5: Exception notification module

15: Electronic fence module

16: Human figure recognition module

17: Faint recognition module

18: Raise your hand for help identification module

19: Flame and smoke identification module

20: Mobile identification module

21: Vehicle identification module

22: Personnel leaving identification module

23:Alarm distress module

231: Distress bracelet

232: warning lights

233: Buzzer

24:Detection module

Figure 1 is a schematic diagram illustrating a preferred embodiment of the new smart identification device; Fig. 2 is a schematic diagram illustrating the signal connection mode of an alarm module in the preferred embodiment; and Figs. 3-13 are schematic diagrams illustrating the implementation mode of the preferred embodiment.

The features and technical contents of the relevant patent applications of the present invention will be clearly presented in the following detailed description of the preferred embodiments with reference to the drawings.

Referring to Fig. 1, 2, and Fig. 3, it is a preferred embodiment of the new intelligent identification device, which includes a monitoring connection module 1, an artificial intelligence processing module 2, a user setting platform 3, and a human-machine interface device 4, and an abnormal notification module 5.

This new model enables a single camera to have multiple AI recognition functions at the same time, and then performs image detection, allowing users to switch on or off the required recognition functions with the icon switch on the mobile phone, which is a rare operation method in the market.

The installation structure of this new model is that the existing monitoring equipment (analog camera, or network camera) is connected with the intelligent identification device (also known as the intelligent identification box in Figure 3). Video Recorder, DVR) connection, each smart recognition device can connect 4 cameras (CH1~CH4). In this way, the user can connect the existing surveillance equipment to the smart identification device, so that the camera can be easily upgraded with plug-and-play, and it can be upgraded to an AI smart camera with identification function.

The monitoring connection module 1 is used to capture image information of a plurality of target areas. The setting field of the monitoring connection module 1 can be places such as homes, factories, offices, storefronts, laneways, parking lots, warehouses, laboratories, residences of elderly people living alone, etc., and then adjust the monitoring connection module 1 according to user needs. Detected video images, such as gate entrances, kitchen stoves, parking lot entrances, warehouse entrances, yard fences, front and rear doors of stores, etc. Herein, the user may be the owner, family member, etc. of the above setting field.

The artificial intelligence processing module 2 is signal-connected with the monitoring connection module 1 to receive and store the image information of the target area, and use the AI model to calculate and identify the image information of the plurality of target areas, which includes an electronic fence model Group 15, a human figure identification module 16, a fainting identification module 17, a hand-raised identification module 18, a fire and smoke identification module 19, a mobile identification module 20, a vehicle identification module 21, and a person Leaving identification module 22 , an alarm module 23 for help, and a detection module 24 .

With reference to FIG. 4 , the user setting platform 3 is signal-connected to the artificial intelligence processing module 2 for users to log in and view the image information through a wireless network. Further, the user logs in the user setting platform 3 through a handheld device or a computer, and can remotely set the image of any target area, and simultaneously activate or deactivate one or more identification modules. The user can log in to the user setting platform 3 through the local mode and the remote mode for setting. The local mode needs to use the same network as the device. If the user is in a foreign country, he can log in to the user setting platform 3 through the remote mode, and set the login account and password requirements to improve the security of the network connection sex. In addition, all users who can log in to the platform are listed, and the usage rights can be managed according to the account roles.

The man-machine interface device 4 includes a social networking platform or an instant messaging software, the social networking platform is Instagram, snapchat, FACEBOOK, Twitter, and the instant messaging software is LINE, WeChat, Messenger, QQ, SKYPE, Wechat, Whatapp, the user The setting platform 3 is embedded in the social networking platform or instant messaging software. Anyone who can publish text, pictures, video and other information channels can implement it, and it is not limited to this.

Users can log in to the user setting platform 3 through human-machine interface devices 4 such as mobile phones, desktop computers, notebook computers, and tablet computers. In this preferred embodiment, the user setting platform 3 is embedded in LINE, and the user can click on the smart identification box through the mobile phone to check the current image status of the target area, change the display name of the smart identification box, configure settings, and turn off automatic notification. The identification modes of different functions are described below The abnormal trigger notification issued by the group is sent to the mobile phone of the user bound to the device or to the computer through the abnormal notification module 5 through LINE, and the notification is a screenshot of the target area and a video of the event.

It is worth mentioning that this new model is different from the conventional identification equipment, which transmits data to the cloud for identification. There are security concerns about data leakage. When an exception is triggered, use LINE to send it to the mobile phone or computer of the user bound to the device, which can truly protect the user's security and privacy from leakage.

Referring to Figures 5 to 8, the new model has seven identification functions, such as human figure, fainting, raising hand for help, smoke, movement, vehicle, and personnel leaving. Users can log in to the user setting platform 3 to set the monitoring connection module 1 The abnormal trigger time, abnormal trigger sensitivity, alarm interval time, timing report, trigger alarm sound, event recording, warning area setting and other trigger conditions of the captured target area. This new smart identification device can be connected to 4 cameras, so that each Each camera has the above 7 recognition functions, and multiple combinations of recognition functions can be set, such as human figure + smoke, human figure + fainting + raising hand for help...etc. In actual implementation, the identification function can be purchased according to user needs.

In addition, each identification function can set different trigger conditions and schedules. The trigger time can be set in three time periods, such as 00:00~23:59 on weekends, 17:30~07:30 on weekdays, and 18:00~00:00, human figure recognition is set on weekends, and smoke recognition is set all the time. The timing report can be set to report every hour or every few hours. When the preset timing report time is reached, the image of each target area will be captured and sent to the mobile phones of multiple users through LINE.

With reference to Fig. 9, the electronic fence module 15 (also known as the fence) allows the user to frame a warning zone (the yellow dotted line box in the figure) in the image of the target area by itself, as a specific monitoring position detection trigger. The user can log in to the user setting platform 3 and select the camera that wants to set the warning zone, drag the image of the target area, and through the design of the electronic fence module 15, it can be placed in the yard of the home (such as As shown in Figure 9), a warning zone is set at the fence of the house, the entrance of the garage, the window of the factory, etc. When an abnormal image is detected in the warning zone, an abnormal notification is triggered. Preferably, seven identification functions, such as human figure, fainting, raising hand for help, smoke, movement, vehicle, and personnel leaving, can be used to frame the warning area in the image of the target area to meet different use requirements.

Referring to FIGS. 10 and 11 , the human figure recognition module 16 (also known as a human figure) identifies whether the image of the target area is intruded by a human figure, and sends a human figure notification corresponding to the intruding human figure. When the human figure recognition module 16 recognizes that there is a human figure in the target area, it captures the image of the human figure and uses a red square frame to surround the human figure (as shown in Figure 10). If the image of the human face is captured, It also enlarges the face and sends the photo, and sends the picture through LINE to notify the user immediately, as evidence of intrusion for subsequent alarms. If it is an event video, record a video of the event process about 30 to 50 seconds before the pedestrian appears in the target area and after the human figure appears (as shown in Figure 11). The user only needs to click the link in the notification to view it During the incident, when the user is in the field, he can know the actual situation without rushing back to the scene for investigation, and make follow-up disposal in the first time. It is particularly noted that the notifications issued by each identification module of this new model can be sent through LINE to send pictures or event videos to immediately notify users, and the following descriptions will not be repeated.

The function of the human figure recognition module 16 can be used when the owner is not at home and a thief breaks in. The owner can be notified immediately, and the alarm can be used to prevent the door from being broken into. Further, image screenshots and event recordings can be used as follow-up legal proceedings evidence. In addition, there is a false positive report link below the image screenshots and event recordings that can be sent to the user, which will be used for subsequent identification and analysis of the device to improve the identification accuracy of the device.

The fainting recognition module 17 (also known as fainting) recognizes whether the image of the target area has a fainted human figure, and sends a faint notification corresponding to the fainted human figure. When it is recognized that there is a fainted human figure in the target area, it will capture the image of the fainted human figure and use a red box to surround the human figure frame, and send the picture through LINE Notify users immediately. Wherein, the fainting recognition module 17 recognizes whether the image of the target area has a human figure fainted, and sends the fainting notification if there is no specific activity within a certain period of time. The certain time (also referred to as the fainting time in the figure) can be set for several hours or tens of minutes according to the usage requirements.

With reference to Fig. 12, it is a screenshot of a user raising his hand for help when a third party invades his home. The raising of his hand for help recognition module 18 (also known as raising his hand) recognizes whether the image of the target area has a human figure raising his hand for help, and responds to the raising of his hand for help. The phantom of the hand asking for help sends out a notification of raising a hand for help. The image of the puppet raising its hand for help is captured in the picture and surrounded by a red box, which immediately notifies the preset users and creates a safe living environment for users.

With reference to FIG. 13 , it is a screenshot of smoke detected in the office. The fire and smoke identification module 19 (also known as smoke) identifies whether the image of the target area has smoke, and sends a fire and smoke notification corresponding to the smoke generation, such as The image captures two red boxes produced by smoke. If the use situation is a fire at home, the fire and smoke recognition module 19 detects smoke, and sends pictures and videos through LINE to notify the user in real time, wake up the sleeping family members, and immediately extinguish the fire or escape immediately to achieve home safety The effect of protection.

The movement identification module 20 (also referred to as movement) identifies whether there is an object moving in the image of the target area, and sends a movement notification corresponding to the moving object. For example, a thief stealthily goes outside the fence of a house to survey the terrain to prepare for theft. The movement recognition module 20 detects a humanoid movement, takes a photo and video to notify the homeowner, and prevents it in advance.

The vehicle identification module 21 (also referred to as a vehicle) is used to identify whether a vehicle enters the image of the target area, and sends a vehicle notification corresponding to the vehicle entering. The warehouse is unattended at night on holidays or at night, and the warning area is set in the warehouse at the same time. The vehicle identification module 21 detects that a vehicle enters the warning area, and immediately sends photos and videos to the designated personnel. In addition, it can also activate sound or light alarm Stop suspicious persons.

The personnel leaving identification module 22 (also called personnel leaving) is used to identify whether there are personnel leaving the image of the target area, and when it is recognized that the personnel leaving exceeds a preset time, a personnel leaving notice is issued. The preset time (also known as the leaving time) can be set from a few minutes to a few hours according to the usage requirements. The personnel leaving identification module 22 can be used in offices or factories, if the employees leave without returning to their posts for more than 30 minutes, an alarm notification will be triggered.

If the monitoring connection module 1 is installed in the home of an elderly person living alone, in order to prevent the elderly from being sick or dead at home, the time of normal activities can be set as an abnormal trigger. If there are human-like activities in the image of the target area, it will trigger the notification to the family members or neighbors who do not live together. Further, through the electronic fence module 15, the bed bottom of the bedroom is selected as the warning area, and the sleep time is set as an abnormal trigger. If the elderly falls under the bed, an alarm will also be triggered to meet different use needs.

The alarm module 23 includes a distress bracelet 231 worn on the user's limbs in the target area, and a warning light 232 arranged in the target area, the distress bracelet 231 and the warning light 232 signal connection, when the user in the target area presses the distress bracelet 231, the warning light 232 emits a red display light, and the alarm distress module 23 recognizes the red display light, and correspondingly sends out an emergency The SOS notification allows the third party to know in time that the user wearing the SOS bracelet 231 is in urgent need of help, so as to obtain medical assistance within the prime time and avoid accidents.

Further, the alarm module 23 further includes a buzzer 233 arranged in the target area. When the user in the target area presses the distress bracelet 231, the buzzer 233 will sound at the same time, and the alarm calls for help. When the module 23 recognizes the red display light, it sends out the emergency call correspondingly.

The detection module 24 is used to identify whether the environment of the target area is abnormal, and when it detects that the object activity in the target area exceeds a preset time or an abnormality occurs, a detection reminder is issued. Wherein, the detection module 24 is selected from one or a combination of cameras, anti-theft devices, smoke sensors, motion detection sensors, infrared sensors, and smart sockets. For example, the user installs a camera next to the fire in the kitchen to point at the fire, and when the fire continues to heat for more than a preset time and no one takes care of it, an abnormal notification will be triggered. Another embodiment is that the monitoring and connection module 1 captures images inside the factory. The smart socket is installed in the boss's home and a buzzer is inserted. When the fire and smoke identification module 19 recognizes smoke, it triggers The buzzer 233 makes a sound, and at the same time transmits on-site photos and videos to the boss, ensuring that the user receives the notification at the first time, and achieves the effect of A detecting the alarm of B.

The abnormal notification module 5 is connected to the artificial intelligence processing module 2, and according to the identification results of each identification module, finds the object to be notified through the cloud mechanism, and reports to the outside through the Line or remote computer or handheld device. Further, according to one or more of the person-shaped notification, the fainting notification, the hand-raised notification, the fire smoke notification, the moving notification, the vehicle notification, the person leaving notification, the emergency SOS notification, and the detection reminder notification A notification is notified to the user through the wireless network.

To sum up, this new smart identification device uses AI model calculations to identify and detect 7 functions, including human form, fainting, raising hands for help, smoke, movement, vehicles, and people leaving, without the need for users to check the images of the target area at any time , or go to the scene to confirm by yourself, but use the mobile phone to receive the real-time pictures and videos sent by LINE and the remote computer to notify the user of the awareness, so as to achieve the effect of detecting the alarm of the second place, and provide the user with intelligent safety protection in advance Moreover, the smart identification device can be connected to existing monitoring equipment, so that the camera can be easily upgraded with plug-and-play, and it can be upgraded to an AI smart camera with identification function, which greatly reduces the installation cost, improves the competitiveness in the market, and uses In addition, the device can set multiple combinations of identification functions, and each identification function can be set with different trigger conditions and schedules to meet the needs of users with multiple functions. A single camera can simultaneously have multiple AI recognition function Can, and then carry out image detection, for the user to use the icon switch on the mobile phone to switch on or off the required identification function, which is a rare operation method in the market, so it can indeed achieve the purpose of the new model.

But the above is only a preferred embodiment of the new model, and should not limit the scope of the new model implementation, that is, all simple equivalent changes and modifications made according to the patent scope of the new model application and the new description content, All still belong to the scope that this new patent covers.

1: Monitoring connection module

2: Artificial intelligence processing module

3: User setting platform

4: Man-machine interface device

5: Exception notification module

15: Electronic fence module

16: Human figure recognition module

17: Faint recognition module

18: Raise your hand for help identification module

19: Flame and smoke identification module

20: Mobile identification module

21: Vehicle identification module

22: Personnel leaving identification module

23:Alarm distress module

24:Detection module

Claims (15)

An intelligent identification device, comprising: a monitoring connection module for capturing image information of a plurality of target areas; an artificial intelligence processing module connected with the monitoring connection module for receiving and storing the image information of the target area Image information, and use the AI model to calculate and identify the image information of the multiple target areas. The artificial intelligence processing module includes: a human figure recognition module, which identifies whether the image of the target area is intruded by a human figure, and sends a human figure notification corresponding to the intruding human figure ; a faint recognition module, which identifies whether the image of the target area has a human figure fainted, and sends a faint notification corresponding to the fainted human figure; a hand-raised for help recognition module, which identifies whether the image of the target area has a human figure who raises his hand for help, and A fire and smoke notification is issued corresponding to the human figure raising the hand for help; a fire and smoke recognition module identifies whether there is smoke in the image of the target area, and sends a fire and smoke notification corresponding to the smoke generation; a movement recognition module identifies the target Whether there is an object moving in the image of the area, and a movement notification is issued corresponding to the moving object; and a user setting platform is connected with the signal of the artificial intelligence processing module for the user to log in to view the image information through the wireless network. According to the intelligent identification device described in Claim 1, wherein the artificial intelligence processing module further includes a vehicle identification module, which is used to identify whether a vehicle enters the image of the target area, and sends a vehicle notification corresponding to the entering vehicle. According to the intelligent identification device described in claim 1, wherein the artificial intelligence processing module further includes a person leaving identification module, which is used to identify whether there is a person leaving the image of the target area, when it is recognized that the person leaving exceeds a preset Time to issue a personnel departure notice. According to the intelligent identification device described in claim 1, wherein the artificial intelligence processing module further includes an electronic fence module for the user to enclose a warning area in the image frame of the target area as a detection area for a specific monitoring position. test trigger. According to the intelligent identification device described in claim 1, wherein the artificial intelligence processing module further includes an alarm and emergency module, which includes a distress bracelet worn on the limb of the user in the target area, and a The warning signal set in the target area, the distress bracelet is connected to the warning signal signal, when the user presses the distress bracelet, the warning signal emits a red display signal, and the alarm module recognizes The red indicator light corresponds to an emergency call for help. According to the smart identification device described in claim 5, wherein the alarm module further includes a buzzer installed in the target area, when the user presses the distress bracelet, the buzzer sounds simultaneously, the When the alarm module recognizes the red display light, it sends out the corresponding emergency notification. According to the intelligent identification device described in claim 1, the artificial intelligence processing module further includes a detection module for identifying whether the environment of the target area is abnormal, and when it is detected that the object activity in the target area exceeds A detection reminder notification will be issued at a preset time or when an abnormality occurs. According to the smart identification device described in claim 7, wherein the detection module is selected from one or a combination of cameras, anti-theft devices, smoke sensors, motion detection sensors, infrared sensors, smart sockets, etc. According to the smart recognition device described in Claim 1, wherein the fainting recognition module recognizes whether the image of the target area has passed out, and sends the fainting notification if there is no specific activity within a certain period of time. According to the smart identification device described in claim 1, it further includes a human-machine interface device, which includes a social platform or an instant messaging software. The social platform is Instagram, snapchat, FACEBOOK, Twitter, and the instant messaging software is LINE, WeChat, Messenger, QQ, SKYPE, Wechat, Whatapp, and the user setting platform is embedded in the social networking platform or instant messaging software. According to the intelligent identification device described in claim 1, it further includes an abnormal notification module connected to the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the One or more notifications for mobile notifications, notifying the user over the wireless network. According to the intelligent identification device described in claim 2, it further includes an abnormality notification module connected to the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the The mobile notification, one or more notifications of the vehicle notification, is notified to the user via the wireless network. According to the intelligent identification device described in claim 3, it further includes an abnormality notification module connected to the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the A moving notification, one or more notifications of the person leaving the notification, notifies the user over the wireless network. According to the intelligent identification device described in claim 5, it further includes an abnormality notification module connected to the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the One or more notifications of the mobile notification and the SOS notification are notified to the user through the wireless network. According to the intelligent identification device described in claim 7, it further includes an abnormal notification module connected to the signal of the artificial intelligence processing module, according to the notification of the human figure, the notification of the fainting, the notification of raising the hand for help, the notification of the fire and smoke, the notification of the The mobile notification, one or more notifications of the detection alert notification, are notified to the user via the wireless network.

Images