Voice photo alarm device and usage method based on 4G network and AI technology

Abstract

The present invention relates to the field of alarm technology, and in particular to a voice and photographing alarm device based on 4G network and AI technology and a use method thereof. The voice and photographing alarm device based on 4G network and AI technology comprises a communication module, an alarm module, a voice recognition module, a photographing module and a human presence detection module, wherein the communication module, the alarm module, the voice recognition module, the photographing module and the human presence detection module are installed inside a bottom shell and a main shell, and the bottom shell and the main shell are fixedly connected by bolts; the alarm module comprises a speaker and an indicator light, the voice recognition module comprises a talking microphone and an AI voice microphone, and the photographing module comprises a camera. In the present invention, during the process of alarming by pressing, a vibration effect can be formed by mutual interference between a vibration needle and a conical groove to realize vibration feedback, and at the same time, a counterweight ball can be provided to improve the vibration effect, and during the pressing process, pressing feedback can be provided to a pressing person to reduce the psychological burden of the pressing person after pressing.

Description

Technical Field

[0001] The present invention relates to the field of alarm technology, and specifically to a voice and photo alarm device based on 4G network and AI technology and a use method thereof. Background Art

[0002] The voice and photo alarm device based on 4G network and AI technology is an intelligent security device that integrates voice recognition, photo taking and alarm functions; It uses AI voice recognition technology to monitor specific keywords in the environment in real time (such as "help" and "robbery", etc.). Once triggered, the alarm program is immediately activated. At the same time, the device has a built-in high-definition camera that can take pictures and upload them to the cloud for remote monitoring and subsequent evidence collection. In addition, the device supports 4G networks and can quickly transmit data without a wired network, ensuring that the alarm information is delivered to the management platform in a timely manner. It is widely used in campuses, communities, commercial places and other scenarios to enhance security prevention capabilities; In the existing voice and photo alarm device based on 4G network and AI technology, when pressing buttons to alarm, if it is done in a hurry, it is difficult to determine whether the pressing is effective, which will bring a certain psychological burden to the person doing the pressing. Therefore, in order to solve the above problems, a voice and photo alarm device based on 4G network and AI technology and a method for using the same are proposed. Summary of the invention

[0003] The purpose of the present invention is to provide a voice and photo alarm device based on 4G network and AI technology and a method of use, so as to solve the problem that in the process of alarming by pressing a button in an anxious situation, it is difficult to determine whether the pressing is effective, which will bring a certain psychological burden to the person pressing the button.

[0004] To achieve the above object, the present invention provides the following technical solutions: A voice and photo alarm device based on 4G network and AI technology and a method of using the same, the voice and photo alarm device based on 4G network and AI technology comprises a communication module, an alarm module, a voice recognition module, a photo module and a human presence detection module, the communication module, the alarm module, the voice recognition module, the photo module and the human presence detection module are installed inside a bottom shell and a main shell, the bottom shell and the main shell are fixedly connected by bolts; the alarm module comprises a speaker and an indicator light, the voice recognition module comprises a call microphone and an AI voice microphone, the photo module comprises a camera, and a button is installed in the middle of one side of the upper end of the main shell. Key alarm, the side of the main shell is provided with an adjustment port, an external call button connecting the machine, a power supply port and a network port; the key alarm includes a pressing block, elastic connecting parts are installed around the pressing block, a reset spring for controlling the resetting of the pressing block is installed at the bottom of the pressing block, and an oscillation mechanism is installed on both sides of the lower end of the pressing block, and a blocking mechanism is provided on the side of the oscillation mechanism; the blocking mechanism includes a fixed arc plate, a conical groove is opened on the inner side of the fixed arc plate, and the oscillation mechanism includes a movable arc plate, a vibration needle is fixedly connected to the side of the movable arc plate away from the center point of the pressing block, and a counterweight ball is fixedly connected to the middle of the vibration needle.

[0005] As a further optimization of the present invention, the upper end of the fixed arc plate is fixedly connected to the main shell, and the fixed arc plate corresponds one to one with the movable arc plate, the upper end of the movable arc plate is fixedly connected to the lower pressure block, any one of the fixed arc plates is provided with a plurality of conical grooves on the inner side, and a plurality of vibration needles are fixedly connected to the outer side of the movable arc plate, and the vibration needles correspond one to one with the counterweight balls, and the end of the vibration needle away from the movable arc plate is conically set.

[0006] As further optimized content of the present invention, the communication module adopts 4G full network access technology, the alarm module also includes handle alarm, button alarm and voice alarm, and the handle alarm is connected to the alarm device through an external call button connector.

[0007] As further optimized content of the present invention, the speech recognition module supports the dialect voice wake-up function, and the speech recognition module is capable of self-learning, so as to gradually optimize the speech recognition accuracy and functional response according to the usage time and user interaction.

[0008] As a further optimized content of the present invention, the camera has a built-in photo taking function, which is used to quickly take photos of the scene when an alarm is triggered and save them automatically.

[0009] As a further optimization content of the present invention, the human presence detection module adopts an AIoT millimeter wave sensor to accurately sense and detect whether there is movement, micro-movement or human presence in the area.

[0010] As a further optimization of the present invention, the communication module supports two-way communication through voice, key or dialing, so as to facilitate the user to communicate with the police responding personnel in real time after the alarm.

[0011] As a further optimized content of the present invention, the following steps are included: Step 1: The user triggers the alarm module by connecting the external call button to the handle connected to the machine, pressing the alarm button or using voice commands; The pressing process of the button alarm is as follows: the pressing block is pressed to control the alarm. During this process, the moving arc plate drives multiple vibration needles to move accordingly. During the movement of the vibration needle, the position where the top of the vibration needle is conically set passes through the conical groove. During the passing process, the vibration needle negotiates, bends and resets. The weighted ball increases the amplitude and efficiency of the vibration, and realizes the tactile feedback during the pressing process of the pressing block 81. Step 2: After the alarm is triggered, the speech recognition module automatically recognizes the user's instructions and performs corresponding operations, and uses the optimized deep neural network loss function for training to improve recognition accuracy. The loss function is:

[0012] In the formula, is the loss function, is the output predicted by the model, is the true label, is the sample size, is the regularization coefficient, is the regularization term of the model; Step 3: Users can communicate with the police in real time through two-way voice and video calls, provide on-site information, and ensure low latency and high quality of voice calls based on the optimized 4G network scheduling algorithm. The algorithm formula for optimized scheduling is:

[0013] In the formula, is the data transmission time, is the packet size, is the current available bandwidth, for queue delay; Step 4: When an alarm is triggered, the camera in the camera module automatically takes and saves photos of the scene, and uses a network optimization algorithm to ensure the timeliness of image transmission. The network optimization algorithm formula is:

[0014] In the formula, is the signal spectrum after denoising, is the spectrum of the input signal, is the noise spectrum estimate, is the time-varying scaling factor; Step 5: The human presence detection module continuously monitors whether there is a human body in the area. The detection uses a dynamically weighted millimeter wave signal processing model to improve the detection accuracy. The processing formula is:

[0015] In the formula, For the final human body detection result, and is the weight coefficient, which is dynamically adjusted to optimize the recognition of action or static state. and Represents detected motion and presence signals, respectively.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. In the present invention, by setting the pressing alarm, in the process of pressing to alarm, the mutual interference between the vibration needle and the conical groove can form a vibration effect to achieve vibration feedback. At the same time, by setting the counterweight ball, the vibration effect can be improved, and the pressing feedback can be provided to the pressing personnel in the process of pressing, thereby reducing the psychological burden of the pressing personnel after pressing; 2. In the present invention, through a variety of alarm modes such as handles, buttons and voice commands, users can choose the most suitable alarm mode according to different emergency situations. This design greatly improves the adaptability and response speed of the system, ensures that the alarm can be effectively triggered in various environments, and enhances the practicality of the system; 3. In the present invention, the camera module can automatically take photos of the scene when an alarm is sounded, and transmit them together with the alarm information, which greatly improves the accuracy and completeness of information transmission, and provides the alarm receivers with real-time on-site conditions, helping them make more accurate emergency decisions and improving rescue efficiency; 4. In the present invention, the human presence detection module accurately senses the movement and stillness of the human body through the AIoT millimeter wave sensor, and can effectively monitor the safety status of the user's area. This module improves the safety of the device, ensures that the system can detect potential dangerous factors in a timely manner, and provide users with all-round safety protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall structure of the present invention; Figure 2 This is a schematic diagram of the button alarm structure of the present invention; Figure 3 It is a schematic diagram of the structure of the blocking mechanism of the present invention; Figure 4 For the present invention Figure 3 Schematic diagram of the structure at A in the middle; Figure 5 This is a schematic diagram of the structure of the installation position of the vibration needle of the present invention; Figure 6 For the present invention Figure 5 Schematic diagram of the structure at B in the middle; Figure 7 For the present invention Figure 5 Schematic diagram of the structure at point C in the middle.

[0018] In the figure: 1. bottom shell; 2. main shell; 3. speaker; 4. indicator light; 5. camera; 6. human presence detection module; 7. call microphone; 8. Button alarm; 81. Pressing block; 82. Elastic connecting piece; 83. Reset spring; 84, blocking mechanism; 841, fixed arc plate; 842, conical groove; 85, oscillating mechanism; 851, moving arc plate; 852, vibrating needle; 853, weighted ball; 9. Debug port; 10. External call button connector; 11. Power port; 12. Network port; 13. AI voice microphone. DETAILED DESCRIPTION

[0019] See also Figure 1-7 , the present invention provides a technical solution: The voice and photo alarm device based on 4G network and AI technology and the method of using the voice and photo alarm device based on 4G network and AI technology include a communication module, an alarm module, a voice recognition module, a photo module and a human presence detection module. The communication module, the alarm module, the voice recognition module, the photo module and the human presence detection module 6 are installed inside the bottom shell 1 and the main shell 2, and the bottom shell 1 and the main shell 2 are fixedly connected by bolts; the alarm module includes a speaker 3 and an indicator light 4, the voice recognition module includes a call microphone 7 and an AI voice microphone 13, the photo module includes a camera 5, a button alarm 8 is installed in the middle of one side of the upper end of the main shell 2, and the side of the main shell 2 is provided with an adjustment port 9, an external call button connecting machine 10, a power supply port 11 and a network port 12; the button alarm 8 includes a pressing block 81, Elastic connectors 82 are installed around the lower pressing block 81, and a reset spring 83 for controlling the reset of the lower pressing block 81 is installed at the bottom of the lower pressing block 81. Oscillating mechanisms 85 are installed on both sides of the lower end of the lower pressing block 81, and blocking mechanisms 84 are arranged on the sides of the oscillating mechanism 85; the blocking mechanism 84 includes a fixed arc plate 841, and a conical groove 842 is opened on the inner side of the fixed arc plate 841. The oscillating mechanism 85 includes a movable arc plate 851, and a vibration needle 852 is fixedly connected to the side of the movable arc plate 851 away from the center point of the lower pressing block 81, and a counterweight ball 853 is fixedly connected in the middle of the vibration needle 852. This structural design ensures the compactness and safety of each module, can effectively reduce the movement of internal components of the device, enhance the stability and seismic resistance of the overall device, and improve the long-term use reliability of the device; As a technical solution for further implementation of this solution, the upper end of the fixed arc plate 841 is fixedly connected to the main shell 2, and the fixed arc plate 841 corresponds to the movable arc plate 851 one by one, and the upper end of the movable arc plate 851 is fixedly connected to the lower pressing block 81, and there are multiple conical grooves 842 opened on the inner side of any fixed arc plate 841, and there are multiple vibration needles 852 fixedly connected to the outer side of the movable arc plate 851, and there are multiple vibration needles 852 and counterweight balls 853. The end of the vibration needle 852 away from the movable arc plate 851 is conical, which enhances the feedback effect of the button alarm module, provides more accurate and intuitive vibration feedback, and improves the user's operating experience in emergency situations. The accuracy of the vibration feedback helps the user to quickly confirm that the alarm has been triggered in a stressful situation; As a technical solution for further implementation of this solution, the communication module adopts 4G full network access technology, and the alarm module also includes handle alarm, button alarm 8 and voice alarm. The handle alarm is connected to the alarm device through an external call button connector. By supporting multiple alarm modes, users can flexibly select the alarm mode according to their needs in different environments and emergency situations, which increases the adaptability and reliability of the device and ensures that distress signals can be sent out in a timely manner in a variety of situations. As a technical solution for further implementation of this solution, the voice recognition module supports the dialect voice wake-up function, and the voice recognition module can self-learn, which is used to gradually optimize the voice recognition accuracy and function response according to the usage time and user interaction. Through self-learning and dialect support, the device can adapt to the pronunciation characteristics of different users, improve the accuracy and versatility of voice recognition, ensure that users in all regions can use the device without obstacles, and improve the user experience; As a technical solution for further implementation of this scheme, the camera 5 has a built-in photo function, which is used to quickly take photos of the scene when an alarm is triggered and automatically save them. This function automatically saves photos of the scene when an alarm is triggered, providing strong evidence for subsequent investigation and processing, and improving the practicality and processing efficiency of the device in emergency situations; As a technical solution for further implementation of this solution, the human presence detection module 6 adopts AIoT millimeter wave sensors to accurately sense and detect whether there is movement, micro-movement or human presence in the area. The use of advanced AIoT millimeter wave sensors can accurately detect the presence of human bodies in complex environments, greatly improving the security function, reducing false alarms, and ensuring the efficiency and accuracy of the alarm system; As a technical solution for further implementation of this solution, the communication module supports two-way communication through voice, key or dialing, which is used for users to communicate with the police in real time after the alarm. Multiple communication methods ensure that users can communicate with the police in time in different situations, especially in emergency situations with inconvenient hands, and provide on-site information, which enhances the flexibility and real-time performance of the alarm system. As a technical solution for further implementing this plan, the following steps are included: Step 1: The user triggers the alarm module by connecting the external call button to the handle connected to the machine 10, pressing the alarm button 8 or voice command; In the process of alarming by pressing the button alarm 8, the pressing block 81 is pressed to control the alarm. In this process, the moving arc plate 851 drives the multiple vibration pins 852 to move accordingly. In the process of moving the vibration pins 852, the position of the cone-shaped setting at the top of the vibration pins 852 passes through the cone groove 842. In the process of passing, the vibration pins 852 negotiate, bend and reset, thereby achieving a vibration effect. The weighted ball 853 can increase the amplitude and efficiency of the vibration, and realize the tactile feedback during the pressing process of the pressing block 81. The pressing information can be intuitively fed back to the pressing personnel, which can avoid the problem that the pressing personnel are uncertain about the pressing information in a panic, and can improve the safety feeling of the pressing personnel. Step 2: After the alarm is triggered, the speech recognition module automatically recognizes the user's instructions and performs corresponding operations, and uses the optimized deep neural network loss function for training to improve recognition accuracy. The loss function is:

[0020] In the formula, is the loss function, is the output predicted by the model, is the true label, is the sample size, is the regularization coefficient, is the regularization term of the model; Step 3: Users can communicate with the police in real time through two-way voice and video calls, provide on-site information, and ensure low latency and high quality of voice calls based on the optimized 4G network scheduling algorithm. The algorithm formula for optimized scheduling is:

[0021] In the formula, is the data transmission time, is the packet size, is the current available bandwidth, for queue delay; Step 4: When an alarm is triggered, the camera 5 in the camera module automatically takes and saves photos of the scene, and ensures the timeliness of image transmission through a network optimization algorithm. The network optimization algorithm formula is:

[0022] In the formula, is the signal spectrum after denoising, is the spectrum of the input signal, is the noise spectrum estimate, is the time-varying scaling factor; Step 5: The human presence detection module 6 continuously monitors whether there is a human body in the area. The detection adopts a dynamically weighted millimeter wave signal processing model to improve the detection accuracy. The processing formula is:

[0023] In the formula, For the final human body detection result, and is the weight coefficient, which is dynamically adjusted to optimize the recognition of action or static state. and They represent detected motion and presence signals respectively. Through the combination of optimized algorithms and multiple functions, the efficiency and accuracy of the alarm device are ensured. Users can flexibly choose the alarm method, the voice recognition and camera modules can be executed in time, the call quality is guaranteed, and the human body detection module can accurately perceive safety hazards and provide a comprehensive emergency response plan, which enhances the emergency response capability and practicality of the entire system.

[0024] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the method and core ideas of the present invention. The above is only a preferred implementation of the present invention. It should be pointed out that due to the limitations of textual expression and the objective existence of infinite specific structures, ordinary technicians in this technical field can make several improvements, modifications or changes without departing from the principles of the present invention, and can also combine the above technical features in an appropriate manner; these improvements, modifications, changes or combinations, or the direct application of the inventive concept and technical solution to other occasions without improvement, should be regarded as the protection scope of the present invention.

Claims

1. A voice and photo alarm device based on 4G network and AI technology, comprising a communication module, an alarm module, a voice recognition module, a photo module and a human presence detection module, characterized in that: The communication module, alarm module, voice recognition module, camera module and human presence detection module (6) are installed inside the bottom shell (1) and the main shell (2), and the bottom shell (1) and the main shell (2) are fixedly connected by bolts; The alarm module comprises a speaker (3) and an indicator light (4); the voice recognition module comprises a call microphone (7) and an AI voice microphone (13); the camera module comprises a camera (5); a button alarm (8) is installed in the middle of one side of the upper end of the main shell (2); and the side of the main shell (2) is provided with an adjustment port (9), an external call button connecting machine (10), a power supply port (11) and a network port (12); The button alarm (8) comprises a lower pressing block (81), elastic connecting members (82) are installed around the lower pressing block (81), a return spring (83) for controlling the return of the lower pressing block (81) is installed at the bottom end of the lower pressing block (81), and oscillation mechanisms (85) are installed on both sides of the lower end of the lower pressing block (81), and a blocking mechanism (84) is provided on the side of the oscillation mechanism (85); The blocking mechanism (84) comprises a fixed arc plate (841), the inner side of which is provided with a conical groove (842); the oscillating mechanism (85) comprises a movable arc plate (851), a vibrating needle (852) is fixedly connected to a side of the movable arc plate (851) away from the center point of the lower pressing block (81), and a counterweight ball (853) is fixedly connected to the middle of the vibrating needle (852).

2. The voice photo alarm device based on 4G network and AI technology according to claim 1 is characterized by: The upper end of the fixed arc plate (841) is fixedly connected to the main shell (2), and the fixed arc plate (841) corresponds to the movable arc plate (851) in one-to-one relationship. The upper end of the movable arc plate (851) is fixedly connected to the lower pressing block (81). A plurality of conical grooves (842) are provided on the inner side of any of the fixed arc plates (841). A plurality of vibration needles (852) are fixedly connected to the outer side of the movable arc plate (851), and the vibration needles (852) correspond to the counterweight balls (853) in one-to-one relationship. The end of the vibration needle (852) away from the movable arc plate (851) is conical.

3. The voice photo alarm device based on 4G network and AI technology according to claim 1 is characterized in that: The communication module adopts 4G full network access technology, and the alarm module also includes a handle alarm, a button alarm (8) and a voice alarm. The handle alarm is connected to the alarm device via an external call button connector.

4. The voice photo alarm device based on 4G network and AI technology according to claim 1 is characterized in that: The speech recognition module supports a dialect voice wake-up function, and the speech recognition module is capable of self-learning, and is used to gradually optimize speech recognition accuracy and functional response according to usage time and user interaction.

5. The voice photo alarm device based on 4G network and AI technology according to claim 1 is characterized in that: The camera (5) has a built-in photo taking function, which is used to quickly take photos of the scene when an alarm is triggered, and automatically save the photos.

6. The voice photo alarm device based on 4G network and AI technology according to claim 1, characterized in that: The human presence detection module (6) uses an AIoT millimeter wave sensor to accurately sense and detect whether there is movement, micro-movement or human presence in the area.

7. The voice photo alarm device based on 4G network and AI technology according to claim 1 is characterized in that: The communication module supports two-way communication through voice, key or dialing, which is used for users to communicate with the police in real time after the alarm.

8. A method for using the voice and photo alarm device based on 4G network and AI technology according to any one of claims 1 to 7, characterized in that: The following steps are involved: Step 1: The user triggers the alarm module by using a handle connected to the machine (10), a key alarm (8) or a voice command via an external call button; The pressing process of the button alarm (8) is as follows: the pressing block (81) is pressed to control the alarm. During this process, the moving arc plate (851) drives the plurality of vibration needles (852) to move accordingly. During the movement of the vibration needles (852), the positions at which the cones at the tops of the vibration needles (852) are set pass through the cone grooves (842). During the passing process, the vibration needles (852) negotiate, bend and reset. The weighted balls (853) increase the amplitude and efficiency of the vibration, thereby realizing tactile feedback during the pressing process of the pressing block 81. Step 2: After the alarm is triggered, the speech recognition module automatically recognizes the user's instructions and performs corresponding operations, and uses the optimized deep neural network loss function for training to improve recognition accuracy. The loss function is: In the formula, is the loss function, is the output predicted by the model, is the true label, is the sample size, is the regularization coefficient, is the regularization term of the model; Step 3: Users can communicate with the police in real time through two-way voice and video calls, provide on-site information, and ensure low latency and high quality of voice calls based on the optimized 4G network scheduling algorithm. The algorithm formula for optimized scheduling is: In the formula, is the data transmission time, is the packet size, is the current available bandwidth, for queue delay; Step 4: When an alarm is triggered, the camera (5) in the camera module automatically takes and saves photos of the scene, and uses a network optimization algorithm to ensure the timeliness of image transmission. The network optimization algorithm formula is: In the formula, is the signal spectrum after denoising, is the spectrum of the input signal, is the noise spectrum estimate, is the time-varying scaling factor; Step 5: The human presence detection module (6) continuously monitors whether there is a human body in the area. The detection uses a dynamically weighted millimeter wave signal processing model to improve the detection accuracy. The processing formula is: In the formula, For the final human body detection result, and is the weight coefficient, which is dynamically adjusted to optimize the recognition of action or static state. and Represents detected motion and presence signals, respectively.

Images