A smart identification system based on neural network model and its use method
The smart management system with neural network-based face recognition and integrated data management mechanisms addresses outdoor recognition challenges, ensuring accurate and efficient face recognition and data exchange across regions, enhancing system stability and security.
Patent Information
- Application Number
- CN202210432478.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-24
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2042-04-24
AI Technical Summary
Existing face recognition systems in service terminals lack accuracy, stability, and efficiency due to outdoor recognition challenges, single-terminal limitations, and inadequate data connectivity between different areas, leading to inefficiencies and security vulnerabilities.
A smart management system utilizing a neural network model with a protection box equipped with adjustable shading, heating, and demisting mechanisms, combined with a networked smart module for face recognition and data management, ensuring accurate and secure face recognition and efficient data exchange across regions.
Enhances face recognition accuracy and adaptability, improves security, and optimizes data acquisition and processing efficiency by adapting to environmental conditions and facilitating cross-regional data collaboration.
Smart Images

Figure CN114863610B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of service technology, and more specifically, to an intelligent recognition system based on a neural network model and a method for using the system. Background Art
[0002] How to quickly provide high-quality services to the public through smart systems at service terminals has become an important direction for the development of informatization.
[0003] Chinese patent document CN107067567A discloses a human resources and social security self-service system, which includes a self-service terminal, an employment service system, an on-site recruitment hall, a recruitment website, an employment card, a social security system, a call center, and a text message platform. The invention realizes the collection of personal information through a self-service terminal and feeds back the information to the employment service system and the social security service system, but in actual use, there is a lack of good registration and tracing means. When the person to be processed enters the processing window, the identity verification and identification of the person to be processed still need to be performed, and the existing face recognition system does not have the accuracy and stability required for outdoor recognition and screening, and the number is obtained through a single terminal, which is easy to produce a production line where the number is passed or grabbed, affecting the processing efficiency of the window. At the same time, the data is only linked within a single area, and the information connection between different areas is relatively inconvenient, and a long waiting time for data communication is required, which cannot meet the efficiency of the service well. Summary of the invention
[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide an intelligent recognition system based on a neural network model and a method of using the system, which has the characteristics of adjusting the light irradiation angle and providing recognition accuracy to ensure durable life protection.
[0005] To achieve the above object, the present invention provides the following technical solutions:
[0006] A smart management system based on a neural network model comprises a protection box arranged at the entrance of a service terminal, wherein a service system host is arranged in the inner cavity of the protection box, a mainboard is loaded in the service system host, and single-chip microcomputers and control chips of each module of the management system body are connected to the service system host, and the management system body is connected to the server through a network communication, a supporting mechanism is arranged at the bottom of the protection box, a sealing mechanism is embedded in one side of the protection box, a heating and demisting identification mechanism is arranged in the inner cavity of the sealing mechanism, an adjustable shading mechanism is arranged on the outer wall of the identification mechanism, and a rain shelter mechanism is movably connected to the top of the protection box.
[0007] By adopting the above technical solution, the face recognition accuracy, adaptability and security of the hall can be optimized and adapted, and the rapid acquisition of data can be achieved through the recognition of face information.
[0008] Furthermore, the support mechanism includes a support plate, insertion holes are opened at the four corners of the top of the support plate, a fixing groove is embedded in the top of the support plate, a support column is clamped in the fixing groove, and the top of the support column is fixedly connected to the bottom of the protection box.
[0009] By adopting the above technical solution, the safety of the protection box installation is improved, and displacement caused by deviation, shaking and impact is avoided.
[0010] Furthermore, the sealing mechanism includes a sealing ring gasket, which is embedded in one side of the protection box, and a plurality of sealing spacers are fixedly connected to the inner cavity of the sealing ring gasket, and the sealing spacer is sleeved on the outside of the identification mechanism, and the sealing spacer and the sealing ring gasket are silicone pads.
[0011] By adopting the above technical solution, the protection capability of the set identification mechanism is improved.
[0012] Furthermore, the identification mechanism includes a transmission plate, one side of the transmission plate is hinged to one side of the inner cavity of the protection box through a first pin shaft, a lens shell is embedded in the inner cavity of the transmission plate, a periscope lens and a photosensitive lens are fixedly installed in the inner cavity of the lens shell, a driving mechanism is fixedly installed at the bottom of the transmission plate, a filling sleeve is provided on the outside of the lens shell, a heating seat is provided in the inner cavity of the filling sleeve, a lens piece is embedded in the inner cavity of the heating seat, and the lens piece is fit with one side of the lens shell.
[0013] By adopting the above technical solution, the influence of fog accumulation on the accuracy of face recognition can be avoided, and the recognition efficiency can be improved through the coordinated recognition of two sets of lenses.
[0014] Furthermore, the driving mechanism includes a second electric push rod, which is fixedly mounted at the bottom of the inner cavity of the protection box, and a mounting plate is fixedly connected to the top of the second electric push rod. Both the top of the mounting plate and the bottom of the transmission plate are fixedly connected to hinge blocks, and a connecting rod is hinged between the hinge blocks on both sides through a second pin shaft.
[0015] By adopting the above technical solution, the drive adjustment of the identification mechanism can be achieved to avoid aging and wear caused by common sense exposure.
[0016] Furthermore, the shading mechanism includes a shading hood, an elastic pad is integrally connected between the two sides of the shading hood, fixed blocks are fixedly connected on both sides of the shading hood, the inner cavity of the fixed block is threadedly connected to a first electric push rod through a nut, and the first electric push rod is fixedly connected to one side of the outside of the protective box.
[0017] By adopting the above technical solution, the shielding of sunlight can be adjusted to meet the stability of face image collection in different weather conditions.
[0018] Furthermore, the rain shelter mechanism includes a slide rail seat, which is fixedly connected to the top of the protection box. A slide groove is provided on one side of the slide rail seat, and a slider is slidably connected in the slide groove. A baffle is fixedly connected to one side of the slider, and a plurality of through grooves are provided on the top of the baffle.
[0019] A smart management system based on a neural network model comprises a management system body, and the management system body comprises a smart module arranged on a network server, and the network module is used for performing overall management services on information through network communication services, the input end of the smart module is communicatively connected with a cloud platform storage module and a user terminal unit, and is used for cloud storage of platform data and control call of user terminals, the input end of the smart module is connected with a service terminal, and the service terminal comprises a face recognition module and a micro-motion alarm module, and is used for identifying face information of the service terminal and performing joint defense alarm for the safety of the terminal service hall, the input end of the face recognition module is connected with a daylight detection module and a fill light control module, and the face recognition module is used for improving the accuracy and efficiency of face recognition through a neural network algorithm, the input end of the micro-motion alarm module is connected with a security detection module, and is used for linkage alarm, the input end of the smart module is bidirectionally connected with a first information management unit, a second information management unit and a regional tracing comparison module, and is used for performing overall tracing comparison management on information, the input end of the smart module is connected with a business collaboration module, and the input end of the business collaboration module is respectively electrically connected with a data receiving module and a department verification module, and is used for the communication of collaborative office data.
[0020] By adopting the above technical solutions, the joint defense alarm and service interaction capabilities of the intelligent system are realized, and the information interaction capabilities between multiple regions are guaranteed, reducing the process volume required for the service.
[0021] Furthermore, the user terminal unit includes an appointment queuing module, a Beidou positioning module and a customer service module. The appointment queuing module is used for remote appointment terminal services, the Beidou positioning module is used to achieve regional positioning through connectivity with a mobile phone positioning module, and the customer service module is used to provide feedback on information consultation through messages, phone calls, videos, and APP notifications.
[0022] By adopting the above technical solutions, users' personal appointment, consultation and information acquisition capabilities on smart terminals are improved.
[0023] The present invention provides a method for using an intelligent management system based on a neural network model. When entering a recognition area and face recognition begins, the second electric push rod at the bottom works to extend, and the mounting plate moves upward. The movement of the mounting plate pushes the hinge block to rotate around the second pin shaft and pushes the connecting rod upward to drive the hinge block and the transmission plate on the other side to move. The transmission plate moves to rotate around the first pin shaft and drives the lens housing on one side to enter the sealing ring gasket and the sealing spacer. The first electric push rod works to shorten and pull the fixed block to move. The movement of the fixed block can pull the sunshade and the elastic pad to move backward. The sunshade pulls the elastic pad to unfold by moving, and the unfolded elastic pad can adjust the light transmittance through the transparent material.
[0024] In summary, the present invention has the following beneficial effects:
[0025] 1. By setting up a smart module and connecting the face recognition module through the smart module, it is possible to perform face recognition at the service terminal, identify face information through a neural network model, and control the first electric push rod to adjust the angle of the sunshade through the fill light control module to achieve the accuracy of face recognition. After face recognition and positioning, the first information management unit, the second information management unit and the regional traceability unit are pre-called through the appointment and numbering module of the user terminal unit to improve the efficiency of business handling. The data linkage between departments in different regions is realized through the business collaboration module, and the data accuracy is included through the department extension verification module, and then the efficiency of cross-regional business handling can be achieved through the collaborative data call of multiple departments.
[0026] 2. In this scheme, a driving mechanism is provided to pull the transmission plate to realize the lifting and hiding of the lens shell, and the lens piece can be heated by heating the heating seat on one side. The lens piece can heat and defog the lens shell and the internal photosensitive lens and the front lens, effectively avoiding the influence of lens fogging on the recognition accuracy, and the hidden lens shell can prevent the lens shell from being exposed outside the protective box for a long time to cause dust accumulation, effectively avoiding the dust from affecting the recognition accuracy and efficiency, effectively meeting the safety and stability requirements of the identification mechanism placed outdoors, and realizing the identity identification and positioning of the business personnel by setting up a face recognition system in the service hall, effectively reducing the identity identification time of the service window, and improving the efficiency of business handling;
[0027] 3. By setting up a shading mechanism, when the external light changes, the first electric push rod can be controlled to shorten and pull the fixed block to move. The movement of the fixed block can pull the shading hood and the elastic pad to move backward. When the shading hood moves, the relative distance between the shading hood can be expanded by pulling the fixed blocks on both sides, thereby improving the adjustment of the light entering the internal lens shell, and thus effectively avoiding the lack of face recognition accuracy caused by overexposure or low brightness. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 is a system block diagram of the present invention;
[0029] Figure 2 is a system block diagram of a user terminal unit of the present invention;
[0030] Figure 3 Schematic diagram of the three-dimensional split structure of the present invention
[0031] Figure 4 Schematic diagram of the three-dimensional structure of the present invention
[0032] Figure 5 It is a schematic diagram of the three-dimensional structure of the identification mechanism of the present invention;
[0033] Figure 6 It is a schematic diagram of the three-dimensional structure of the driving mechanism of the present invention.
[0034] In the figure:
[0035] 1. Protective box; 2. Sealing mechanism; 201. Sealing ring; 202. Sealing spacer; 3. Support mechanism; 301. Support plate; 302. Insertion hole; 303. Fixing groove; 4. Rain shelter mechanism; 401. Slide rail seat; 402. Sliding block; 403. Baffle; 404. Through groove; 5. Identification mechanism; 501. Transmission plate; 502. Lens shell; 503. Periscope lens; 504. Photosensitive lens; 505. Lens piece; 506. Heating seat; 507. Filling sleeve; 6. Shading mechanism; 601. First electric push rod; 602. Fixing block; 603. Nut cap; 604. Shading hood; 605. Elastic pad; 7. Driving mechanism; 701. Second electric push rod; 70 2. Mounting plate; 703. Articulated block; 704. Second pin shaft; 705. Connecting rod; 8. Management system body; 801. Smart module; 802. First information management unit; 803. Second information management unit; 804. Regional tracing and comparison module; 805. Cloud platform storage module; 806. Business collaboration module; 807. Department verification module; 808. Data receiving module; 809. User terminal unit; 8091. Appointment number module; 8092. Beidou positioning module; 8093. Customer service module; 810. Daylight detection module; 811. Fill light control module; 812. Security detection module; 813. Face recognition module; 814. Micro-motion alarm module. DETAILED DESCRIPTION
[0036] Example:
[0037] The following is combined with Figure 1-6 The present invention is described in further detail.
[0038] See also Figure 1-2The present invention provides a technical solution: a smart management system based on a neural network model, including a management system body 8 arranged in a system host inside a smart management device, the management system body 8 realizes control of service management through a system mainboard and various mechanisms loaded in the system host, and the management system body 8 includes a smart module 801 arranged in a network server, a network module for coordinating information management services through network communication services, the input end of the smart module 801 is communicatively connected with a cloud platform storage module 805 and a user terminal unit 809, which are used for cloud storage of platform data and control calls to user terminals, the input end of the smart module 801 is connected with a service terminal, and the service terminal includes a face recognition module 813 and a micro-motion alarm module 814, wherein the face recognition module 813 is a HI3516EV300 face recognition module single board computer, and the micro-motion alarm module 814 is a GST-L The D-8300 alarm module is used to identify the facial information of the service terminal and to conduct joint defense alarm for the security of the terminal service hall. The input end of the face recognition module 813 is connected to the daylight detection module 810, namely the ITR-9909 light sensor and the fill light control module 811, and the face recognition module 813 is used to improve the accuracy and efficiency of face recognition through a neural network algorithm. The input end of the micro-motion alarm module 814 is connected to the security detection module 812 for linkage alarm. The input end of the smart module 801 is bidirectionally connected to the first information management unit 802, the second information management unit 803 and the regional tracing and comparison module 804 for coordinated tracing and comparison management of information. The input end of the smart module 801 is connected to the business collaboration module 806. The input ends of the business collaboration module 806 are respectively electrically connected to the data receiving module 808, namely DTD433M and the department verification module 807 for collaborative office data communication.
[0039] In this embodiment: the cloud platform storage module 805, i.e., the EEPROM storage module, can store the platform data in the cloud server to ensure data security and storage cost. The micro-motion alarm module 814 of the service terminal can realize security detection and linkage alarm of the in-and-out personnel through the front-end security detection module 812, realize the business handling security capability of the service terminal, and realize the rapid identification of face data through the learning function of the neural network model, and register and reserve information through the smart module 801, so as to facilitate the rapid positioning of user information at the service window of the service terminal without re-proofreading, and realize the management input and output of information through the first information management unit 802, the second information management unit 803 and the regional tracing comparison module 804, and realize the information call transcription between different regions through the regional tracing comparison module 804, thereby effectively improving the information processing service efficiency. The smart module 801 is an Internet network module, which is used to realize the data exchange and communication service of the network module of the terminal, and communicates with the terminal system set in the smart device through the network.
[0040] like Figure 6 As shown, the user terminal unit 809 includes an appointment queuing module 8091, a Beidou positioning module 8092 and a customer service module 8093. The appointment queuing module 8091 is used for remote appointment terminal services, the Beidou positioning module 8092 and the UM4B0 module are used to achieve regional positioning through connectivity with the mobile phone positioning module, and the customer service module 8093 is used to provide feedback on information consultation through messages, phone calls, videos, and APP notifications.
[0041] In this embodiment: the user terminal unit 809 is used to realize the rapid reservation processing of the service number, and the Beidou positioning module 8092 can realize the rapid division of the area and the rapid recommendation of the business department.
[0042] like Figure 3-6 As shown, a smart management system based on a neural network model includes a protection box 1 arranged at the entrance of a service terminal, and a service system host is arranged in the inner cavity of the protection box 1. The service system host is loaded with a motherboard, the motherboard model is Intel5000V, and the chip model is Intel Xeon 5405, the mainboard integrates the Windows software system and communicates with the smart module 801. The smart module 801 uses the server for data processing. The processor of the smart module 801 cooperates with the Windows software system through the user terminal unit 809, the face recognition module 813 and the micro-motion alarm module 814 in the load mainboard to realize the software control of the support mechanism 3, the sealing mechanism 2 and the identification mechanism 5 in the management device and the support of face recognition. The single-chip microcomputers and control chips of each module of the management system main body 8 are all arranged on the system mainboard, and the management system main body 8 is connected to the smart module 801 server through the network chip integrated with the system mainboard. Specifically, the bottom of the protection box 1 is provided with a support mechanism 3, and a sealing mechanism 2 is embedded in one side of the protection box 1. The inner cavity of the sealing mechanism 2 is provided with a heating and demisting identification mechanism 5. The identification mechanism 5 realizes software recognition control by communicating with the face recognition module 813 in the service system. The outer wall of the identification mechanism 5 is provided with an adjustable shading mechanism 6. The top of the protection box 1 is movably connected with a rain shelter mechanism 4, and the rain shelter mechanism 4 is controlled by the single-chip microcomputer loaded by the micro-motion alarm module 814.
[0043] In this embodiment: the sealing mechanism 2 can achieve sound insulation and noise reduction for the protection box 1, and can quickly clear the fog through self-heating, and ensure that the shading mechanism 6 adjusts the light input of the identification mechanism 5.
[0044] like Figure 3As shown, the support mechanism 3 includes a support plate 301, and insertion holes 302 are opened at the four corners of the top of the support plate 301. A fixing groove 303 is embedded in the top of the support plate 301. A support column is clamped in the fixing groove 303, and the top of the support column is fixedly connected to the bottom of the protection box 1.
[0045] In this embodiment: the fixing groove 303 can ensure the stability of the support column insertion, and the multiple insertion holes 302 can be connected to the embedded column.
[0046] like Figure 3 As shown, the sealing mechanism 2 includes a sealing ring gasket 201, which is embedded in one side of the protection box 1. The inner cavity of the sealing ring gasket 201 is fixedly connected with a plurality of sealing spacers 202. The sealing spacers 202 are sleeved on the outside of the identification mechanism 5. The sealing spacers 202 and the sealing ring gasket 201 are silicone pads.
[0047] In this embodiment: the sealing partition and the sealing ring gasket 201 made of silicone pad material can ensure the sleeve limit of the identification mechanism 5, and the ring gasket is arranged in a ring shape to ensure the passability of the identification mechanism 5.
[0048] like Figure 3 and 5 As shown, the identification mechanism 5 includes a transmission plate 501, one side of the transmission plate 501 is hinged to one side of the inner cavity of the protection box 1 through a first pin shaft, a lens shell 502 is embedded in the inner cavity of the transmission plate 501, a periscope lens 503 and a photosensitive lens 504 are fixedly installed in the inner cavity of the lens shell 502, a driving mechanism 7 is fixedly installed at the bottom of the transmission plate 501, a filling sleeve 507 is provided on the outer side of the lens shell 502, a heating seat 506 is provided in the inner cavity of the filling sleeve 507, a lens piece 505 is embedded in the inner cavity of the heating seat 506, and the lens piece 505 is in contact with one side of the lens shell 502.
[0049] In this embodiment: the lens piece 505 is snapped into the heating seat 506 to achieve heat conduction, and the filling sleeve 507 can ensure the snap-fit filling effect with the external shading mechanism 6, while the lens shell 502 can protect the periscope lens 503 and the photosensitive lens 504, and the lens piece 505 can improve the isolation and blocking of external dust to ensure the overall durable life.
[0050] like Figure 6 As shown, the driving mechanism 7 includes a second electric push rod 701, which is fixedly mounted at the bottom of the inner cavity of the protective box 1. The top of the second electric push rod 701 is fixedly connected to a mounting plate 702. The top of the mounting plate 702 and the bottom of the transmission plate 501 are both fixedly connected to hinge blocks 703, and a connecting rod 705 is hinged between the hinge blocks 703 on both sides through a second pin shaft 704.
[0051] In this embodiment, the L-shaped transmission plate 501 can ensure that the bottom driving mechanism 7 is turned over when pulled, and then the stability of the rotational deviation of the transmission plate 501 in the protection box 1 can be ensured.
[0052] like Figure 3-4 As shown, the shading mechanism 6 includes a shading cover 604, and elastic pads 605 are integrally connected between the two sides of the shading cover 604. Fixed blocks 602 are fixedly connected on both sides of the shading cover 604. The inner cavity of the fixed block 602 is threadedly connected to the first electric push rod 601 through a nut cap 603. The first electric push rod 601 is fixedly connected to one side of the outside of the protective box 1.
[0053] In the present embodiment: the elastic pad 605 integrally connected to the inner cavity of the light shield 604 is a transparent pad body, and when the two sides of the light shield 604 are pulled apart, the light shield 604 can realize the expansion of the elastic pad 605, and then the expansion range of the elastic pad 605 can be adjusted by pulling the first electric push rod 601, thereby improving the precision of adjusting the input light, and the filling of the light shield 604 and the filling sleeve 507 can avoid axial displacement of the lens housing 502, thereby improving the assembly precision.
[0054] like Figure 3-4 As shown, the rain shelter mechanism 4 includes a slide rail seat 401, which is fixedly connected to the top of the protection box 1, a slide groove is provided on one side of the slide rail seat 401, and a slider 402 is slidably connected in the slide groove, a baffle 403 is fixedly connected to one side of the slider 402, and a plurality of through grooves 404 are provided on the top of the baffle 403;
[0055] In this embodiment: the baffle 403 on the top of the protective box 1 can be quickly slid into and connected with the slide groove on one side of the slide rail seat 401 through the slider 402 on the rear side, which is convenient for cleaning the baffle 403. The baffle 403 can reduce the impact of external falling objects and achieve rain and moisture-proof effects. The T-shaped slider 402 slides more stably in the slide rail seat 401, effectively avoiding axial pulling that causes the slider 402 to move out of the slide rail seat 401.
[0056] The present invention provides a method for using an intelligent management system based on a neural network model. When entering the recognition area and face recognition begins, the second electric push rod 701 at the bottom works to extend, and the mounting plate 702 moves upward. The mounting plate 702 moves to push the hinge block 703 to rotate around the second pin shaft 704 and push the connecting rod 705 to drive the hinge block 703 on the other side and the transmission plate 501 to move upward. The transmission plate 501 moves to rotate around the first pin shaft and drives one side of the lens housing 502 to enter the sealing ring gasket 201 and the sealing spacer 202. The first electric push rod 601 works to shorten and pull the fixed block 602 to move. The fixed block 602 moves to pull the light shield 604 and the elastic pad 605 to move backward. The light shield 604 moves to pull the elastic pad 605 to unfold, and the unfolded elastic pad 605 realizes light transmittance adjustment through the transparent material.
[0057] When in use, the protection box 1 is inserted into the support plate 301 through the bottom support column, and the support plate 301 is installed through the insertion holes 302 at the four corners of the top.
[0058] This specific embodiment is merely an explanation of the present invention and is not a limitation of the present invention. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed. However, as long as they are within the scope of the claims of the present invention, they are protected by the patent law.
Claims
1. An intelligent management system based on a neural network model, characterized by: The invention comprises a protection box (1) arranged at the entrance of a service terminal, wherein the inner cavity of the protection box (1) is provided with a service system host, the service system host is loaded with a main board, and each module single chip microcomputer and control chip of a management system main body (8) are connected to the service system host, and the management system main body (8) is connected to a server through a network, the bottom of the protection box (1) is provided with a support mechanism (3), one side of the protection box (1) is embedded with a sealing mechanism (2), the inner cavity of the sealing mechanism (2) is provided with a heating and demisting identification mechanism (5), the outer wall of the identification mechanism (5) is provided with an adjustable shading mechanism (6), and the top of the protection box (1) is movably connected with a rain shelter mechanism (4); the support mechanism (3) comprises a support plate (301), the support plate Insertion holes (302) are provided at the four corners of the top of the support plate (301); a fixing groove (303) is embedded in the top of the support plate (301); a support column is clamped in the fixing groove (303), and the top of the support column is fixedly connected to the bottom of the protection box (1); the sealing mechanism (2) comprises a sealing ring gasket (201); the sealing ring gasket (201) is embedded in one side of the protection box (1); a plurality of sealing spacers (202) are fixedly connected to the inner cavity of the sealing ring gasket (201); the sealing spacers (202) are sleeved on the outside of the identification mechanism (5); the sealing spacers (202) and the sealing ring gasket (201) are silicone pads; the identification mechanism (5) comprises a transmission plate (501); one side of the transmission plate (501) is connected by a first pin shaft The transmission plate (501) is hinged to one side of the inner cavity of the protection box (1); a lens shell (502) is embedded in the inner cavity of the transmission plate (501); a periscope lens (503) and a photosensitive lens (504) are fixedly installed in the inner cavity of the lens shell (502); a driving mechanism (7) is fixedly installed at the bottom of the transmission plate (501); a filling sleeve (507) is provided on the outside of the lens shell (502); a heating seat (506) is provided in the inner cavity of the filling sleeve (507); a lens piece (505) is embedded in the inner cavity of the heating seat (506); and the lens piece (505) is fitted with one side of the lens shell (502); the driving mechanism (7) comprises a second electric push rod (701); the second electric push rod (701) is fixedly installed at the bottom of the inner cavity of the protection box (1). The top of the second electric push rod (701) is fixedly connected to a mounting plate (702), the top of the mounting plate (702) and the bottom of the transmission plate (501) are fixedly connected to hinge blocks (703), and a connecting rod (705) is hinged between the hinge blocks (703) on both sides via a second pin shaft (704); the shading mechanism (6) comprises a light shield (604), elastic pads (605) are integrally connected between the two sides of the light shield (604), and the two sides of the light shield (604) are fixedly connected to fixed blocks (602), the inner cavity of the fixed block (602) is threadedly connected to the first electric push rod (601) via a nut cap (603), and the first electric push rod (601) is fixedly connected to one side of the outside of the protection box (1);The rain shelter mechanism (4) comprises a slide rail seat (401), the slide rail seat (401) is fixedly connected to the top of the protection box (1), a slide groove is provided on one side of the slide rail seat (401), a slider (402) is slidably connected in the slide groove, a baffle (403) is fixedly connected to one side of the slider (402), and a plurality of through grooves (404) are provided on the top of the baffle (403). ; 2. According to claim 1, a smart management system based on a neural network model is characterized in that: The invention comprises a management system body (8), wherein the management system body (8) comprises an intelligent module (801) arranged on a network server, which is a network module for providing a coordinated management service for information through a network communication service. The input end of the intelligent module (801) is communicatively connected to a cloud platform storage module (805) and a user terminal unit (809), which is used for cloud storage of platform data and control calls to user terminals. The input end of the intelligent module (801) is connected to a service terminal, and the service terminal comprises a face recognition module (813) and a micro-motion alarm module (814), which are used for identifying face information of the service terminal and providing a joint defense alarm for the safety of the terminal service hall. The face recognition module (81 3) The input end is connected to a daylight detection module (810) and a fill light control module (811), and the face recognition module (813) is used to improve the accuracy and efficiency of face recognition through a neural network algorithm. The input end of the micro-motion alarm module (814) is connected to a security detection module (812) for linkage alarm. The input end of the smart module (801) is bidirectionally connected to a first information management unit (802), a second information management unit (803) and a regional tracing comparison module (804) for module (806). The input end of the business collaboration module (806) is respectively electrically connected to a data receiving module (808) and a department verification module (807) for collaborative office data communication.
3. The intelligent management system based on a neural network model according to claim 2 is characterized in that: The user terminal unit (809) includes an appointment number queuing module (8091), a Beidou positioning module (8092) and a customer service module (8093), wherein the appointment number queuing module (8091) is used for remote appointment terminal service, the Beidou positioning module (8092) is used for realizing regional positioning through the connection of the mobile phone positioning module, and the customer service module (8093) is used for feedback on information consultation through information, telephone, video and APP notification.
4. The method for using a smart management system based on a neural network model according to claim 3 is characterized in that: When entering the recognition area and face recognition begins, the second electric push rod (701) at the bottom works to extend, the mounting plate (702) moves upward, the mounting plate (702) moves to push the hinge block (703) to rotate around the second pin shaft (704) and push the connecting rod (705) to drive the hinge block (703) on the other side and the transmission plate (501) to move upward, the transmission plate (501) moves to rotate around the first pin shaft and drives the lens housing (502) on one side to enter the sealing ring gasket (201) and the sealing spacer (202), the first electric push rod (601) works to shorten and pull the fixed block (602) to move, the fixed block (602) moves to pull the light shielding hood (604) and the elastic pad (605) to move backward, the light shielding hood (604) moves to pull the elastic pad (605) to expand, and the elastic pad (605) after expansion realizes light transmittance adjustment through transparent material.
