Community nb device information collection and access system and method based on an ocr algorithm model

Abstract

The present application relates to a kind of community NB equipment information acquisition and access system and method based on OCR algorithm model.The system includes: the APP terminal equipment based on OCR image recognition, by detecting-identification algorithm model, to be accessed NB-IoT equipment is positioned, output key information, and encapsulation is reported to the unified access platform of wisdom community internet of things equipment;The unified access platform of wisdom community internet of things equipment is used for the data matching and data correlation perfect of the key information of reported NB-IoT equipment;Message bus is used for the uplink and downlink service message data between the front-end access equipment gateway and the unified access platform of wisdom community internet of things equipment, and message data is stored and forwarded;Front-end access equipment gateway is registered and business communication by telecommunication CTWing platform to access NB-IoT equipment.The system and method of the present application are beneficial to the rapid information registration of wisdom community NB internet of things equipment, equipment platform registration, on-line and off-line and the analysis processing of service information data.

Description

Technical Field

[0001] This invention belongs to the field of Internet of Things (IoT) technology, specifically relating to a community NB device information collection and access system and method based on an OCR algorithm model. Background Technology

[0002] With the development and progress of society and the economy, people's demands for their living environment are gradually increasing. Smart communities, a new concept of community management under the smart city framework, represent a new model of community management innovation in the new era. Introducing smart security communities into community construction integrates information and communication technologies such as the Internet of Things and sensors, combining property management with resident services. This creates a safe, convenient, and comfortable modern living environment for community residents, forming a new management model based on information-based and intelligent community management and services. Currently, the implementation of smart community construction involves a large number of devices such as NB-IoT (Internet Protocol version 1) devices. The management of these devices is primarily based on on-site construction personnel, resulting in relatively rudimentary equipment data and access management. This leads to numerous operations, low efficiency, and time-consuming and labor-intensive debugging. Furthermore, in the later stages of equipment maintenance, there is a lack of specific on-site equipment images and other information for reference, making retrieval and maintenance difficult. Summary of the Invention

[0003] The purpose of this invention is to provide a community NB device information collection and access system and method based on an OCR algorithm model. This system and method facilitate efficient and convenient access to community networked NB devices, while also retaining basic information of the relevant devices in real time, providing data and image support for future operation and maintenance.

[0004] To achieve the above objectives, the technical solution of the present invention is: a community NB device information collection and access system based on an OCR algorithm model, comprising:

[0005] The APP terminal device based on OCR image recognition constructs a detection-recognition algorithm model based on OCR image recognition technology. The detection-recognition algorithm model is used to identify and locate the NB-IoT device to be connected, and outputs key information of the NB-IoT device, including device code, device model and device manufacturer. The data is then packaged into a unified data format and reported to the unified access platform for IoT devices in the smart community.

[0006] The unified access platform for IoT devices in smart communities is used to perform data matching and data association improvement on the key information of reported NB-IoT devices. If the key information of the corresponding NB-IoT device is not correctly matched on the unified access platform for IoT devices in smart communities, the corresponding NB-IoT device protocol needs to be developed and registered on the unified access platform for IoT devices in smart communities, and then data matching and data association improvement will be performed. The unified access platform for IoT devices in smart communities will push the improved key information to the message bus for unified processing.

[0007] The message bus is a data connection layer that connects the front-end access device gateway and the unified access platform for smart community IoT devices. It is used to asynchronously process uplink and downlink business message data between the front-end access device gateway and the unified access platform for smart community IoT devices, and to store and forward the message data.

[0008] The front-end access device gateway registers and communicates with NB-IoT devices through the CTWing platform. Its functions include registering key information of NB-IoT devices on the CTWing platform and parsing and processing services according to the device manufacturer's profile protocol.

[0009] This invention also provides a method for collecting and accessing community NB device information based on an OCR algorithm model using the above-mentioned system, comprising the following steps:

[0010] Step S1: The APP terminal device based on OCR image recognition takes pictures of the device nameplate information, detects and recognizes it, and outputs key information of the NB-IoT device, including the device code, device model and device manufacturer.

[0011] Step S2: The APP terminal device based on OCR image recognition uploads key information such as device code, device model and device manufacturer to the unified access platform for IoT devices in the smart community, performs device retrieval, matching and association improvement, and sends it back to the APP. After the user confirms the final information, the NB-IoT device asset information is entered on the platform, and the person who entered the information and the time of entry are automatically saved.

[0012] Step S3: After the user confirms the online status, the key information of the NB-IoT device entered in the registration will be further associated with the corresponding device's protocol processing library, front-end access gateway IP address, account, and password information in the unified access platform for smart community IoT devices. The basic associated data will be encrypted, encapsulated, and pushed to the data bus, which will then uniformly push it to the corresponding front-end access device gateway.

[0013] Step S4: When the current access device gateway receives the device online information pushed by the data bus, it will search in the existing device list to determine whether the NB-IoT device has been online; for devices that have been registered on the Telecom CTWing platform, the online information of the device will be returned directly to the unified access platform for IoT devices in the smart community.

[0014] Step S5: If the front-end access device gateway detects that the device is not in the list of online services, then based on the key information of the NB-IoT device, including the manufacturer number, on the unified access platform for smart community IoT devices, perform the NB-IoT device online registration operation on the access application of the telecom CTWing platform.

[0015] Step S6: After successful registration on the CTWing platform, the front-end access device gateway will attempt to call the status query instruction in the profile protocol library specified by the device to send an instruction to the NB-IoT device. If the NB-IoT device normally responds with a status protocol message and parses it normally, the current NB-IoT device is successfully online, and the gateway sends an online success message to the unified access platform for smart community IoT devices.

[0016] Step S7: If the front-end access device gateway receives a message, but the message information is abnormal and cannot be effectively parsed, it will send the received message information and the corresponding parsing protocol content as fault information to the unified access platform of smart community IoT devices. The operator will then confirm the fault information and reprocess it.

[0017] Step S8: If the front-end access device gateway attempts to issue different commands three times in a row at 10s, 20s, and 30s without receiving any information messages, the front-end access device gateway will temporarily remove the NB-IoT device from the online list queue and register the operation status and abnormal information as log information on the unified access platform for IoT devices in the smart community. At the same time, the operator and the back-end administrator will push the automatic online failure information of the NB-IoT device.

[0018] Step S9: The unified access platform for smart community IoT devices finally confirms the access status of the NB-IoT device based on the feedback information.

[0019] Compared to existing technologies, this invention offers the following advantages: By using a handheld terminal app based on an OCR image recognition engine to photograph and recognize devices, this invention quickly enables basic data entry and management on the platform. Simultaneously, it automatically compares and matches relevant devices against the platform's protocol library, matching the protocol parsing library with the corresponding access gateway server. Through application registration, device command and control, and service message parsing on the CTWing platform, it achieves automatic device network connection. Therefore, this invention has strong practicality and broad application prospects in the field of IoT access for NB-IoT devices. Attached Figure Description

[0020] Figure 1 This is a system architecture diagram of an embodiment of the present invention.

[0021] Figure 2 This is a flowchart illustrating the method implementation of an embodiment of the present invention.

[0022] Figure 3 This is a graph showing the prediction results of the measured-recognition algorithm model of the present invention. Detailed Implementation

[0023] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] This invention discloses a community NB device information collection and access system based on an OCR algorithm model, comprising:

[0025] The APP terminal device based on OCR image recognition constructs a detection-recognition algorithm model based on OCR image recognition technology. The detection-recognition algorithm model is used to identify and locate the NB-IoT device to be connected, and outputs key information of the NB-IoT device, including device code, device model and device manufacturer. The data is then packaged into a unified data format and reported to the unified access platform for IoT devices in the smart community.

[0026] The unified access platform for IoT devices in smart communities is used to perform data matching and data association improvement on the key information of reported NB-IoT devices. If the key information of the corresponding NB-IoT device is not correctly matched on the unified access platform for IoT devices in smart communities, the corresponding NB-IoT device protocol needs to be developed and registered on the unified access platform for IoT devices in smart communities, and then data matching and data association improvement will be performed. The unified access platform for IoT devices in smart communities will push the improved key information to the message bus for unified processing.

[0027] The message bus is a data connection layer that connects the front-end access device gateway and the unified access platform for smart community IoT devices. It is used to asynchronously process uplink and downlink business message data between the front-end access device gateway and the unified access platform for smart community IoT devices, and to store and forward the message data.

[0028] The front-end access device gateway registers and communicates with NB-IoT devices through the CTWing platform. Its functions include registering key information of NB-IoT devices on the CTWing platform and parsing and processing services according to the device manufacturer's profile protocol.

[0029] This invention also provides a method for collecting and accessing community NB device information based on an OCR algorithm model using the above-mentioned system, comprising the following steps:

[0030] Step S1: The APP terminal device based on OCR image recognition takes pictures of the device nameplate information, detects and recognizes it, and outputs key information of the NB-IoT device, including the device code, device model and device manufacturer.

[0031] Step S2: The APP terminal device based on OCR image recognition uploads key information such as device code, device model and device manufacturer to the unified access platform for IoT devices in the smart community, performs device retrieval, matching and association improvement, and sends it back to the APP. After the user confirms the final information, the NB-IoT device asset information is entered on the platform, and the person who entered the information and the time of entry are automatically saved.

[0032] Step S3: After the user confirms the online status, the key information of the NB-IoT device entered in the registration will be further associated with the corresponding device's protocol processing library, front-end access gateway IP address, account, and password information in the unified access platform for smart community IoT devices. The basic associated data will be encrypted, encapsulated, and pushed to the data bus, which will then uniformly push it to the corresponding front-end access device gateway.

[0033] Step S4: When the current access device gateway receives the device online information pushed by the data bus, it will search in the existing device list to determine whether the NB-IoT device has been online; for devices that have been registered on the Telecom CTWing platform, the online information of the device will be returned directly to the unified access platform for IoT devices in the smart community.

[0034] Step S5: If the front-end access device gateway detects that the device is not in the list of online services, then based on the key information of the NB-IoT device, including the manufacturer number, on the unified access platform for smart community IoT devices, perform the NB-IoT device online registration operation on the access application of the telecom CTWing platform.

[0035] Step S6: After successful registration on the CTWing platform, the front-end access device gateway will attempt to call the status query instruction in the profile protocol library specified by the device to send an instruction to the NB-IoT device. If the NB-IoT device normally responds with a status protocol message and parses it normally, the current NB-IoT device is successfully online, and the gateway sends an online success message to the unified access platform for smart community IoT devices.

[0036] Step S7: If the front-end access device gateway receives a message, but the message information is abnormal and cannot be effectively parsed, it will send the received message information and the corresponding parsing protocol content as fault information to the unified access platform of smart community IoT devices. The operator will then confirm the fault information and reprocess it.

[0037] Step S8: If the front-end access device gateway attempts to issue different commands three times in a row at 10s, 20s, and 30s without receiving any information messages, the front-end access device gateway will temporarily remove the NB-IoT device from the online list queue and register the operation status and abnormal information as log information on the unified access platform for IoT devices in the smart community. At the same time, the operator and the back-end administrator will push the automatic online failure information of the NB-IoT device.

[0038] Step S9: The unified access platform for smart community IoT devices finally confirms the access status of the NB-IoT device based on the feedback information.

[0039] The following are specific implementation examples of the present invention.

[0040] Please see Figure 1 This invention provides a rapid information collection and access system for community NB devices based on an OCR algorithm model, including an APP terminal based on an OCR image detection-recognition algorithm model, a unified access platform for smart community IoT devices, a front-end access gateway, and a message bus.

[0041] The APP terminal based on the OCR image detection and recognition algorithm model is used to take photos in real time and access the device, and detect and recognize information such as the device manufacturer code and type as basic device data and input it into the unified access platform.

[0042] The unified access platform for IoT devices in the smart community serves as the core hub of the system, responsible for managing basic device data, device access protocol library, device online / offline status, device fault management, and device business information.

[0043] The message bus is a middleware for message communication, used for pushing and receiving uplink and downlink message data between the device unified access platform and the access gateway server.

[0044] The aforementioned front-end access gateway service is the foundation for NB devices to access and register with the telecom CTWing platform, perform online / offline operations, and parse the specific working business status and process operation commands.

[0045] This system's access method is based on an APP terminal using an OCR image detection and recognition algorithm model. This terminal is used for real-time photo capture of accessing devices and identifies device manufacturer codes, types, and other information as basic device data, which is then entered into the unified access platform. The unified access platform for smart community IoT devices serves as the core hub of the system, primarily responsible for managing basic device data, device access protocol libraries, device online / offline management, device fault management, and device business information management. The message bus is the middleware for message communication, used for pushing and receiving uplink and downlink message data between the unified access platform and the access gateway server. The front-end access gateway service is the working foundation for NB devices to register with the telecom CTWing platform and for processing specific business instructions for device access. This system and method facilitate rapid information registration, device platform registration, online / offline status management, and business information data parsing and processing for NB IoT devices in smart communities.

[0046] Please see Figure 2 This invention provides a method for rapid information collection and access for community NB devices based on an OCR algorithm model, including the following steps:

[0047] Step S1: Based on the OCR image detection-recognition algorithm model, the data is converted into a standard time series format and noise reduction is performed to a certain extent. The data processing includes the following parts:

[0048] (1) Data import

[0049] (2) Data partitioning

[0050] (3) Data PCA noise reduction and normalization processing

[0051] (4) Constructing time series data

[0052] After the above processing, a sequential model is built using TensorFlow 2.0 to construct an LSTM model. The LSTM model mainly consists of two LSTM layers, two Dropout layers, and one Dense layer. Dropout is used to prevent overfitting, and Dense is used for output. Finally, the model uses the Adam algorithm as the optimizer, and the MSE loss function is selected. The model prediction results are presented in a two-dimensional coordinate graph, see details. Figure 3 .

[0053] Step S2: The terminal APP device based on the OCR image detection-recognition algorithm model takes a picture of the device nameplate and recognizes and extracts text information such as device manufacturer code and manufacturer model from the generated image text, coded letters and other features.

[0054] The device information is defined as shown in Table 1 below:

[0055] Table 1

[0056] KEY Full name byte illustrate id id 8 Device reporting ID number fid fatherId 8 Associated device parent node ID dk devicekey 32 Device Code dn devicename 32 Equipment Name dt devicetype 16 Equipment type df devicefactory 32 Equipment manufacturers pic picture 64 Images captured by the device lng longitude 16 Equipment longitude information lat latitude 16 Device Dimension Information

[0057] Step S3: Based on information such as device code and manufacturer model, the handheld terminal APP uploads the relevant information to the unified access platform for IoT devices in the smart community, performs device retrieval, matching, association and improvement, and sends it back to the APP. After the user's final confirmation, the device asset information is entered on the platform, and the person who entered the information and the entry time are automatically saved.

[0058] Step S4: After the user confirms online, the device information entered in the registration record will be further associated with the corresponding device's protocol processing library, front-end access gateway IP address, account, password and other information in the unified access platform for IoT devices in the smart community. The basic associated data will be encrypted, encapsulated and pushed to the data bus, and then uniformly pushed to the corresponding front-end access gateway server by the data bus.

[0059] Step S5: When the current access gateway server receives the device online information pushed by the data bus, it will search the existing device list to see if the device has been online. For devices that have been registered on the Telecom CTWing platform, the online information of the device will be returned directly to the unified access platform.

[0060] Step S6: If the front-end access gateway service detection device is not in the list of online services, then register the device online in the access application on the Telecom CTWing platform according to the device manufacturer number and other information.

[0061] Device registration instruction script:

[0062]

[0063]

[0064] Command response script:

[0065]

[0066] Step S7: After successful registration on the CTWing platform, the gateway server will attempt to call the status query instruction (such as querying the device's battery level) in the profile protocol library specified by the device to send an instruction to the device. If the device normally responds with a status (battery level) protocol message and parses it normally, the current device is successfully online, and the gateway sends an online success message to the unified access platform.

[0067] Command delivery script:

[0068]

[0069]

[0070] Device information feedback message:

[0071]

[0072] Step S8: If the gateway receives a message, but the message information is abnormal and cannot be effectively parsed, it will send the received message information and the corresponding parsing protocol content as fault information to the platform front end. The operator will then confirm the fault information and reprocess it (e.g., abnormal device type protocol input, re-offline editing and re-online, etc.).

[0073] Equipment normal analysis

[0074] Step S9: If the gateway attempts to issue different commands three times in a row at 10s, 20s, and 30s without receiving any feedback messages, the gateway will temporarily remove the device from the online list queue and register the operation status and abnormal information as log information on the unified access platform. At the same time, the operator and the backend administrator will be notified of the device's automatic online failure.

[0075] Step S10: The unified access platform for smart community IoT devices confirms the access status of the device based on the feedback information.

[0076] The above are preferred embodiments of the present invention. Any changes made to the technical solution of the present invention that do not exceed the scope of the technical solution of the present invention shall fall within the protection scope of the present invention.

Claims

1. A community NB device information collection and access system based on an OCR algorithm model, characterized in that, include: The APP terminal device based on OCR image recognition identifies and locates the NB-IoT device to be connected, outputs the key information of the NB-IoT device, and encapsulates the data in a unified format to report to the unified access platform for IoT devices in the smart community. The unified access platform for IoT devices in smart communities is used to perform data matching and data association improvement on the key information of reported NB-IoT devices. If the key information of the corresponding NB-IoT device is not correctly matched on the unified access platform for IoT devices in smart communities, the corresponding NB-IoT device protocol is developed and registered on the unified access platform for IoT devices in smart communities, and then data matching and data association improvement are performed. The improved key information is pushed to the message bus for unified processing. The message bus is a data connection layer that connects the front-end access device gateway and the unified access platform for smart community IoT devices; The front-end access device gateway registers and enables business communication with NB-IoT devices through the CTWing platform of China Telecom. The system's data acquisition and access methods include: Step S1: The APP terminal device based on OCR image recognition takes pictures, detects and recognizes the device nameplate information, and outputs key information of the NB-IoT device, including the device code, device model and device manufacturer. Step S2: Upload key information to the unified access platform for IoT devices in the smart community, perform device retrieval, matching, association and improvement, and send it back to the APP. After final confirmation by the user, log in the NB-IoT device asset information on the platform and automatically save the person who entered the information and the entry time. Step S3: After the user confirms online, the key information entered in the login will be further associated with the corresponding device's protocol processing library, front-end access gateway IP address, account, and password information in the unified access platform for IoT devices in the smart community. The basic associated data will be encrypted, encapsulated, and pushed to the data bus, which will then push it to the corresponding front-end access device gateway. Step S4: When the current access device gateway receives the device online information pushed by the data bus, it searches the existing device list to determine whether the NB-IoT device has been online; for devices that have been registered on the Telecom CTWing platform, the device online information is directly returned to the unified access platform for smart community IoT devices. Step S5: If the front-end access device gateway detects that the device is not in the list of online services, then based on the key information of the NB-IoT device, including the manufacturer number, on the unified access platform for smart community IoT devices, perform the NB-IoT device online registration operation on the access application of the telecom CTWing platform. Step S6: After successful registration on the CTWing platform, the front-end access device gateway will attempt to call the status query instruction in the profile protocol library specified by the device to send an instruction to the NB-IoT device. If the NB-IoT device normally responds with a status protocol message and parses it normally, the current NB-IoT device is successfully online, and the gateway sends an online success message to the unified access platform for IoT devices in the smart community. Step S7: If the front-end access device gateway receives a message, but the message information is abnormal and cannot be effectively parsed, the received message information and the corresponding parsing protocol content are sent as fault information to the unified access platform of smart community IoT devices. The operator confirms the fault information and processes it again. Step S8: If the front-end access device gateway attempts to issue different commands three times in a row at 10s, 20s, and 30s without receiving any information messages, the front-end access device gateway will temporarily remove the NB-IoT device from the online list queue and register the operation status and abnormal information as log information on the unified access platform for IoT devices in the smart community. At the same time, the operator and the back-end administrator will push the automatic online failure information of the NB-IoT device. Step S9: The unified access platform for smart community IoT devices finally confirms the access status of the NB-IoT device based on the feedback information.

2. The community NB device information collection and access system based on the OCR algorithm model according to claim 1, characterized in that, The APP terminal device based on OCR image recognition uses a detection-recognition algorithm model to quickly recognize and locate the NB-IoT device nameplate, output the key information of the NB-IoT device, and retrieve and match it with the smart community IoT device unified access platform. After the association is completed and the user confirms the information, the user can quickly log in to the platform. If the corresponding NB-IoT device cannot be matched normally, the key information and image information will be automatically saved and fed back to the backend detection-recognition algorithm model for deep retraining, and then reported to the development team.

3. The community NB device information collection and access system based on the OCR algorithm model according to claim 1 or 2, characterized in that, The detection-recognition algorithm model first converts the image into a grayscale image through image preprocessing, then converts it into a binary image, and performs edge detection and contour processing. Finally, it trains the detection model and the recognition model to achieve the output results of the detection-recognition model.

4. The community NB device information collection and access system based on the OCR algorithm model according to claim 1, characterized in that, The unified access platform for smart community IoT devices associates and completes the specific initialization information of NB-IoT devices based on the identified key information, including device type, device manufacturer's protocol parsing library, corresponding front-end access gateway IP address, account, and password. After data encapsulation and encryption, the data is pushed to the data bus, which then pushes it to the corresponding front-end access device gateway.

Images