A carbon metering and energy consumption acquisition and monitoring method and storage medium based on image recognition

Abstract

This invention discloses a carbon metering energy consumption acquisition and monitoring method and storage medium based on image recognition. The method includes: Step S1: Data acquisition; acquiring instrument images or video streams from various metering devices at the monitoring site through image recognition; Step S2: Data transmission; uploading the instrument images or video streams obtained in Step S1 to a cloud server via a wired or wireless network; Step S3: Image recognition; running a pre-trained machine learning image recognition model on the cloud server to recognize the received instrument images and extract meter reading information; Step S4: Data verification and storage; performing logical verification on the identified meter reading information; Step S5: Alarm step; sending an alarm message to the system administrator when the identified meter reading information shows data abnormality or the data stream is interrupted for more than a preset time. The storage medium is implemented based on the above method. This invention has the advantages of simple principle, wide applicability, and better compatibility.

Description

Technical Field

[0001] This invention mainly relates to the field of intelligent carbon emission metering technology, specifically a carbon metering energy consumption acquisition and monitoring method and storage medium based on image recognition. Background Technology

[0002] To address the severe challenges of global climate change, China has formulated a "dual carbon" strategic goal: to peak carbon dioxide emissions before 2030 and strive to achieve carbon neutrality before 2060. Researching and developing carbon emission measurement technologies and establishing relevant standards systems can guide energy-intensive enterprises to optimize their energy structures, scientifically and efficiently reduce corporate carbon emissions, and lay the foundation for peak carbon emissions and carbon neutrality.

[0003] To accurately and effectively reflect the carbon emission characteristics of enterprises' production activities, a robust real-time carbon emission data acquisition method and accounting system are indispensable. High-energy-consuming and high-carbon-emission enterprises face challenges such as complex carbon emission sources, broad monitoring dimensions, and an incomplete routine monitoring system. For example, the carbon emissions of a high-energy-consuming enterprise may involve multiple sources, including fossil fuel combustion, industrial production process emissions, indirect emissions such as electric heating, and carbon-containing gas leaks. While most carbon emission sources have mature metering and monitoring systems, the monitoring devices and systems are managed by different enterprises, and their data transmission channels are not interconnected, making it difficult to meet the centralized processing and calculation needs of multi-source carbon emission data for carbon metering. Adding additional carbon metering equipment to various energy metering pathways also faces numerous obstacles related to confidentiality, technology, and policy. Summary of the Invention

[0004] The technical problem to be solved by this invention is: in view of the technical problems existing in the prior art, this invention provides a carbon metering energy consumption acquisition and monitoring method and storage medium based on image recognition that is simple in principle, has a wide range of applications, and better compatibility.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A carbon metering and energy consumption acquisition and monitoring method based on image recognition, comprising: Step S1: Data Acquisition; Acquire instrument images or video streams from various metering devices at the monitoring site using image recognition; Step S2: Data transmission; Upload the instrument image or video stream obtained in step S1 to the cloud server via a wired or wireless network; Step S3: Image recognition; In the cloud server, a pre-trained machine learning image recognition model is run to recognize the received instrument image and extract the meter reading information; Step S4: Data verification and storage; perform logical verification on the identified meter reading information; Step S5: Alarm step; When the identified meter reading information shows abnormal data or the data stream is interrupted for more than a preset time, an alarm message is sent to the system administrator.

[0006] As a further improvement to the method of the present invention: in step S1, at least one network camera deployed at the metering device site periodically collects instrument images or video streams containing the dial or digital display area.

[0007] As a further improvement to the method of the present invention: in step S1, the metering device is a variety of sensors and / or detection components arranged at the monitoring site, including one or more of a natural gas meter, a water meter, and a steam meter.

[0008] As a further improvement to the method of the present invention: in step S3, the identification includes instrument positioning, digital area segmentation and character recognition.

[0009] As a further improvement to the method of the present invention: in step S4, if the verification passes, the meter reading information, along with the corresponding meter identifier and timestamp, is stored in the cloud database.

[0010] As a further improvement to the method of the present invention: in step S3, the machine learning image recognition model is a model based on a convolutional neural network; the machine learning image recognition model is trained by a labeled image dataset containing various lighting conditions, different angles, and different models of instruments.

[0011] As a further improvement to the method of the present invention: between step S3 and step S4, a threshold judgment is also included; whether the confidence level of the recognition result output by the machine learning image recognition model is higher than a preset threshold; if it is lower than the threshold, it is marked as low confidence data, and triggers re-acquisition of the image or issues a manual review alarm.

[0012] As a further improvement to the method of this invention: an image acquisition layer, a data transmission layer, a cloud processing layer, an alarm and human interaction layer, and a carbon metering application layer are arranged at the monitoring site; including: The image acquisition layer consists of network cameras arranged near each measuring device; The data transmission layer consists of a wireless network communication module connected to the network camera, used to upload the camera's instrument images to the cloud; The cloud-based analytics and computing layer consists of image recognition algorithm models and data verification algorithms deployed on cloud servers, used to identify and verify the code readings in images. The carbon measurement data platform communicates with the cloud identification service module to receive and store the verified meter readings, and calculates, analyzes and visualizes carbon emissions based on the readings and carbon emission factors.

[0013] As a further improvement to the method of the present invention: the cloud recognition service module integrates a model update unit, which is used to receive data after manual review and correction, and to incrementally train the machine learning image recognition model.

[0014] The present invention further provides a storage medium that can be read by a computer or processor, wherein the storage medium stores a computer program for executing any of the above methods.

[0015] Compared with the prior art, the advantages of the present invention are as follows: 1. The carbon metering energy consumption acquisition and monitoring method and storage medium based on image recognition of the present invention are simple in principle, have a wide range of applications and better compatibility. It adopts the method of installing network cameras near the original gas meters, water meters and other metering devices, and reads the real-time data of various metering devices in real time through machine learning image recognition and then uploads it to the carbon metering platform, without physical interference with the original metering path.

[0016] 2. The carbon metering energy consumption acquisition and monitoring method and storage medium based on image recognition of the present invention can make full use of various existing metering instruments in enterprises without replacing or adding additional acquisition devices, avoiding complex wiring and circuit modifications. The present invention only requires the addition of a general-purpose network camera, greatly reducing hardware costs and deployment difficulty. It also circumvents the restriction that certain metering loops cannot be connected to other devices; this solution has high versatility, and through cloud-based machine learning models, it can adapt to various metering devices of different environmental conditions and specifications; through confidence level judgment, multi-level logical verification mechanisms, and final manual review mechanisms, data recognition errors can be effectively avoided, ensuring the accuracy of the final data stored. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the method of the present invention in a specific embodiment.

[0018] Figure 2 This is a schematic diagram of the overall layout of the monitoring site in a specific embodiment of the present invention. Detailed Implementation

[0019] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] This invention is a method for collecting energy consumption data of various energy sources in current carbon emission metering. It targets existing metering devices for energy sources such as natural gas, steam, and water by adding network cameras and using machine learning image recognition on a cloud server to collect and identify meter readings from gas meters, water meters, and other metering devices in real time. The data is then uploaded to a carbon metering cloud platform for aggregation and analysis of carbon emission data.

[0021] like Figure 1 As shown, this invention discloses a carbon metering and energy consumption acquisition and monitoring method based on image recognition, which includes: Step S1: Data Acquisition; The instrument images or video streams of various metering devices on site are collected through image recognition. In a specific application example, this invention can periodically capture instrument images or video streams, including the dial or digital display area, using at least one network camera deployed at the metering device site. The metering device can be various sensors and / or detection components, such as a natural gas meter, water meter, steam meter, etc.

[0022] Step S2: Data transmission; The instrument image or video stream obtained in step S1 is uploaded to the cloud server via a wired or wireless network; Step S3: Image recognition; In the cloud server, a pre-trained machine learning image recognition model is run to recognize the received instrument images and extract the meter reading information. In specific application examples, the identification includes instrument positioning, digital area segmentation, and character recognition.

[0023] Step S4: Data verification and storage; Perform logical verification on the identified meter reading information; If the verification passes, the meter reading information, along with the corresponding meter identifier and timestamp, will be stored in the cloud database.

[0024] Step S5: Alarm procedure; When the identified meter reading information shows abnormal data or the data stream is interrupted for more than a preset time, an alarm message is sent to the system administrator.

[0025] In a specific application example, in step S3, the machine learning image recognition model is a convolutional neural network-based model. Furthermore, the machine learning image recognition model is trained using a labeled image dataset containing various lighting conditions, different angles, and different instrument models.

[0026] In a specific application example, between steps S3 and S4, the present invention further includes a threshold judgment; that is, whether the confidence level of the recognition result output by the machine learning image recognition model is higher than a preset threshold; if it is lower than the threshold, it is marked as low confidence data, and triggers re-acquisition of the image or issues a manual review alarm.

[0027] The present invention also provides a storage medium that can be read by a computer or processor, wherein the storage medium stores a computer program for executing any of the above methods.

[0028] Those skilled in the art will understand that the above embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0029] like Figure 2 The diagram shown illustrates the overall architecture of this invention in a carbon metering and energy consumption monitoring field. The overall architecture is divided into a field image acquisition layer, a data transmission layer, a cloud processing layer, an alarm and human interaction layer, and a carbon metering application layer, which includes: The image acquisition layer consists of network cameras placed in appropriate locations near meters such as gas meters, water meters, and steam meters, which take and upload photos at a frequency of once every 20 seconds. The data transmission layer consists of a wireless network communication module connected to the network camera, which is responsible for uploading the camera's instrument images to the cloud; The cloud-based analytics and computing layer consists of image recognition algorithm models and data verification algorithms deployed on cloud servers, used to identify and verify the code readings in images. The carbon measurement data platform communicates with the cloud identification service module to receive and store the verified meter readings, and calculates, analyzes and visualizes carbon emissions based on the readings and carbon emission factors.

[0030] Corresponding to the overall architecture described above, the specific monitoring setup at the site includes: The image acquisition module includes multiple network cameras deployed at the site of each energy metering device for acquiring instrument images; The data transmission module is communicatively connected to the image acquisition module and is used to upload instrument images to the cloud. Cloud-based analysis and computing module: Deployed on a cloud server, including a machine learning image recognition unit and a data verification unit, used to identify and verify meter readings in images; Carbon metering data platform, which communicates with the cloud-based identification service module, used to receive and store the verified meter readings, and to calculate, analyze, and visualize carbon emissions based on the readings and carbon emission factors.

[0031] In practical applications, according to the method of this invention, after receiving the instrument image, the instrument area in the image is first located. Then, the meter code display area is segmented from the instrument area, and the specific meter code digit is identified through a convolutional neural network. The confidence level of the identification result is then calculated. When the confidence level meets the requirements, the meter code reading is output. When the confidence level does not meet the requirements, image acquisition and identification are repeated, and an alarm signal is sent to the administrator.

[0032] Furthermore, the network camera can be an IP camera with Wi-Fi or 4G / 5G mobile network access capabilities.

[0033] Furthermore, the cloud-based recognition service module may be equipped with a model update unit, which is used to receive data after manual review and correction, and to incrementally train the machine learning image recognition model to continuously optimize the recognition accuracy.

[0034] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A carbon metering and energy consumption acquisition and monitoring method based on image recognition, characterized in that, include: Step S1: Data Acquisition; The system uses image recognition to collect instrument images or video streams from various metering devices at the monitoring site. Step S2: Data transmission; Upload the instrument image or video stream obtained in step S1 to the cloud server via a wired or wireless network; Step S3: Image recognition; In the cloud server, a pre-trained machine learning image recognition model is run to recognize the received instrument image and extract the meter reading information; Step S4: Data verification and storage; perform logical verification on the identified meter reading information; Step S5: Alarm step; When the identified meter reading information shows abnormal data or the data stream is interrupted for more than a preset time, an alarm message is sent to the system administrator.

2. The carbon metering and energy consumption acquisition and monitoring method based on image recognition according to claim 1, characterized in that, In step S1, at least one network camera deployed at the metering device site periodically captures instrument images or video streams containing the dial or digital display area.

3. The carbon metering and energy consumption acquisition and monitoring method based on image recognition according to claim 2, characterized in that, In step S1, the metering device is a variety of sensors and / or detection components arranged at the monitoring site, including one or more of a natural gas meter, a water meter, and a steam meter.

4. The carbon metering and energy consumption acquisition and monitoring method based on image recognition according to any one of claims 1-3, characterized in that, In step S3, the identification includes instrument positioning, digital area segmentation, and character recognition.

5. The carbon metering and energy consumption acquisition and monitoring method based on image recognition according to any one of claims 1-3, characterized in that, In step S4, if the verification passes, the meter reading information, along with the corresponding meter identifier and timestamp, is stored in the cloud database.

6. The carbon metering and energy consumption acquisition and monitoring method based on image recognition according to any one of claims 1-3, characterized in that, In step S3, the machine learning image recognition model is a model based on a convolutional neural network; the machine learning image recognition model is trained using a labeled image dataset containing various lighting conditions, different angles, and different models of instruments.

7. The carbon metering and energy consumption acquisition and monitoring method based on image recognition according to any one of claims 1-3, characterized in that, Between steps S3 and S4, a threshold determination is also included: whether the confidence level of the recognition result output by the machine learning image recognition model is higher than a preset threshold. If the data is below the threshold, it is marked as low-confidence data, and an alarm is triggered to re-acquire the image or issue a manual review warning.

8. The carbon metering and energy consumption acquisition and monitoring method based on image recognition according to any one of claims 1-3, characterized in that, The monitoring site is equipped with an image acquisition layer, a data transmission layer, a cloud processing layer, an alarm and human interaction layer, and a carbon metering application layer. These include: The image acquisition layer consists of network cameras arranged near each measuring device; The data transmission layer consists of a wireless network communication module connected to the network camera, used to upload the camera's instrument images to the cloud; The cloud-based analytics and computing layer consists of image recognition algorithm models and data verification algorithms deployed on cloud servers, used to identify and verify the code readings in images. The carbon measurement data platform communicates with the cloud identification service module to receive and store the verified meter readings, and calculates, analyzes and visualizes carbon emissions based on the meter readings and carbon emission factors.

9. The carbon metering and energy consumption acquisition and monitoring method based on image recognition according to claim 8, characterized in that, The cloud-based recognition service module integrates a model update unit, which is used to receive data after manual review and correction, and to incrementally train the machine learning image recognition model.

10. A storage medium capable of being read by a computer or processor, characterized in that, The storage medium stores a computer program for executing any one of the methods of claims 1 to 9.

Images