A carbon emission data audit system and method

Abstract

The present application discloses a carbon emission data audit method and system. The scheme includes a detection terminal and an audit terminal, and the method includes: obtaining the collection time period of the carbon emissions recorded in the carbon emission electronic report; selecting at least one time period to be audited from different collection time periods; reading the corresponding data to be audited from the detection terminal according to the selected time period to be audited, and the data to be audited includes an image to be audited; after performing image recognition on the image to be audited using the second carbon emission calculation method, determining the confidence of the carbon emission data corresponding to the data to be audited according to the image recognition result; and using a labeling method corresponding to the confidence to label and display the corresponding carbon emission data. It can be seen that by using this scheme to audit carbon emission data, not only the work efficiency is improved, but also the front-end collected data can be automatically verified and audited, thereby improving the accuracy of data auditing as a whole.

Description

Technical Field

[0001] The present invention relates to data processing and analysis technology, and in particular to a data audit analysis and processing system and method for carbon emissions. Background Art

[0002] At present, in order to realize the collection of carbon emission data for vehicles on roads or in some large parks, greenhouse gas detection devices or cameras with speed measurement functions are usually set up beside roads or in large parks to collect corresponding data on vehicle carbon emissions in the area, and then the carbon emission situation in the area is judged based on the collected carbon emission data, so as to provide sufficient and effective data basis for the subsequent formulation of green development strategies for this area or the optimization of relevant carbon emission management systems. It can be seen that the collected carbon emission data must be reviewed to ensure the accuracy of the carbon emission situation judged based on these data. However, the current review of carbon emission data is basically manual review, and it is difficult to trace and analyze the reasons for invalid and / or low trust in the data. Therefore, this not only imposes a heavy workload on the auditors, resulting in low work efficiency, but also makes it difficult to trace the reasons for invalid and / or low trust in the data, so it is difficult to provide guiding optimization and improvement suggestions for the collection and processing of carbon emissions, which is not conducive to the subsequent optimization and promotion of the carbon emission management system. Summary of the invention

[0003] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a carbon emission data audit system and method, which can improve the processing efficiency of the audit work.

[0004] In a first aspect, an embodiment of the present application provides a carbon emission data audit method, which is applied to an audit terminal, and the method includes:

[0005] S1. Obtain the collection time period of carbon emissions recorded in the carbon emissions electronic report;

[0006] S2. Select at least one time period to be reviewed from the different collection time periods obtained;

[0007] S3, reading from the detection terminal according to the selected time period to be reviewed to obtain at least one corresponding data to be reviewed, wherein the data to be reviewed includes an image to be reviewed;

[0008] S4. After performing image recognition on the image to be audited using the second carbon emission calculation method, an image recognition result is obtained, and the confidence level of the carbon emission data corresponding to the data to be audited is determined according to the image recognition result;

[0009] S5. Use a labeling method corresponding to the confidence level to label and display the corresponding carbon emission data.

[0010] In some embodiments, the second carbon emission calculation method includes a first image recognition algorithm, a second image recognition algorithm and a third image recognition algorithm, and the step S4 may specifically include:

[0011] S401, using a first image recognition algorithm to identify a vehicle model in the image to be reviewed, using a second image recognition algorithm to identify a stain area in the image to be reviewed, and using a third image recognition algorithm to identify a static target in the image to be reviewed;

[0012] S402: when it is determined that the recognition results of the vehicle models are all correct and there is no stain area or static target in the image to be reviewed, the confidence level of the carbon emission data corresponding to the image to be reviewed is set to a first confidence level;

[0013] S403: when it is determined that there is an incorrect recognition result of the vehicle model recognition result, and there is no stain area and static target in the image to be reviewed, a confidence confirmation process is performed on the incorrect recognition result;

[0014] S404: when it is determined that there is a stain area and / or a static target in the image to be reviewed, the confidence level of the carbon emission data corresponding to the image to be reviewed is set to a second confidence level;

[0015] The first confidence level is greater than the second confidence level.

[0016] In some embodiments, the confidence confirmation process for the incorrect recognition result in step S403 may specifically include:

[0017] S4031, obtaining the carbon emissions of a first vehicle corresponding to a historical vehicle target object, and recalculating the carbon emissions of a second vehicle corresponding to the historical vehicle target object after obtaining the corresponding carbon emission estimation model based on the newly identified vehicle model;

[0018] S4032, calculating a first difference between the carbon emissions of the first vehicle and the carbon emissions of the second vehicle;

[0019] S4033: when the first difference is less than the first threshold, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a third confidence level;

[0020] S4034: when the first difference is greater than or equal to the first threshold and less than the second threshold, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a fourth confidence level, wherein the fourth confidence level is in inverse proportion to the first difference;

[0021] S4035: when the first difference is greater than or equal to the second threshold, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a fifth confidence level;

[0022] The first confidence level is greater than or equal to the third confidence level, the third confidence level is greater than the maximum value of the fourth confidence level, and the minimum value of the fourth confidence level is greater than the fifth confidence level and the second confidence level.

[0023] In some embodiments, the second carbon emission calculation method further includes a fourth image recognition algorithm, and the step S4 may further include the following steps:

[0024] S405, after using the fourth image recognition algorithm to recognize the vehicle target object in the image to be reviewed, when it is determined that the vehicle target object is recognized in the image to be reviewed but the license plate cannot be recognized, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a sixth confidence level;

[0025] The minimum value of the fourth confidence level is greater than the sixth confidence level.

[0026] In some embodiments, the method further comprises the following steps:

[0027] S6. Determine the processing priority of the carbon emission electronic report based on the confidence level of the carbon emission data corresponding to the data to be reviewed.

[0028] In some embodiments, step S6 may specifically include:

[0029] S601: When the confidence level of the carbon emission data in the carbon emission electronic report is at least one of the second confidence level, the fifth confidence level, and the sixth confidence level, the processing priority of the carbon emission electronic report is the first priority level;

[0030] S602: When the confidence level of the carbon emission data in the carbon emission electronic report does not have the second confidence level, the fifth confidence level, and the sixth confidence level, the confidence levels corresponding to the plurality of carbon emission data in the carbon emission electronic report are summed up, and the processing priority of the carbon emission electronic report is determined to be the second priority level according to the calculated sum, and the second priority level is in inverse proportion to the sum;

[0031] The first priority is higher than the second priority.

[0032] In some embodiments, step S5 may specifically include:

[0033] S501, obtaining a recommended strategy corresponding to the confidence level;

[0034] S502. Use a labeling method corresponding to the confidence level to label the corresponding carbon emission data differently, and set a form button at a position corresponding to the carbon emission data, wherein the form button is used to pop up a first window after being clicked, and the first window displays the corresponding confidence level and recommended strategy.

[0035] In some embodiments, step S3 may specifically include:

[0036] S301, sending a query read request to the detection terminal, so that the detection terminal performs verification according to the ID of the audit terminal contained in the query read request, and after successful verification, querying and reading the pending audit data collected and recorded in the pending audit time period according to the pending audit time period contained in the query read request, and then encrypting the pending audit data read by the query using the first secret key, and returning the encrypted pending audit data to the audit terminal;

[0037] S302: After receiving the encrypted data to be reviewed returned by the detection terminal, the encrypted data to be reviewed is decrypted using the second secret key.

[0038] In some embodiments, the first image recognition algorithm is different from the image recognition algorithm recorded in the carbon emission electronic report, and the accuracy of the first image recognition algorithm is higher than the accuracy of the image recognition algorithm recorded in the carbon emission electronic report.

[0039] In a second aspect, an embodiment of the present application provides a carbon emission data audit system, the system comprising:

[0040] The detection terminal is used to detect and collect data on the carbon emissions of the vehicle and record the collected data;

[0041] An audit terminal, comprising at least one processor, for executing the steps of the carbon emission data audit method as described above;

[0042] The detection terminal is communicatively connected with the audit terminal.

[0043] The present application can achieve at least one of the following technical effects: after selecting the collection time period of the carbon emissions recorded in the carbon emissions electronic report, the present application scheme obtains the corresponding data to be audited from the detection terminal according to the selected time period to be audited, and then performs image recognition on the audited data, determines the confidence of the corresponding carbon emissions data based on the recognition result, and finally performs different annotations and displays on the carbon emissions data according to the different confidence levels. In this way, there is no need for auditors to manually trace the original data and verify the data offline, thereby improving work processing efficiency and accuracy. In addition, auditors can quickly understand the corresponding carbon emissions data through different annotation displays, which is conducive to auditors to quickly read and deal with subsequent corresponding solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments.

[0045] Figure 1 A schematic diagram of the steps of a carbon emission data collection method is provided for an embodiment of the present application;

[0046] Figure 2 A schematic diagram of the steps of a carbon emission data audit method is provided for an embodiment of the present application;

[0047] Figure 3 A schematic diagram showing the first window in step S502 provided in an embodiment of the present application;

[0048] Figure 4 A structural schematic diagram of a carbon emission data audit system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described through the implementation method with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0050] In order to facilitate the formulation of green transportation development strategies and the optimization and improvement of the management system of vehicle carbon emissions by enterprises or parks, it is necessary to collect vehicle carbon emission data, so that carbon emissions can be monitored in a timely manner and provide effective data basis for further analysis of carbon emissions. In order to ensure the accuracy and reliability of the data, enterprises or organizations need to submit the collected carbon emission data in the form of a report to a designated third-party authority for verification and review. However, the verification and review of the data of these carbon emission reports is basically manual review, and the content recorded in the report is limited. Therefore, when the auditors question the data, it is difficult to analyze and trace the validity / credibility of these data, and what are the corresponding causes of their generation, so it is difficult to provide modification suggestions and strategies. It can be seen that a carbon emission data audit solution that can reduce the workload of auditors, improve processing efficiency, and facilitate analysis and tracing is one of the problems that need to be solved urgently.

[0051] First of all, it should be noted that, since different carbon emission collection schemes involve different types of carbon emission data, in this embodiment, subsequent review and processing is mainly performed on the data involved in the following carbon emission data collection method.

[0052] Reference Figure 1 The carbon emission data collection method described below is mainly implemented based on the mapping relationship between the vehicle speed and fuel consumption, and the vehicle speed detection is implemented using image recognition technology. Specifically, the method steps included in the carbon emission data collection method are as follows.

[0053] C1. Acquire a first image captured by a first camera, wherein the first image includes at least one vehicle target object, and the first camera is a camera installed in a wide area / large place such as a highway and / or a park. The area / place mentioned here is an area / place where vehicles are allowed to travel.

[0054] C2. Perform vehicle type recognition on the vehicle target object in the first image to obtain a vehicle type recognition result.

[0055] Specifically, different types of vehicles have different properties such as the architecture they adopt, the functions they have, and the drive type, so the mapping relationship between the speed and fuel consumption of different types of vehicles will be different. Therefore, the type of vehicle captured in the image is identified, so that the corresponding carbon emission estimation model can be accurately and appropriately called to calculate the carbon emissions of different types of vehicles, thereby obtaining the carbon emissions of the highway area.

[0056] Among them, step C2 specifically includes the following steps.

[0057] C201. Select to use the license plate number recognition method to identify the vehicle type, and / or select to use the recognition method combining the license plate color, vehicle shape outline and vehicle logo to identify the vehicle type.

[0058] ①. The above-mentioned method of using the license plate number to identify the vehicle type may specifically include the following steps:

[0059] A1. Recognize the license plate number of the vehicle target object in the first image. For this step, the algorithm steps mainly used may be license plate location, character segmentation and character recognition in order to determine the license plate number.

[0060] A2. According to the identified license plate number, the vehicle model corresponding to the license plate number is queried from a preset database.

[0061] A3. Determine the driving type of the vehicle target object according to the vehicle model.

[0062] Specifically, the preset database can also store the mapping relationship between the vehicle model and the drive type, so that when the vehicle model is determined, the drive type corresponding to the vehicle model can be directly obtained from the preset database, so that the drive type of the vehicle target object can be obtained. Of course, if the preset database does not store the mapping relationship between the vehicle model and the drive type, then the vehicle model can be directly used to identify and match the corresponding drive type in other public vehicle information databases.

[0063] ②. The above-mentioned method of combining the license plate color, vehicle outline and vehicle logo to identify the vehicle type may specifically include the following steps:

[0064] B1. Recognize the license plate color, vehicle outline and vehicle logo of the vehicle target object in the first image.

[0065] B2. Filter out several preliminary selected vehicle models corresponding to the recognized license plate color from a preset database.

[0066] B3. Selecting a number of candidate vehicle models corresponding to the identified vehicle logo from the number of initially selected vehicle models.

[0067] B4. After calculating the similarity between the identified vehicle outline and the vehicle outlines corresponding to several candidate vehicle models, the vehicle model corresponding to the highest similarity is used as the final vehicle model.

[0068] B5. Determine the driving type of the target vehicle object according to the final vehicle model.

[0069] Specifically, since the license plate color of fuel vehicles is blue and the license plate color of new energy vehicles is green, the vehicle models belonging to fuel vehicles or new energy vehicles can be screened out using the identified license plate color to achieve rough screening; then, the vehicle models belonging to fuel vehicles or new energy vehicles can be further screened using the identified vehicle logo, which can reduce the number of vehicle contour matching and identification required; finally, based on the matching of the identified vehicle contour with the vehicle contour stored in the database, the vehicle contour with the highest matching degree (i.e., the highest similarity) can be identified; finally, based on the vehicle contour with the highest matching degree, the final vehicle model can be determined, thereby obtaining the drive type corresponding to the vehicle model, i.e., the drive type of the vehicle target object. It can be seen that by screening based on the license plate color, vehicle logo, and vehicle contour step by step, the drive type of the vehicle target object can be quickly and accurately identified.

[0070] In addition, it should be noted that compared with the matching and identification of license plate color and car logo, the algorithm of vehicle outline is the most complex, and in this field, vehicles of different brands may have very similar outlines, or fuel vehicles and new energy vehicles of the same brand may have very similar outlines. Therefore, only using the vehicle outline to identify and match the outline with the highest similarity may result in multiple recognition results. It can be seen that the implementation method of reducing the above steps B2 and / or B3 will result in low efficiency and accuracy.

[0071] C3. According to the vehicle type recognition result, a corresponding carbon emission estimation model is obtained, wherein the carbon emission estimation model is a mapping relationship model between vehicle speed and carbon emissions.

[0072] Specifically, since the vehicle type recognition result includes the driving type of the vehicle target object, and the driving type includes a fuel driving type, a pure electric driving type, and a hybrid driving type, the step C3 may specifically include the following steps:

[0073] C301. When the driving type is a fuel driving type, obtaining a corresponding first carbon emission estimation model; and / or,

[0074] C302. When the driving type is a hybrid driving type, the current driving mode of the vehicle target object is obtained from the information access port of the vehicle system; when it is determined that the current driving mode is a fuel driving mode, the corresponding second carbon emission estimation model is obtained.

[0075] Alternatively, even if the driving type is fuel-driven, the corresponding carbon emission estimation models may be different for different vehicle models. Therefore, for C3, it may also include:

[0076] C303. Obtain a corresponding third carbon emission estimation model based on the vehicle model and the drive type. That is, based on the vehicle model and its drive type, a more refined and appropriate carbon emission estimation model is further determined. For example, although all fuel-driven vehicles have different vehicle models, their corresponding carbon emission estimation models may be different. In order to accurately obtain the corresponding carbon emission estimation model, the final required carbon emission estimation model may be determined based on the vehicle model and the drive type for subsequent use.

[0077] Specifically, for the above-mentioned mapping relationship model between vehicle speed and carbon emissions, the expression function formula adopted can be:

[0078]

[0079] Where, E represents the carbon emissions per 100 kilometers, in mg / (100km); v is the speed of the vehicle, in km / h; a, b, c are coefficients obtained after curve fitting, all of which are constants; ρ is the density of the fuel liquid, in kg / L; Q is the default net calorific value of the fuel, in MJ / kg; F i is the emission factor based on net calorific value, in mg / MJ; I i is the characteristic factor. n is the total number of emission factors and characteristic factors based on net calorific value.

[0080] C4. After obtaining the speed of the vehicle target object, the carbon emissions of the area are determined according to the obtained speed and the carbon emissions estimation model.

[0081] It can be seen that by adopting the above-mentioned collection method, the carbon emission estimation model corresponding to each vehicle target object traveling in the area can be determined, and then the carbon emissions of the vehicle target objects traveling in the area can be obtained based on the speed of the identified vehicle target objects and the corresponding carbon emission estimation model. Therefore, by counting the carbon emissions of the vehicle target objects appearing in the area in any time period, the vehicle carbon emissions in the area in different collection time periods can be obtained, and the total carbon emissions can be obtained by summing up the carbon emissions in different time periods.

[0082] In order to ensure the reliability and accuracy of the above-mentioned data collection scheme, regular data verification and audit are required, which will facilitate the subsequent accurate analysis of carbon emissions, thereby facilitating the reliable and accurate formulation and optimization of subsequent related carbon emission management strategies and green planning strategies.

[0083] After analysis, it was found that the mapping relationship between the vehicle model and the driving type of the vehicle target object, as well as the mapping relationship between the driving type and the carbon emission estimation model, are basically pre-stored for the carbon emission results in the carbon emission report, and can be directly called when used. Therefore, the main reasons for the error influence include: 1. The error of the image recognition algorithm. Since the error of the image recognition algorithm adopted by different enterprises, organizations or units will be different, the error of the data collected based on the above collection scheme will be different; 2. The physical error of the camera. For example, the lens of the camera equipment is stained, which causes the license plate of the vehicle in the captured image to be stained. The camera is covered by stains, or, because the camera equipment is installed outdoors, it is easily affected by artificial or natural external factors, which causes the camera equipment to flip, so that the installation angle of the camera lens cannot be well aimed at the vehicle license plate, for example, rotating upward so that the camera always shoots nearby plants, buildings or the sky, or rotating downward so that the camera can only shoot the roof. It can be seen that this will greatly affect the detection results of carbon emissions of vehicles traveling in the area; 3. The calibration frequency of the equipment. If the calibration frequency is higher, the confidence of the carbon emission results detected by the collection is higher, and vice versa. Even, if the calibration record of the equipment cannot be found, the data collected in the corresponding time period should be marked as low confidence or invalid. It can be seen that for the above-mentioned collection scheme, the main reason for the error in the collected carbon emissions comes from the front-end image capture equipment. Therefore, by reviewing the data collected by this equipment, it can be verified whether the carbon emission data collected by this scheme is accurate / reliable.

[0084] Reference Figure 2 The steps of the data review and analysis processing method for the carbon emission report formed by the data collected by the above collection scheme can be as follows. Among them, this carbon emission data review and analysis processing method is mainly used in the audit terminal.

[0085] S1. Obtain the collection time period of the carbon emissions recorded in the carbon emissions electronic report.

[0086] Specifically, the general carbon emission report will contain at least two parts: 1. Carbon emission calculation method, including the algorithm, formula and data source used, as well as the processing error rate of the image algorithm and formula used, that is, the processing accuracy; 2. Carbon emission results, generally including total carbon emissions, carbon emissions collected in different time periods, that is, different collection time periods and carbon emission data collected within the collection time period. It can be seen that several carbon emission collection time periods can be obtained from the carbon emission electronic report.

[0087] S2. Select at least one time period to be reviewed from the different acquired collection time periods.

[0088] Specifically, for all the collection time periods in the carbon emission report, you can choose to audit and verify the source data corresponding to each collection time period, or you can select at least one collection time period from all the collection time periods as the time period to be audited according to a preset rule or random sampling. Preferably, it is best to use all the collection time periods as the time period to be audited, that is, for step S2, it can be: all the acquired collection time periods are used as the time period to be audited.

[0089] S3. Reading at least one corresponding data to be reviewed from the detection terminal according to the selected time period to be reviewed, wherein the data to be reviewed includes an image to be reviewed.

[0090] Specifically, a carbon emission report generally includes the purpose and scope of the report, carbon emission calculation methods, carbon emission results, etc., but does not include the data source corresponding to the carbon emission results, that is, the data collected by the front end. For example, based on the above-mentioned collection scheme, since the final carbon emission result is a data result obtained after relevant processing of the image data collected by the front-end camera equipment, the image data collected by the front-end camera equipment can be the data source corresponding to the carbon emission result.

[0091] In view of this, in order to verify and audit the carbon emission results, the audit terminal needs to obtain authorization from the detection terminal, and the two are connected in communication. The audit terminal can successfully query and obtain the data collected during the selected time period from the detection terminal as the data to be audited, where the data to be audited mainly includes the images collected during the collection time period, that is, the images to be audited. For example, the selected time period to be audited is T0~T1, T3~T4,..., Tn-1~Tn, and for the detection terminal, it will record and save the collected data and the corresponding time, so the corresponding collected data can be obtained from the detection terminal according to the above-mentioned time to be audited, that is, the data to be audited.

[0092] In some embodiments, the data detected by the detection terminal involves the business operation data of an enterprise or organization. Therefore, when the detection terminal is authorized to open the communication interface, in order to avoid being stolen by an illegal third party and causing data leakage, in addition to the need for the detection terminal to verify the terminal ID of the audit terminal, the data to be transmitted also needs to be encrypted. Therefore, for the step S3, it can specifically include the following steps.

[0093] S301. Send a query read request to the detection terminal to enable the detection terminal to perform verification according to the ID of the audit terminal contained in the query read request. After the verification is successful, query and read the pending audit data collected and recorded within the pending audit time period contained in the query read request, and then use the first secret key to encrypt the pending audit data read by the query, and return the encrypted pending audit data to the audit terminal.

[0094] S302: After receiving the encrypted data to be reviewed returned by the detection terminal, the encrypted data to be reviewed is decrypted using the second secret key.

[0095] Specifically, the first key and the second key may be the same or different, which mainly depends on whether it is a symmetric encryption algorithm or an asymmetric encryption algorithm. In this embodiment, since the data transmission security is higher than the data processing efficiency, an asymmetric encryption algorithm is preferred. As for the second key of the audit terminal, it can be obtained from other channels, which is not limited in detail in this embodiment.

[0096] After the encrypted data to be reviewed is decrypted in step S302, the plain text of the data to be reviewed is obtained, and then the confidence analysis and processing can be performed on the plain text of the data to be reviewed in subsequent steps.

[0097] S4. After performing image recognition on the image to be audited using the second carbon emission calculation method, an image recognition result is obtained, and the confidence level of the carbon emission data corresponding to the data to be audited is determined according to the image recognition result.

[0098] Specifically, the data collected by the front-end device is audited and verified, and the confidence level of the corresponding carbon emission results can be determined based on different audit and verification results. For example, if the audit and verification results of the data are unqualified, qualified, accurate, etc., the corresponding confidence levels will be different. The confidence level mainly refers to the degree of trust or reliability of the data.

[0099] In the embodiment of the present application, the data to be reviewed is mainly the first image obtained by photographing the area. Therefore, for the second carbon emission calculation method, it also mainly includes different image recognition algorithms, so as to perform different image recognition on the first image, and then determine whether the vehicle model is accurately identified when the carbon emission data is collected and processed based on the content identified from the image, and whether there are adverse external factors that make it impossible to correctly identify the vehicle model from the image, or even make it impossible to identify the vehicle model when there is a vehicle target object, thereby resulting in the inability to obtain the correct carbon emission estimation model in the future to calculate the accurate carbon emission results.

[0100] Based on this, for the second carbon emission calculation method, it may include a first image recognition algorithm, a second image recognition algorithm and a third image recognition algorithm to perform image recognition of different targets on the image to be reviewed; and for step S4, it may specifically include the steps as shown below.

[0101] S401, using a first image recognition algorithm to identify the vehicle model in the image to be reviewed, using a second image recognition algorithm to identify the stain area in the image to be reviewed, and using a third image recognition algorithm to identify the static target in the image to be reviewed.

[0102] Specifically, the first image recognition algorithm, the second image recognition algorithm, and the third image recognition algorithm are generally different algorithms. Of course, if there is an algorithm that can identify the vehicle model, the stain area, and the static target when processing the image, then the first to third image recognition algorithms can also be referred to as the same algorithm. Therefore, no further limitation is made in this embodiment.

[0103] In addition, for the above-mentioned first image recognition algorithm, it is mainly used to identify the vehicle model in the image, and the carbon emission electronic report also records the image recognition algorithm used in its carbon emission calculation scheme to identify the vehicle model, that is, the historical image recognition algorithm. Therefore, in order to further improve the accuracy of data audit verification, it should be preferred that for the selection of the first image recognition algorithm, this step is: obtain the image recognition algorithm (fifth image recognition algorithm) for vehicle model recognition recorded in the carbon emission electronic report, and then select the first image recognition algorithm that is different from the fifth image recognition algorithm and has higher image recognition accuracy. That is to say, in one embodiment, for the first image recognition algorithm, it is preferably different from the fifth image recognition algorithm, and the recognition accuracy of the first image recognition algorithm is higher than the recognition accuracy of the fifth image recognition algorithm. Of course, if based on other considerations, it is necessary to select the first image recognition algorithm and the fifth image recognition algorithm as the same, it is also possible.

[0104] For the above three different types of recognition, the specific recognition processing steps are respectively as follows.

[0105] S4011. Use a first image recognition algorithm to identify the vehicle model in the image to be reviewed.

[0106] S40111. After using the first image recognition algorithm to recognize the vehicle model in the image to be reviewed, a newly recognized vehicle model is obtained.

[0107] S40112: Query the historical vehicle model corresponding to the image to be reviewed from the historical data recorded and stored in the detection terminal. The historical vehicle model refers to the historical vehicle model obtained after vehicle model recognition of the first image during the historical collection period.

[0108] S40113. Determine whether the newly identified vehicle model is the same as the historical vehicle model. If so, it means that the recognition result of the vehicle model identified in the image to be reviewed is correct. Otherwise, it means that the recognition result of the vehicle model identified in the image to be reviewed is wrong.

[0109] Specifically, in order to further ensure the correctness of the verification and review, the above step S40113 may include:

[0110] The image to be reviewed, the newly identified vehicle model, the historical vehicle model, and the judgment result of whether they are the same are all displayed on a first pop-up window, wherein the first pop-up window is provided with a first button and a second button; wherein the first button is used to indicate that the newly identified vehicle signal is judged to be the same as the historical vehicle model, and the second button is used to indicate that the newly identified vehicle signal is judged to be different from the historical vehicle model;

[0111] When it is detected that the first button is clicked, it means that the recognition result of the vehicle model recognized in the image to be reviewed is correct, and when it is detected that the second button is clicked, it means that the recognition result of the vehicle model recognized in the image to be reviewed is wrong.

[0112] It can be seen that through the specific implementation of the above step S40113, the image to be reviewed, the newly identified vehicle model, the historical vehicle model and the judgment result of whether they are the same can be further displayed to the reviewer for review. When the reviewer reviews and confirms that the displayed information is correct, he clicks the first button to confirm submission, otherwise, he clicks the second button to confirm submission. It can be seen that through this method, the accuracy of the judgment result of whether the newly identified vehicle model is the same as the historical vehicle model can be further guaranteed.

[0113] S4012: Use a second image recognition algorithm to identify the stain area in the image to be reviewed.

[0114] Specifically, since the camera equipment is set up outdoors, the lens of the equipment may be stained, which will cause a stained area to exist at the corresponding position in the captured image, resulting in the situation that even if a vehicle passes by, the license plate cannot be recognized from the captured image. In other words, due to the presence of a stained area on the captured image, even if a vehicle passes by, it cannot be recognized, which will cause the final calculated carbon emissions value to be less than the actual carbon emissions generated by the vehicle, or even much less. Therefore, it is necessary to identify the stained area in the image to be reviewed.

[0115] In one embodiment, if the area of ​​the stain region is small enough and is located at the edge of the image to be reviewed, the stain region will not affect the license plate recognition in the image, so step S4012 may include the following steps.

[0116] S40121. When a number of candidate stain regions are identified from the image to be reviewed, the areas and positions of the candidate stain regions are determined.

[0117] S40122. When the areas of all candidate stain regions are smaller than the first area threshold, and all candidate stain regions are at the edge of the image to be reviewed, it is determined that there is no stain region in the image to be reviewed. It should be noted that the edge of the image to be reviewed may be a preset position area, and its width may be limited as required, so the shape and size of the edge are not specifically limited here, and may be set according to actual conditions.

[0118] S40123. When the area of ​​at least one candidate stain region is greater than or equal to the first area threshold, and / or at least one candidate stain region is not at the edge of the image to be reviewed, it is determined that there is a stain region in the image to be reviewed. In other words, as long as the area of ​​a candidate stain region is greater than or equal to the first area threshold, no matter where it is located in the image, it is determined that there is a stain region in the image to be reviewed. Similarly, as long as a candidate stain region is not at the edge of the image to be reviewed, no matter how large its area is, it is determined that there is a stain region in the image to be reviewed.

[0119] It can be seen that by adopting the above judgment method, the misjudgment rate of audit verification can be reduced. In addition, for the identification of the stain area, it can be identified by using the existing image recognition algorithm, which is not elaborated and limited here.

[0120] S4013. Use a third image recognition algorithm to identify static objects in the image to be reviewed.

[0121] Specifically, since the camera device is set outdoors, it is easy to be turned upside down or left and right due to external forces, which will cause the lens of the camera device to deviate from the driving area of ​​the vehicle. There will be some known static targets near the driving area of ​​the vehicle, such as iconic buildings, trees, shops, etc., and even if the camera device is turned upside down, the static target will be the sky. Therefore, the third image recognition algorithm is used to identify the static targets in the image to be reviewed. Its main purpose is to identify the known static targets, so as to determine whether the installation angle of the camera device has changed, so as to determine that the area captured by the camera has deviated from the driving area of ​​the vehicle. If so, it means that even if a vehicle passes through the area, the image captured by the camera device can no longer correctly recognize the license plate, which will also cause the final calculated carbon emissions to be less than the actual carbon emissions generated by the driving vehicle, or even much less. Therefore, it is necessary to identify some specific static targets in the image to be reviewed to determine whether the installation angle of the camera has deviated from the effective angle.

[0122] In addition, for the identification of static targets, it can be identified by using existing image recognition algorithms, which are not elaborated and limited here. Moreover, when there are a certain number of candidate static targets in the image, what is the proportion of the candidate static targets in the image, and what is the position distribution of the candidate static targets in the image, it is finally determined whether there are static targets in the image to be reviewed. These preset conditions can be set and adjusted according to the actual situation of different regions, and are not specifically limited here.

[0123] S402: When it is determined that the vehicle model recognition results are all correct and there is no stain area or static target in the image to be reviewed, the confidence level of the carbon emission data corresponding to the image to be reviewed is set to the first confidence level. It can be seen that in this case, the confidence level of the carbon emission data processed based on the image to be reviewed is the highest.

[0124] S403: When it is determined that there are incorrect recognition results of the vehicle model, and there is no stain area and static target in the image to be reviewed, a confidence confirmation process is performed on the incorrect recognition result.

[0125] Specifically, although there are incorrect identification results in the vehicle model identification results, if the newly identified vehicle model and the historical vehicle model have similar corresponding carbon emission estimation models, resulting in similar differences in the carbon emission data finally calculated and processed, then the confidence level of the carbon emission data can also be set higher. However, in the subsequent feedback, the enterprise or third party organization can be told that an error occurred in the verification and review of the data source, but the data results in the report do not need to be modified because they are within the allowable error range.

[0126] In view of this, the confidence confirmation process for the incorrect recognition result in step S403 may specifically include the following steps.

[0127] S4031. Obtain the carbon emissions of a first vehicle corresponding to a historical vehicle target object, and after obtaining a corresponding carbon emission estimation model based on a newly identified vehicle model, recalculate the carbon emissions of a second vehicle corresponding to the historical vehicle target object.

[0128] Specifically, since the carbon emission data corresponding to the historical vehicle model of the historical vehicle target object has been recorded and stored in the detection terminal, it can be directly obtained. For the vehicle model re-identified for the historical vehicle target object, after obtaining the corresponding carbon emission estimation model, based on the newly obtained carbon emission estimation model and the historical vehicle speed corresponding to the historical vehicle target object (this is also the data recorded and stored in the detection terminal, which can be directly read and obtained), the carbon emission data corresponding to the historical vehicle target object is recalculated, that is, the second vehicle carbon emission.

[0129] S4032. Calculate a first difference between the carbon emissions of the first vehicle and the carbon emissions of the second vehicle.

[0130] S4033: When the first difference is less than the first threshold, the confidence level of the carbon emission data corresponding to the image to be reviewed is set to a third confidence level.

[0131] Specifically, the first threshold value can be set to a relatively small value so that it is within the allowable error range. Furthermore, if the first threshold value is small enough, the first difference value can be considered to be negligible. At this time, the third confidence level is similar to the first confidence level, that is, the third confidence level can be the same as the first confidence level. However, in general, the third confidence level should be smaller than the first confidence level.

[0132] S4034. When the first difference is greater than or equal to the first threshold and less than the second threshold, the confidence level of the carbon emission data corresponding to the image to be reviewed is set to a fourth confidence level, wherein the fourth confidence level is in inverse proportion to the first difference.

[0133] Specifically, since the first difference is still within the acceptable range and does not fall within the unqualified line, the fourth confidence level of the carbon emission data corresponding to the image to be reviewed can be determined based on the first difference. The larger the first difference, the lower the fourth confidence level. In other words, the fourth confidence level has a value range of [a, b].

[0134] S4035: When the first difference is greater than or equal to the second threshold, the confidence level of the carbon emission data corresponding to the image to be reviewed is set to a fifth confidence level.

[0135] Specifically, when the first difference is greater than or equal to the second threshold, the error has already reached the unqualified line, and the confidence value of the carbon emission data should be lower than the fourth confidence value.

[0136] From the above, it can be obtained that the first confidence level is greater than or equal to the third confidence level, the third confidence level is greater than the maximum value b of the fourth confidence level, and the minimum value a of the fourth confidence level is greater than the fifth confidence level.

[0137] S404. When it is determined that there are stain areas and / or static targets in the image to be reviewed, the confidence of the carbon emission data corresponding to the image to be reviewed is set to the second confidence. It can be seen that in this case, the license plate cannot be correctly and effectively identified from the image to be reviewed, which will result in a vehicle passing by. However, since the license plate cannot be identified, it will be misjudged as no vehicle passing by. This will cause the carbon emission data processed based on the image to be reviewed to have a relatively large error, so its second confidence will be relatively low, lower than the first confidence, that is, the first confidence is greater than the second confidence. Further, based on the above steps, it can be seen that the minimum value a of the fourth confidence is greater than the second confidence.

[0138] It should be noted that, since both the fifth confidence level and the second confidence level are confidence levels corresponding to unqualified situations, the difference between the fifth confidence level and the second confidence level does not need to be specifically limited in this embodiment.

[0139] In one embodiment, the second carbon emission calculation method further includes a fourth image recognition algorithm, and the step S4 may further include the following steps:

[0140] S405, after using the fourth image recognition algorithm to recognize the vehicle target object in the image to be reviewed, when it is determined that the vehicle target object is recognized in the image to be reviewed but the license plate cannot be recognized, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a sixth confidence level;

[0141] The minimum value of the fourth confidence level is greater than the sixth confidence level.

[0142] Specifically, when the installation angle of the camera device is flipped downward, the lens of the camera device can only capture the outline of the moving vehicle, but cannot capture the license plate of the moving vehicle. For example, when the camera device is flipped downward, its camera lens is almost pointing to the ground, then only the roof of the vehicle can be captured, and the license plate cannot be identified from the captured image. Therefore, when the image recognition algorithm is used to identify the vehicle from the image to be reviewed, but the license plate cannot be identified, the final calculated carbon emissions value will be less than the actual carbon emissions generated by the moving vehicle, or even much less. It can be seen that when this situation is identified from the image to be reviewed, the confidence level of the corresponding carbon emissions data should also be set relatively low.

[0143] Similarly, in view of the fact that the above situation is an unqualified situation, the sixth confidence level and its difference from the fifth confidence level and the second confidence level do not need to be specifically limited in this embodiment.

[0144] S5. Use a labeling method corresponding to the confidence level to label and display the corresponding carbon emission data.

[0145] Specifically, different situations identified for the image to be reviewed correspond to different confidence levels, so the corresponding problems can be determined according to the different confidence levels. In this way, according to the different identified confidence levels, the corresponding modification suggestion strategy content can be queried and obtained. In this way, different degrees of annotation are performed on the carbon emission data with different confidence levels, which is beneficial for the auditor to review and review, and the corresponding modification suggestion strategy can also be displayed. This can further save the workload of the auditor's review and modification feedback and improve the work processing efficiency. Therefore, in one embodiment, the step S5 may include the following steps.

[0146] S501: Obtain a recommended strategy corresponding to the confidence level.

[0147] Specifically, the suggested strategy mainly includes the image recognition result for the image to be reviewed and the corresponding suggested modification strategy. For example, if the image recognition result is that the installation angle of the camera device is deviated, then the suggested modification strategy may be to further inspect the installation of the camera device.

[0148] S502: using a labeling method corresponding to the confidence level, labeling the corresponding carbon emission data differently, and setting a form button at a position corresponding to the carbon emission data, wherein the form button is used to pop up a first window after being clicked, and the first window displays the corresponding confidence level and the recommended strategy. A close button is set on the first window.

[0149] Specifically, since carbon emission data with low confidence levels should receive more attention and be processed with priority, the lower the confidence level, the more obvious the corresponding carbon emission data should be labeled. In addition, in order to facilitate the review of auditors and avoid modifying the report format itself, the following table contains the carbon emission data with low confidence levels: Figure 3 As shown, the display of image recognition results, confidence and recommended strategies is achieved by using a form button and a pop-up window (i.e., the first window). The confidence and recommended strategies are displayed by clicking a form button (“○” is used as the embodiment of the form button in the figure). A close button is provided on the pop-up window, so when you do not need to view it, you can click the close button on the pop-up window to close it.

[0150] S6. Determine the processing priority of the carbon emission electronic report based on the confidence level of the carbon emission data corresponding to the data to be reviewed.

[0151] Specifically, the confidence level of carbon emission data is not only used as the basis for selecting the annotation method, but also as the basis for the processing order of the carbon emission electronic report. Since different pending review time periods correspond to different pending review images, and different pending review images also correspond to different confidence levels, the processing priority of the carbon emission electronic report should be obtained only after all the determined confidence levels are judged and processed.

[0152] In some embodiments, step S6 may specifically include the following steps.

[0153] S601. When the confidence level of the carbon emission data in the carbon emission electronic report has at least one of the second confidence level, the fifth confidence level, and the sixth confidence level, the processing priority of the carbon emission electronic report is the first priority.

[0154] Specifically, since the image verification situations corresponding to the second, fifth, and sixth confidence levels are all high-error unqualified situations, and require a long time to correct, as long as any of the second, fifth, and sixth confidence levels exists, the processing priority of this report is the first priority. After the auditor confirms the review, it can be fed back to the enterprise for corresponding corrections.

[0155] S602. When the confidence level of the carbon emission data in the carbon emission electronic report does not include the second confidence level, the fifth confidence level and the sixth confidence level, the confidence levels corresponding to the several carbon emission data in the carbon emission electronic report are summed up, and the processing priority of the carbon emission electronic report is determined to be the second priority level based on the calculated sum, and the second priority level is inversely proportional to the sum; wherein the first priority level is higher than the second priority level, and there may be a lowest priority level and a highest priority level in the second priority level, in which case the first priority level should be higher than the highest priority level in the second priority level.

[0156] Specifically, when the second confidence level, the fifth confidence level, and the sixth confidence level do not exist, it means that the confidence level of the current carbon emission data is relatively high. Then, the sum of several confidence levels can be calculated, and the corresponding carbon emission report can be reviewed and processed from small to large according to the size of the sum of confidence levels. That is, the higher the sum of confidence levels, the lower the processing priority of the carbon emission report.

[0157] Furthermore, considering that the calibration frequency of the detection terminal can also affect the confidence of the overall report, for example, if the calibration record of the equipment cannot be found from the detection terminal and / or the calibration frequency is too low, the overall confidence of the carbon emission report should be reduced. It can be seen that the step of determining that the processing priority of the carbon emission electronic report is the second priority based on the calculated sum can specifically include:

[0158] When the calibration record of the detection equipment cannot be obtained, obtaining the first confidence adjustment coefficient m1;

[0159] When it is determined that the calibration frequency of the detection device is less than the first frequency threshold, obtaining a second confidence adjustment coefficient m2;

[0160] When it is determined that the calibration frequency of the detection device is greater than or equal to the first frequency threshold, obtaining a third confidence adjustment coefficient m3;

[0161] The calculated sum is multiplied by the obtained confidence adjustment coefficient to obtain the final confidence of the carbon emission electronic report, and the processing priority of the carbon emission electronic report is determined according to the final confidence;

[0162] Wherein, m1<m2<m3. In one embodiment, m3 is 1, m2 is 0.8, and m1 is 0.6.

[0163] It can be seen that by adopting the above solution to realize the audit of carbon emission data, the front-end collected data can be traced and audited and verified, so as to obtain the confidence level of the corresponding carbon emission data, and make corresponding annotations and display the corresponding modification strategies according to different confidence levels. This can greatly reduce the workload of auditors and improve work processing efficiency. In addition, the front-end data can be traced and verified and audited through image recognition algorithms, which can also ensure the accuracy of the data audit and verification.

[0164] Reference Figure 4 The present application also provides a carbon emission data audit system, which specifically includes a detection terminal and an audit terminal, and the detection terminal and the audit terminal are communicatively connected.

[0165] The detection terminal is mainly used to detect and collect data on the carbon emissions of vehicles, and to record the collected data.

[0166] Specifically, the detection terminal includes:

[0167] The first camera is used to photograph the area where the vehicle is traveling;

[0168] The first processing system includes at least one processor for executing the steps of the above-mentioned carbon emission data collection method and recording all data / parameters related to the detection.

[0169] The audit terminal includes at least one processor for executing the steps of the above-mentioned carbon emission data audit analysis and processing method.

[0170] The above-mentioned detection terminals are generally set up in enterprises or organizations that perform carbon emission collection and detection, while the audit terminals are set up in third-party authoritative institutions.

[0171] In view of the fact that the detection terminal and the audit terminal of the system embodiment each include at least one processor for executing the steps of the above method embodiment, the beneficial effects of the system of this embodiment are the same as the beneficial effects of the above method embodiment and will not be elaborated here.

[0172] In addition, the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the above method embodiment. The method herein includes a carbon emission data collection method and / or a carbon emission data audit method.

[0173] For the processors mentioned in the above storage medium embodiment and the system embodiment, the number can be at least 1, and at least any step in the above method embodiment can be executed. When the number is at least two, at least two processors can be connected to each other by communication, not limited to wired or wireless communication connection, and the at least one processor can be connected to various intelligent terminal devices by communication. In addition, the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0174] Finally, it should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0175] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. A carbon emission data audit method, applied to an audit terminal, characterized in that: The method includes: S1. Obtain the collection time period of carbon emissions recorded in the carbon emissions electronic report; S2. Select at least one time period to be reviewed from the different collection time periods obtained; S3, reading from the detection terminal according to the selected time period to be reviewed to obtain at least one corresponding data to be reviewed, wherein the data to be reviewed includes an image to be reviewed; S4. After performing image recognition on the image to be audited using the second carbon emission calculation method, an image recognition result is obtained, and the confidence level of the carbon emission data corresponding to the data to be audited is determined according to the image recognition result; S5. Use a labeling method corresponding to the confidence level to label and display the corresponding carbon emission data; The second carbon emission calculation method includes a first image recognition algorithm, a second image recognition algorithm and a third image recognition algorithm, and the step S4 specifically includes: S401, using a first image recognition algorithm to identify a vehicle model in the image to be reviewed, using a second image recognition algorithm to identify a stain area in the image to be reviewed, and using a third image recognition algorithm to identify a static target in the image to be reviewed; S402: when it is determined that the recognition results of the vehicle models are all correct and there is no stain area or static target in the image to be reviewed, the confidence level of the carbon emission data corresponding to the image to be reviewed is set to a first confidence level; S403: when it is determined that there is an incorrect recognition result of the vehicle model recognition result, and there is no stain area and static target in the image to be reviewed, a confidence confirmation process is performed on the incorrect recognition result; S404: when it is determined that there is a stain area and / or a static target in the image to be reviewed, the confidence level of the carbon emission data corresponding to the image to be reviewed is set to a second confidence level; Wherein, the first confidence level is greater than the second confidence level; The confidence confirmation process for the incorrect recognition result in step S403 specifically includes: S4031, obtaining the carbon emissions of a first vehicle corresponding to a historical vehicle target object, and recalculating the carbon emissions of a second vehicle corresponding to the historical vehicle target object after obtaining the corresponding carbon emission estimation model based on the newly identified vehicle model; S4032, calculating a first difference between the carbon emissions of the first vehicle and the carbon emissions of the second vehicle; S4033: when the first difference is less than the first threshold, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a third confidence level; S4034: when the first difference is greater than or equal to the first threshold and less than the second threshold, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a fourth confidence level, wherein the fourth confidence level is in inverse proportion to the first difference; S4035: when the first difference is greater than or equal to the second threshold, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a fifth confidence level; Wherein, the first confidence level is greater than or equal to the third confidence level, the third confidence level is greater than the maximum value of the fourth confidence level, and the minimum value of the fourth confidence level is greater than the fifth confidence level and the second confidence level; The second carbon emission calculation method further includes a fourth image recognition algorithm, and the step S4 further includes the following steps: S405, after using the fourth image recognition algorithm to recognize the vehicle target object in the image to be reviewed, when it is determined that the vehicle target object is recognized in the image to be reviewed but the license plate cannot be recognized, setting the confidence level of the carbon emission data corresponding to the image to be reviewed to a sixth confidence level; The minimum value of the fourth confidence level is greater than the sixth confidence level.

2. The method according to claim 1, characterized in that The method further comprises the following steps: S6. Determine the processing priority of the carbon emission electronic report based on the confidence level of the carbon emission data corresponding to the data to be reviewed.

3. The method according to claim 2, characterized in that The step S6 specifically includes: S601: When the confidence level of the carbon emission data in the carbon emission electronic report is at least one of the second confidence level, the fifth confidence level, and the sixth confidence level, the processing priority of the carbon emission electronic report is the first priority level; S602: When the confidence level of the carbon emission data in the carbon emission electronic report does not have the second confidence level, the fifth confidence level, and the sixth confidence level, the confidence levels corresponding to the plurality of carbon emission data in the carbon emission electronic report are summed up, and the processing priority of the carbon emission electronic report is determined to be the second priority level according to the calculated sum, and the second priority level is in inverse proportion to the sum; The first priority is higher than the second priority.

4. The method according to any one of claims 1 to 3, characterized in that: The step S5 specifically includes: S501, obtaining a recommended strategy corresponding to the confidence level; S502. Use a labeling method corresponding to the confidence level to label the corresponding carbon emission data differently, and set a form button at a position corresponding to the carbon emission data, wherein the form button is used to pop up a first window after being clicked, and the first window displays the corresponding confidence level and recommended strategy.

5. The method according to any one of claims 1 to 3, characterized in that: The step S3 specifically includes: S301, sending a query read request to the detection terminal, so that the detection terminal performs verification according to the ID of the audit terminal contained in the query read request, and after successful verification, querying and reading the pending audit data collected and recorded in the pending audit time period according to the pending audit time period contained in the query read request, and then encrypting the pending audit data read by the query using the first secret key, and returning the encrypted pending audit data to the audit terminal; S302: After receiving the encrypted data to be reviewed returned by the detection terminal, the encrypted data to be reviewed is decrypted using the second secret key.

6. The method according to any one of claims 1 to 3, characterized in that: The first image recognition algorithm is different from the image recognition algorithm recorded in the carbon emission electronic report, and the accuracy of the first image recognition algorithm is higher than the accuracy of the image recognition algorithm recorded in the carbon emission electronic report.

7. A carbon emission data audit system, characterized in that: The system includes: The detection terminal is used to detect and collect data on the carbon emissions of the vehicle and record the collected data; An audit terminal, comprising at least one processor, for executing the method according to any one of claims 1 to 6; The detection terminal is communicatively connected with the audit terminal.

Images