Method, device and computer storage medium for correcting mutation sweep data

Abstract

The application provides a sudden sweep frequency data correction method, device, equipment and computer storage medium. The sudden sweep frequency data correction method comprises the following steps: obtaining a scene label of a road section corresponding to sudden sweep frequency data; obtaining historical sweep frequency data when the scene label meets a correction condition; and correcting the sudden sweep frequency data according to the historical sweep frequency data. The sudden sweep frequency data correction method, device, equipment and computer storage medium of the application first obtain a scene label of a road section corresponding to sudden sweep frequency data, obtain historical sweep frequency data when the scene label meets a correction condition, and then correct the sudden sweep frequency data according to the historical sweep frequency data, so that the abnormal sweep frequency data in the sweep frequency data is repaired.

Description

Technical Field

[0001] This application belongs to the field of data analysis, and in particular relates to a method, apparatus, device and computer storage medium for correcting mutation sweep frequency data. Background Technology

[0002] Currently, abnormal situations such as data loss, data mutation, and data transmission failure may occur during frequency sweep drive testing. Since there are no professional personnel to follow up with the test, it is impossible to accurately estimate what kind of problem caused the abnormality in a certain area, and the accuracy of the frequency sweep drive test results cannot be guaranteed. Summary of the Invention

[0003] This application provides a method, apparatus, device, and computer storage medium for correcting frequency sweep data during mutations, which can solve the technical problem of inaccurate frequency sweep test results in the prior art.

[0004] In a first aspect, embodiments of this application provide a method for correcting mutation sweep frequency data, the method comprising:

[0005] Obtain the scene labels for the corresponding road segments in the frequency sweep data of sudden changes;

[0006] When the scenario markers meet the correction conditions, historical frequency sweep data is acquired;

[0007] Correct the abrupt frequency sweep data based on historical frequency sweep data.

[0008] Further, in one embodiment, the scenario marker includes:

[0009] Context labels for mutation scanning data collected in non-tunnel environments, context labels for mutation scanning data collected in wide road environments, context labels for mutation scanning data collected in dispersed building environments, context labels for mutation scanning data collected in unobstructed environments, context labels for mutation scanning data collected when the ambient temperature change is less than a first threshold, and context labels for mutation scanning data collected when the ambient humidity change is less than a second threshold.

[0010] Furthermore, in one embodiment, the modification conditions include:

[0011] The scenario label is at least one of the following: scenario label for mutation scanning data collected in a non-tunnel environment, scenario label for mutation scanning data collected in a wide road environment, scenario label for mutation scanning data collected in a dispersed building environment, scenario label for mutation scanning data collected in an obstacle-free environment, scenario label for mutation scanning data collected when the ambient temperature change is less than a first threshold, and scenario label for mutation scanning data collected when the ambient humidity change is less than a second threshold.

[0012] Furthermore, in one embodiment, correcting abrupt frequency sweep data based on historical frequency sweep data includes:

[0013] The model is supplemented by training data based on historical frequency sweep data;

[0014] Input a preset number of historical frequency sweep data that are adjacent to and precede the mutation frequency sweep data into the data supplementation model, and output the results;

[0015] Replace the mutation sweep frequency data with the output results.

[0016] Secondly, embodiments of this application provide a mutation sweep frequency data correction device, the device comprising:

[0017] The acquisition module is used to acquire the scene markers of the road segments corresponding to the mutation frequency sweep data;

[0018] The acquisition module is also used to acquire historical frequency sweep data when the scenario marker meets the correction conditions;

[0019] The correction module is used to correct abrupt frequency sweep data based on historical frequency sweep data.

[0020] Further, in one embodiment, the scenario marker includes:

[0021] Context labels for mutation scanning data collected in non-tunnel environments, context labels for mutation scanning data collected in wide road environments, context labels for mutation scanning data collected in dispersed building environments, context labels for mutation scanning data collected in unobstructed environments, context labels for mutation scanning data collected when the ambient temperature change is less than a first threshold, and context labels for mutation scanning data collected when the ambient humidity change is less than a second threshold.

[0022] Furthermore, in one embodiment, the modification conditions include:

[0023] The scenario label is at least one of the following: scenario label for mutation scanning data collected in a non-tunnel environment, scenario label for mutation scanning data collected in a wide road environment, scenario label for mutation scanning data collected in a dispersed building environment, scenario label for mutation scanning data collected in an unobstructed environment, scenario label for mutation scanning data collected when the ambient temperature change is less than a first threshold, and scenario label for mutation scanning data collected when the ambient humidity change is less than a second threshold.

[0024] Furthermore, in one embodiment, the correction module specifically includes:

[0025] The model is supplemented by training data based on historical frequency sweep data;

[0026] Input a preset number of historical frequency sweep data that are adjacent to and precede the mutation frequency sweep data into the data supplementation model, and output the results;

[0027] Replace the mutation sweep frequency data with the output results.

[0028] Thirdly, embodiments of this application provide a mutation sweep frequency data correction device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements a mutation sweep frequency data correction method.

[0029] Fourthly, embodiments of this application provide a computer-readable storage medium storing an information transmission implementation program, which, when executed by a processor, implements a mutation sweep frequency data correction method.

[0030] The method, apparatus, device, and computer storage medium for correcting abrupt frequency sweep data in this application first obtain the scene markers of the road segments corresponding to the abrupt frequency sweep data. When it is determined that the scene markers meet the correction conditions, historical frequency sweep data is obtained, and then the abrupt frequency sweep data is corrected based on the historical frequency sweep data, thereby realizing the repair of abnormal frequency sweep data in the frequency sweep data. Attached Figure Description

[0031] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the structure of a frequency sweep drive test system provided in one embodiment of this application;

[0033] Figure 2 This is a flowchart illustrating a mutation sweep frequency data correction method provided in one embodiment of this application;

[0034] Figure 3 This is a schematic diagram of the structure of a data supplementation model provided in one embodiment of this application;

[0035] Figure 4 This is a schematic diagram of the structure of a mutation sweep frequency data correction device provided in one embodiment of this application;

[0036] Figure 5 This is a schematic diagram of the structure of a mutation sweep frequency data correction device provided in one embodiment of this application. Detailed Implementation

[0037] The features and exemplary embodiments of various aspects of this application will now be described in detail. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only configured to explain this application and are not configured to limit this application. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by indicating examples of this application.

[0038] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0039] Currently, during frequency sweep road testing, abnormal situations such as data loss, data mutation, and data transmission failure may occur. Since there are no professional personnel to follow the test, it is impossible to accurately estimate what kind of problem caused the abnormality in a certain area. This requires professional personnel to go to the problem area to investigate and troubleshoot, which greatly increases the workload of professional personnel and reduces the automation level of the equipment.

[0040] Existing road test equipment performs frequency scanning according to the planned working route, and then transmits the data to the analysis platform for result analysis. However, currently, no data preprocessing is performed before the collected data is analyzed.

[0041] Existing frequency sweep road test solutions lack a data preprocessing stage, which cannot effectively eliminate abnormal data and reduces the accuracy of test results.

[0042] To address the problems in the prior art, this application provides a method, apparatus, device, and computer storage medium for correcting abrupt frequency sweep data. This application first obtains the scene markers of the corresponding road segments in the abrupt frequency sweep data. When it is determined that the scene markers meet the correction conditions, historical frequency sweep data is obtained, and then the abrupt frequency sweep data is corrected based on the historical frequency sweep data, thus achieving the repair of abnormal frequency sweep data in the sweep data. The method for correcting abrupt frequency sweep data provided in this application is described below.

[0043] The main body executing this method is the frequency sweeping drive test system, which is described below. Compared with traditional frequency sweeping drive test equipment that only has a battery assembly, navigation and positioning unit, data communication unit and phased array antenna unit, the frequency sweeping drive test system of this application adds a temperature and humidity sensor module, a visual detection module and a visual detection module adjustment unit.

[0044] Figure 1 A schematic diagram of the system structure is shown, such as Figure 1 As shown, the system includes:

[0045] Battery assembly, navigation and positioning unit, frequency sweeper communication unit, phased array antenna unit, temperature and humidity sensor module, visual detection module, and visual detection module adjustment unit.

[0046] The system includes a battery module to power the frequency sweeping road test equipment; a navigation and positioning unit for path navigation; a data communication unit to transmit data to the analysis platform; a phased array antenna unit for signal reception and processing, and signal transmission via the frequency sweeping instrument's communication unit. A temperature and humidity sensor module acquires real-time temperature and humidity data from the equipment's operating environment; a visual detection module acquires real-time panoramic data from the equipment's operating environment; and a visual detection module adjustment unit adjusts the illumination compensation of the visual detection module.

[0047] Figure 2 A flowchart illustrating a mutation sweep frequency data correction method according to an embodiment of this application is shown. Figure 2 As shown, the method may include the following steps:

[0048] S102, Obtain the scene label of the road segment corresponding to the sudden change frequency sweep data.

[0049] The scene markers on the road segment reflect the scene type of the road segment.

[0050] For example, a scenario is labeled as mutation scan data collected in a non-tunnel environment, and the corresponding scenario type is that the road section does not have a tunnel.

[0051] The frequency sweeping road test system acquires real-time frequency sweeping data in the equipment's working road segment and generates all scenario tags for the equipment's working road segment based on the real-time frequency sweeping data. All scenario tags include scenario tags for road segments corresponding to sudden changes in frequency sweeping data.

[0052] In one embodiment, scenario tags can be generated through the following steps:

[0053] By identifying visual data in the frequency sweep data, scene tags are generated based on the surrounding vehicle situation, building distribution, and special terrain shown in the visual data.

[0054] In one embodiment, the scenario marker may include scenario markers for mutation scan data collected in a non-tunnel environment; scenario markers for mutation scan data collected in a wide road environment; scenario markers for mutation scan data collected in a dispersed building environment; scenario markers for mutation scan data collected in an obstacle-free environment; scenario markers for mutation scan data collected when the ambient temperature change is less than a first threshold; and scenario markers for mutation scan data collected when the ambient humidity change is less than a second threshold.

[0055] S104. When the scenario flag meets the correction conditions, acquire historical frequency sweep data.

[0056] In one embodiment, the correction condition includes:

[0057] The scenario label is at least one of the following: scenario label for mutation scanning data collected in a non-tunnel environment, scenario label for mutation scanning data collected in a wide road environment, scenario label for mutation scanning data collected in a dispersed building environment, scenario label for mutation scanning data collected in an obstacle-free environment, scenario label for mutation scanning data collected when the ambient temperature change is less than a first threshold, and scenario label for mutation scanning data collected when the ambient humidity change is less than a second threshold.

[0058] In particular, when the scenario marker is the scenario marker for mutation scanning data collected in a non-tunnel environment, the data loss at this time is not due to the influence of special environment, but is a non-realistic loss that cannot truly reflect the signal data of the test environment and requires data correction.

[0059] When the scenario is marked as the scenario marker for mutation scan data collected in a wide road environment, the data missing is due to temporary abnormal interference and cannot accurately reflect the actual situation. In this case, data filling and data compensation can be performed.

[0060] When the scenario is marked as a scenario for abrupt scanning data collected in a dispersed building environment, the data loss is not due to interference from surrounding buildings. The structure of the buildings does not have a significant impact on the signal and will not cause a large deviation in the data before and after. Therefore, the data cannot truly reflect the signal data of the test environment and data correction is required.

[0061] When the scenario is marked as the scenario marker for the mutation scan data collected in an unobstructed environment, the data loss is due to temporary abnormal interference or signal blind spots, which cannot accurately reflect the actual situation. In this case, data filling and data compensation can be performed based on the actual distance size corresponding to the data loss time.

[0062] For example, when the data loss distance is within 10m, the cause is temporary abnormal interference, and data correction is required; however, when the data loss distance is greater than 10m, the data loss is caused by signal blind spots, which truly reflect the signal data of the test environment, and no data correction is required.

[0063] When the scenario is marked as a scenario where the ambient temperature change is less than the first threshold and the collected sudden scan data is marked, there are no physical interference factors in the test environment. The signal strength changes from a certain moment and the ambient temperature before and after changes significantly. The signal strength change at this time is due to temperature change interference and data compensation is required based on the magnitude of the temperature change.

[0064] When the scenario is marked as a scenario where the change in ambient humidity is less than the first threshold and the collected abrupt scan data, there are no physical interference factors in the test environment. The signal strength changes from a certain moment, and the ambient humidity before and after changes significantly. The signal strength change at this time is due to the interference of humidity change, and data compensation is required based on the magnitude of the humidity change.

[0065] S106, correct the abrupt frequency sweep data based on historical frequency sweep data.

[0066] In one embodiment, S106 may include:

[0067] The model is trained based on historical frequency sweep data; a preset number of historical frequency sweep data that are adjacent to and precede the mutation frequency sweep data are input into the model to supplement the data, and the results are output; the output results are used to replace the mutation frequency sweep data, thereby realizing the repair of the mutation frequency sweep data.

[0068] The data supplementation model provided in this application is described below. Figure 3 A schematic diagram of the data supplementation model is shown, such as... Figure 3 As shown, the data supplementation model includes:

[0069] It consists of an LSTM layer, a Dropout layer, a Dense Layer, and an Output Layer.

[0070] At time t, the input to the model is the collected historical data sequence x. t (i.e., a preset number of historical frequency sweep data points that are adjacent to and precede the mutation frequency sweep data collection), the model output is the predicted value y for the next time step. t .

[0071] The specific calculation process is as follows:

[0072] x t Input to LSTM layer, output of LSTM layer

[0073] The LSTM layer output Apply the following formulas:

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] In the formula, i t f t c t o t The outputs of the input gate, forget gate, memory cell, and output gate are respectively; W ix W ih W ic The input information x is respectively t LSTM layer output at the previous time step And the output c of memory cells at the previous moment t-1 The weight matrix to the input gate; W fx W fm W fc The input information x are respectively t LSTM layer output at the previous time step And the output c of memory cells at the previous moment t-1 Weight matrix to the forget gate; W ox W om W oc The input information x are respectively t LSTM layer output at the previous time step And the output c of memory cells at the previous moment t-1 The weight matrix to the output gate; b f b o b c These are the biases for the input gate, output gate, forget gate, and memory cell, respectively; σ(·) is the sigmoid function, and h(·) is the activation function from input to output of the memory cell, usually taken as the hyperbolic tangent function tanh(·).

[0080] Input to Dropout layer, output of Dropout layer

[0081] Input Dense Layer, Dense Layer Output

[0082] in, In the formula, W p for The weight.

[0083] Input to Output Layer, Output Layer outputs y t .

[0084] in, In the formula, W yh is the weight matrix between the dense layer and the output layer; b is the bias of the output layer; g(·) is the activation function, usually the hyperbolic tangent function tanh(·).

[0085] At this point, the actual data y corresponding to the road segment corresponding to the abrupt change in frequency sweep data has been calculated. t , will y t Replacing the mutation sweep frequency data achieves the repair of the mutation sweep frequency data.

[0086] In practical applications, due to the limitation of historical data, N historical frequency sweep data points before the current time are generally used as inputs to the LSTM model for network training and frequency sweep data prediction.

[0087] To prevent overfitting and improve generalization ability, the output of the LSTM layer is processed by a Dropout layer, which can improve the robustness of the neural network.

[0088] The output of an LSTM layer is connected to a Dense Layer, and each neuron in the Dense Layer is connected to the outputs of all neurons in the previous layer. Through the Dense Layer, the LSTM output is multiplied by a matrix and a bias is added to produce the predicted data. The reason for adding a Dense Layer is that although the LSTM output contains the feature information needed for prediction, this feature information cannot fully reflect the relationship between the feature data and the prediction result. The Dense Layer allows us to understand the functional relationship between the feature data and the prediction result.

[0089] The method for correcting abrupt frequency sweep data in this application first obtains the scenario marker of the road segment corresponding to the abrupt frequency sweep data. When it is determined that the scenario marker meets the correction conditions, historical frequency sweep data is obtained, and then the abrupt frequency sweep data is corrected based on the historical frequency sweep data, thereby realizing the repair of abnormal frequency sweep data in the frequency sweep data.

[0090] Figure 1-3 The method for correcting mutation sweep frequency data is described below, in conjunction with the appendix. Figure 4 and attached Figure 5This application describes the apparatus provided in the embodiments.

[0091] Figure 4 This illustration shows a schematic diagram of a mutation sweep frequency correction device provided in one embodiment of this application. Figure 4 Each module in the device shown has the ability to implement Figure 2 The functions of each step in the process are defined, and their corresponding technical effects are achieved. For example... Figure 4 As shown, the device may include:

[0092] The acquisition module 400 is used to acquire the scene markers of the road segments corresponding to the sudden change frequency sweep data.

[0093] The scene markers on the road segment reflect the scene type of the road segment.

[0094] In one embodiment, this application uses a frequency sweeping drive test system to obtain real-time frequency sweeping data in the equipment's working road segment. Based on the real-time frequency sweeping data, all scenario tags for the equipment's working road segment can be generated, including scenario tags for road segments corresponding to sudden frequency sweeping data changes.

[0095] In one embodiment, scenario tags can be generated through the following steps:

[0096] By identifying visual data in the frequency sweep data, scene tags are generated based on the surrounding vehicle situation, building distribution, and special terrain shown in the visual data.

[0097] In one embodiment, the scenario marker may include:

[0098] Context marking for mutation scanning data collected in non-tunnel environments; context marking for mutation scanning data collected in wide road environments; context marking for mutation scanning data collected in dispersed building environments; context marking for mutation scanning data collected in unobstructed environments; context marking for mutation scanning data collected when the ambient temperature change is less than a first threshold; and context marking for mutation scanning data collected when the ambient humidity change is less than a second threshold.

[0099] The frequency sweep data mutations corresponding to the above-mentioned scenario types are not affected by special environments. In this case, the frequency sweep data changes are usually abnormal mutations and data repair is required.

[0100] The acquisition module 400 is also used to acquire historical frequency sweep data when the scenario marker meets the correction conditions.

[0101] Then, historical frequency sweep data is used to repair the abrupt frequency sweep data.

[0102] In one embodiment, the correction condition includes:

[0103] The scenario label is at least one of the following: scenario label for mutation scanning data collected in a non-tunnel environment, scenario label for mutation scanning data collected in a wide road environment, scenario label for mutation scanning data collected in a dispersed building environment, scenario label for mutation scanning data collected in an obstacle-free environment, scenario label for mutation scanning data collected when the ambient temperature change is less than a first threshold, and scenario label for mutation scanning data collected when the ambient humidity change is less than a second threshold.

[0104] The frequency sweep data mutations in the above-mentioned scenario types are not affected by special environments. In this case, the frequency sweep data mutations are usually abnormal mutations rather than normal mutations, and data repair is required.

[0105] In particular, when the scenario marker is the scenario marker for mutation scanning data collected in a non-tunnel environment, the data loss at this time is not due to the influence of special environment, but is a non-realistic loss that cannot truly reflect the signal data of the test environment and requires data correction.

[0106] When the scenario is marked as the scenario marker for mutation scan data collected in a wide road environment, the data missing is due to temporary abnormal interference and cannot accurately reflect the actual situation. In this case, data filling and data compensation can be performed.

[0107] When the scenario is marked as a scenario for abrupt scanning data collected in a dispersed building environment, the data loss is not due to interference from surrounding buildings. The structure of the buildings does not have a significant impact on the signal and will not cause a large deviation in the data before and after. Therefore, the data cannot truly reflect the signal data of the test environment and data correction is required.

[0108] When the scenario is marked as the scenario marker for the mutation scan data collected in an unobstructed environment, the data loss is due to temporary abnormal interference or signal blind spots, which cannot accurately reflect the actual situation. In this case, data filling and data compensation can be performed based on the actual distance size corresponding to the data loss time.

[0109] For example, when the data loss distance is within 10m, the cause is temporary abnormal interference, and data correction is required; however, when the data loss distance is greater than 10m, the data loss is caused by signal blind spots, which truly reflect the signal data of the test environment, and no data correction is required.

[0110] When the scenario is marked as a scenario where the ambient temperature change is less than the first threshold and the collected sudden scan data is marked, there are no physical interference factors in the test environment. The signal strength changes from a certain moment and the ambient temperature before and after changes significantly. The signal strength change at this time is due to temperature change interference and data compensation is required based on the magnitude of the temperature change.

[0111] When the scenario is marked as a scenario where the change in ambient humidity is less than the first threshold and the collected abrupt scan data, there are no physical interference factors in the test environment. The signal strength changes from a certain moment, and the ambient humidity before and after changes significantly. The signal strength change at this time is due to the interference of humidity change, and data compensation is required based on the magnitude of the humidity change.

[0112] The correction module 402 is used to correct abrupt frequency sweep data based on historical frequency sweep data.

[0113] In one embodiment, the correction module 402 may be specifically used for:

[0114] The model is trained based on historical frequency sweep data; a preset number of historical frequency sweep data that are adjacent to and precede the mutation frequency sweep data are input into the model to supplement the data, and the results are output; the output results are used to replace the mutation frequency sweep data.

[0115] The data supplementation model provided in this application is described below. Figure 3 A schematic diagram of the data supplementation model is shown, such as... Figure 3 As shown, the data supplementation model includes:

[0116] It consists of an LSTM layer, a Dropout layer, a Dense Layer, and an Output Layer.

[0117] At time t, the input to the model is the collected historical data sequence x. t (i.e., a preset number of historical frequency sweep data points that are adjacent to and precede the mutation frequency sweep data collection), the model output is the predicted value y for the next time step. t .

[0118] The specific calculation process is as follows:

[0119] x t Input to LSTM layer, output of LSTM layer

[0120] The LSTM layer output Apply the following formulas:

[0121]

[0122]

[0123]

[0124]

[0125]

[0126] In the formula, it f t c t o t The outputs of the input gate, forget gate, memory cell, and output gate are respectively; W ix W ih W ic The input information x are respectively t LSTM layer output at the previous time step And the output c of memory cells at the previous moment t-1 The weight matrix to the input gate; W fx W fm W fc The input information x are respectively t LSTM layer output at the previous time step And the output c of memory cells at the previous moment t-1 Weight matrix to the forget gate; W ox W om W oc The input information x are respectively t LSTM layer output at the previous time step And the output c of memory cells at the previous moment t-1 The weight matrix to the output gate; b f b o b c These are the biases for the input gate, output gate, forget gate, and memory cell, respectively; σ(·) is the sigmoid function, and h(·) is the activation function from input to output of the memory cell, usually taken as the hyperbolic tangent function tanh(·).

[0127] Input to Dropout layer, output of Dropout layer

[0128] Input Dense Layer, Dense Layer Output

[0129] in, In the formula, W p for The weight.

[0130] Input to Output Layer, Output Layer outputs y t .

[0131] in, In the formula, W yhis the weight matrix between the dense layer and the output layer; b is the bias of the output layer; g(·) is the activation function, usually the hyperbolic tangent function tanh(·).

[0132] At this point, the actual data y corresponding to the road segment corresponding to the abrupt change in frequency sweep data has been calculated. t , will y t Replacing the mutation sweep frequency data achieves the repair of the mutation sweep frequency data.

[0133] In practical applications, due to the limitation of historical data, N historical frequency sweep data points before the current time are generally used as inputs to the LSTM model for network training and frequency sweep data prediction.

[0134] To prevent overfitting and improve generalization ability, the output of the LSTM layer is processed by a Dropout layer, which can improve the robustness of the neural network.

[0135] The output of an LSTM layer is connected to a Dense Layer, and each neuron in the Dense Layer is connected to the outputs of all neurons in the previous layer. Through the Dense Layer, the LSTM output is multiplied by a matrix and a bias is added to produce the predicted data. The reason for adding a Dense Layer is that although the LSTM output contains the feature information needed for prediction, this feature information cannot fully reflect the relationship between the feature data and the prediction result. The Dense Layer allows us to understand the functional relationship between the feature data and the prediction result.

[0136] The method for correcting abrupt frequency sweep data in this application first obtains the scenario marker of the road segment corresponding to the abrupt frequency sweep data. When it is determined that the scenario marker meets the correction conditions, historical frequency sweep data is obtained, and then the abrupt frequency sweep data is corrected based on the historical frequency sweep data, thereby realizing the repair of abnormal frequency sweep data in the frequency sweep data.

[0137] The abrupt frequency sweep data correction device of this application first obtains the scene marker of the road segment corresponding to the abrupt frequency sweep data. When it is determined that the scene marker meets the correction conditions, it obtains historical frequency sweep data and then corrects the abrupt frequency sweep data based on the historical frequency sweep data, thereby realizing the repair of abnormal frequency sweep data in the frequency sweep data.

[0138] Figure 5 A schematic diagram of the structure of a mutation sweep frequency data correction device provided in one embodiment of this application is shown. Figure 5 As shown, the device may include a processor 501 and a memory 502 storing computer program instructions.

[0139] Specifically, the processor 501 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0140] Memory 502 may include mass storage for data or instructions. For example, and not limitingly, memory 502 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. In one instance, memory 502 may include removable or non-removable (or fixed) media, or memory 502 may be non-volatile solid-state storage. Memory 502 may be internal or external to the integrated gateway disaster recovery device.

[0141] In one instance, memory 502 may be read-only memory (ROM). In one instance, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0142] The processor 501 reads and executes computer program instructions stored in the memory 502 to achieve... Figure 2 The method / steps S102 to S106 in the illustrated embodiment are achieved. Figure 2 The technical effects achieved by executing the methods / steps shown in the examples are not elaborated here for the sake of brevity.

[0143] In one example, the mutation sweep frequency data correction device may further include a communication interface 503 and a bus 510. Wherein, as Figure 5 As shown, the processor 501, memory 502, and communication interface 503 are connected through bus 510 and complete communication with each other.

[0144] The communication interface 503 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0145] Bus 510 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 510 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0146] The mutation sweep frequency data correction device can execute the mutation sweep frequency data correction method in the embodiments of this application, thereby achieving... Figure 2 The effects of the described mutation sweep frequency data correction method.

[0147] Furthermore, in conjunction with the mutation sweep frequency data correction method in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the mutation sweep frequency data correction methods in the above embodiments.

[0148] It should be clarified that this application is not limited to the specific configurations and processes described above and indicated in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and indicated. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0149] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0150] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0151] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in 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, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0152] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for correcting frequency sweep data due to mutations, characterized in that, include: The scene labels of the road segments corresponding to the sudden change frequency sweep data are obtained. The scene labels are generated by recognizing the visual data in the frequency sweep data and marking them according to the surrounding vehicle situation, building distribution and special terrain shown by the visual data. When the scenario marker meets the correction conditions, historical frequency scanning data is acquired. The correction conditions include at least one of the following: scenario markers for abrupt scanning data collected in a non-tunnel environment, scenario markers for abrupt scanning data collected in a wide road environment, scenario markers for abrupt scanning data collected in a dispersed building environment, scenario markers for abrupt scanning data collected in an obstacle-free environment, scenario markers for abrupt scanning data collected when the ambient temperature change is less than a first threshold, and scenario markers for abrupt scanning data collected when the ambient humidity change is less than a second threshold. The abrupt frequency sweep data is corrected based on the historical frequency sweep data.

2. The mutation sweep frequency data correction method as described in claim 1, characterized in that, The scenario markers include: Context labels for mutation scanning data collected in non-tunnel environments, context labels for mutation scanning data collected in wide road environments, context labels for mutation scanning data collected in dispersed building environments, context labels for mutation scanning data collected in unobstructed environments, context labels for mutation scanning data collected when the ambient temperature change is less than a first threshold, and context labels for mutation scanning data collected when the ambient humidity change is less than a second threshold.

3. The mutation sweep frequency data correction method as described in claim 1, characterized in that, The step of correcting the abrupt frequency sweep data based on the historical frequency sweep data includes: The model is supplemented by training data based on the historical frequency sweep data; Input a preset number of historical frequency sweep data that are adjacent to and precede the mutation frequency sweep data acquisition into the data supplementation model, and output the result; Replace the mutation sweep frequency data with the output results.

4. A device for correcting abrupt frequency sweep data, characterized in that, The device includes: The acquisition module is used to acquire the scene markers of the road segments corresponding to the sudden change frequency sweep data. The scene markers are generated by recognizing the visual data in the frequency sweep data and marking them accordingly based on the surrounding vehicle situation, building distribution and special terrain shown by the visual data. The acquisition module is further configured to acquire historical frequency scanning data when the scenario marker meets the correction conditions, wherein the correction conditions include at least one of the following: scenario markers for abrupt scanning data collected in a non-tunnel environment, scenario markers for abrupt scanning data collected in a wide road environment, scenario markers for abrupt scanning data collected in a dispersed building environment, scenario markers for abrupt scanning data collected in an obstacle-free environment, scenario markers for abrupt scanning data collected when the ambient temperature change is less than a first threshold, and scenario markers for abrupt scanning data collected when the ambient humidity change is less than a second threshold; The correction module is used to correct the abrupt frequency sweep data based on the historical frequency sweep data.

5. The mutation sweep frequency data correction device as described in claim 4, characterized in that, The scenario markers include: Context labels for mutation scanning data collected in non-tunnel environments, context labels for mutation scanning data collected in wide road environments, context labels for mutation scanning data collected in dispersed building environments, context labels for mutation scanning data collected in unobstructed environments, context labels for mutation scanning data collected when the ambient temperature change is less than a first threshold, and context labels for mutation scanning data collected when the ambient humidity change is less than a second threshold.

6. The abrupt frequency sweep data correction device as described in claim 4, characterized in that, The correction module specifically includes: The model is supplemented by training data based on the historical frequency sweep data; Input a preset number of historical frequency sweep data that are adjacent to and precede the mutation frequency sweep data acquisition into the data supplementation model, and output the result; Replace the mutation sweep frequency data with the output results.

7. A mutation sweep frequency data correction device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the mutation sweep frequency data correction method as described in any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an implementation program for information transmission, which, when executed by a processor, implements the mutation sweep frequency data correction method as described in any one of claims 1 to 3.

Images