CN116721316A - Model training and geomagnetic chart optimizing method, device, medium and equipment - Google Patents
Model training and geomagnetic chart optimizing method, device, medium and equipment Download PDFInfo
- Publication number
- CN116721316A CN116721316A CN202311010106.5A CN202311010106A CN116721316A CN 116721316 A CN116721316 A CN 116721316A CN 202311010106 A CN202311010106 A CN 202311010106A CN 116721316 A CN116721316 A CN 116721316A
- Authority
- CN
- China
- Prior art keywords
- geomagnetic
- geomagnetic chart
- chart
- target
- resolution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 82
- 238000012549 training Methods 0.000 title claims abstract description 81
- 238000005457 optimization Methods 0.000 claims abstract description 31
- 230000005358 geomagnetic field Effects 0.000 claims description 23
- 238000003860 storage Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 15
- 238000005259 measurement Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 12
- 230000006872 improvement Effects 0.000 description 11
- 238000009826 distribution Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000013526 transfer learning Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000010979 ruby Substances 0.000 description 1
- 229910001750 ruby Inorganic materials 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/04—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means
- G01C21/08—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means involving use of the magnetic field of the earth
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4046—Scaling of whole images or parts thereof, e.g. expanding or contracting using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/84—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using probabilistic graphical models from image or video features, e.g. Markov models or Bayesian networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Remote Sensing (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Radar, Positioning & Navigation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Automation & Control Theory (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Environmental & Geological Engineering (AREA)
- Probability & Statistics with Applications (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The specification discloses a model training and geomagnetic chart optimizing method, device, medium and equipment. The method comprises the following steps: acquiring an initial geomagnetic chart of a designated area; measuring magnetic field data of a designated area through a preset vector magnetometer, and generating a target geomagnetic chart based on the magnetic field data, wherein the resolution of the target geomagnetic chart is higher than that of an initial geomagnetic chart; inputting the initial geomagnetic chart as a training sample into a generation network in a to-be-trained generation model to generate a super-resolution geomagnetic chart of a designated area through the generation network; inputting the super-resolution geomagnetic chart and the target geomagnetic chart into a discrimination network in a generation model, so as to determine the probability that the super-resolution geomagnetic chart is the target geomagnetic chart through the discrimination network; and training the generated model by taking the minimized deviation between the super-resolution geomagnetic chart and the target geomagnetic chart and the minimized probability of judging the super-resolution geomagnetic chart as the target geomagnetic chart as an optimization target.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a method, an apparatus, a medium, and a device for model training and geomagnetic chart optimization.
Background
The geomagnetic navigation technology is a navigation mode with strong anti-interference capability, no accumulated error and moderate precision, wherein geomagnetic navigation requires geomagnetic graphs as reference graphs for geomagnetic matching, and the accuracy of the geomagnetic graphs determines the accuracy of geomagnetic navigation.
At present, the precision of the original geomagnetic chart is usually improved by adopting a spatial interpolation mode, namely, the spatial interpolation is carried out according to sparse geomagnetic data, and smooth contours are properly drawn.
Therefore, how to accurately improve the accuracy of the geomagnetic chart on the premise of guaranteeing the actual geomagnetic distribution situation and further guarantee the accuracy of geomagnetic navigation is a problem to be solved urgently.
Disclosure of Invention
The present disclosure provides a method, apparatus, medium and device for model training and geomagnetic chart optimization, so as to partially solve the foregoing problems in the prior art.
The technical scheme adopted in the specification is as follows:
the present specification provides a method of model training, comprising:
acquiring an initial geomagnetic chart of a designated area;
measuring magnetic field data of the designated area through a preset vector magnetometer, and generating a target geomagnetic chart based on the magnetic field data, wherein the resolution of the target geomagnetic chart is higher than that of the initial geomagnetic chart;
inputting the initial geomagnetic chart serving as a training sample into a generating network in a generating model to be trained so as to generate a super-resolution geomagnetic chart of the designated area through the generating network;
inputting the super-resolution geomagnetic chart and the target geomagnetic chart into a discrimination network in the generation model to determine the probability that the super-resolution geomagnetic chart is the target geomagnetic chart through the discrimination network;
and training the generated model by taking the minimized deviation between the super-resolution geomagnetic chart and the target geomagnetic chart and the minimized probability of judging the super-resolution geomagnetic chart as the target geomagnetic chart as an optimization target.
Optionally, acquiring an initial geomagnetic chart of the designated area specifically includes:
acquiring a geomagnetic field model corresponding to the designated area;
calculating magnetic field data corresponding to the designated area based on the geomagnetic field model;
and constructing the initial geomagnetic chart according to the calculated magnetic field data.
Optionally, training the generated model with a probability of minimizing a deviation between the super-resolution geomagnetic chart and the target geomagnetic chart and minimizing the super-resolution geomagnetic chart to be distinguished as the target geomagnetic chart as an optimization target specifically includes:
and training the generating network by taking the deviation between the minimized super-resolution geomagnetic chart and the target geomagnetic chart as an optimization target, and training the judging network by taking the probability of the minimized super-resolution geomagnetic chart being judged to be the target geomagnetic chart as the optimization target.
The specification provides a geomagnetic chart optimization method, which comprises the following steps:
acquiring an initial geomagnetic chart of a target area;
inputting the initial geomagnetic chart into a pre-trained generation model to generate a super-resolution geomagnetic chart corresponding to the target area through the generation model, wherein the resolution of the super-resolution geomagnetic chart is higher than that of the initial geomagnetic chart, and the generation model is obtained through training by the model training method.
Optionally, the method further comprises:
obtaining geographic information data of the target area;
according to the geographic information data, the super-resolution geomagnetic chart is adjusted, and adjusted geomagnetic chart data are obtained;
and executing tasks according to the adjusted geomagnetic chart data.
Optionally, the geographic information data includes: at least one of the topographic data and the topographic data of the target area.
Optionally, before acquiring the initial geomagnetic chart of the target area, the method further includes:
and deploying the generated model after training is completed, and deleting the discrimination network in the generated model.
The present specification provides a model training apparatus comprising:
the acquisition module acquires an initial geomagnetic chart of the designated area;
the measurement module is used for measuring magnetic field data of the designated area through a preset vector magnetometer and generating a target geomagnetic chart based on the magnetic field data, wherein the resolution of the target geomagnetic chart is higher than that of the initial geomagnetic chart;
the generation module is used for inputting the initial geomagnetic chart serving as a training sample into a generation network in a to-be-trained generation model so as to generate a super-resolution geomagnetic chart of the designated area through the generation network;
the judging module inputs the super-resolution geomagnetic chart and the target geomagnetic chart into a judging network in the generating model so as to determine the probability that the super-resolution geomagnetic chart is the target geomagnetic chart through the judging network;
and the training module is used for training the generated model by taking the probability of minimizing the deviation between the super-resolution geomagnetic chart and the target geomagnetic chart and the probability of minimizing the super-resolution geomagnetic chart as the target geomagnetic chart as an optimization target.
The present specification provides a computer readable storage medium storing a computer program which when executed by a processor implements the method of model training described above.
The present specification provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing a method of model training as described above when executing the program.
The above-mentioned at least one technical scheme that this specification adopted can reach following beneficial effect:
in the model training method provided by the specification, magnetic field data of a designated area is measured through a preset vector magnetometer, and a target geomagnetic chart is generated based on the magnetic field data, wherein the resolution of the target geomagnetic chart is higher than that of an initial geomagnetic chart; inputting the initial geomagnetic chart as a training sample into a generation network in a to-be-trained generation model to generate a super-resolution geomagnetic chart of a designated area through the generation network; inputting the super-resolution geomagnetic chart and the target geomagnetic chart into a discrimination network in a generation model, so as to determine the probability that the super-resolution geomagnetic chart is the target geomagnetic chart through the discrimination network; and training the generated model by taking the minimized deviation between the super-resolution geomagnetic chart and the target geomagnetic chart and the minimized probability of the super-resolution geomagnetic chart as the target geomagnetic chart as an optimization target.
According to the method, in the process of training the generated model, the countermeasure relation between the generated network and the discrimination network can be utilized, so that the super-resolution geomagnetic chart generated by the generated network gradually approaches to the real target geomagnetic chart along with the training of the model, the discrimination network can more accurately distinguish the super-resolution geomagnetic chart from the target geomagnetic chart along with the training of the model, and the accuracy and the authenticity of the super-resolution geomagnetic chart generated by the generated network are improved along with the improvement of the discrimination accuracy of the discrimination network, so that the discrimination network is deceived, and the accuracy of the geomagnetic chart is accurately improved on the premise of guaranteeing the distribution condition of the actual geomagnetism, and the accuracy of geomagnetic navigation is further ensured.
Drawings
The accompanying drawings, which are included to provide a further understanding of the specification, illustrate and explain the exemplary embodiments of the present specification and their description, are not intended to limit the specification unduly. In the drawings:
FIG. 1 is a schematic flow chart of a model training method provided in the present specification;
FIG. 2 is a schematic diagram of a training process for generating models provided in the present specification;
FIG. 3 is a schematic flow chart of a geomagnetic chart optimization method provided in the present specification;
FIG. 4 is a schematic diagram of a model training apparatus provided in the present specification;
fig. 5 is a schematic diagram of a geomagnetic chart generating apparatus provided in the present specification;
fig. 6 is a schematic view of an electronic device corresponding to fig. 1 or fig. 3 provided in the present specification.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the present specification more apparent, the technical solutions of the present specification will be clearly and completely described below with reference to specific embodiments of the present specification and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present specification. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are intended to be within the scope of the present disclosure.
The following describes in detail the technical solutions provided by the embodiments of the present specification with reference to the accompanying drawings.
Fig. 1 is a flow chart of a method for model training provided in the present specification, including the following steps:
s101: and acquiring an initial geomagnetic chart of the designated area.
S102: and measuring magnetic field data of the designated area through a preset vector magnetometer, and generating a target geomagnetic chart based on the magnetic field data, wherein the resolution of the target geomagnetic chart is higher than that of the initial geomagnetic chart.
At present, various spatial interpolation methods are mainly adopted for constructing the regional high-precision geomagnetic map, namely, spatial interpolation is carried out according to geomagnetic measurement point data, and contours of the optical bones are properly drawn within an error range, so that the geomagnetic map with higher precision is obtained. The spatial interpolation method includes a kriging interpolation method, a radial basis function interpolation method, an inverse distance weighted interpolation method, a nearest interpolation method, a linear interpolation method, a minimum curvature method, a polynomial fitting method and the like. However, spatial interpolation presets a specific geomagnetic field characteristic form, which limits the improvement of magnetometer accuracy to geomagnetic map accuracy.
Based on the above, the specification provides a model training method, which trains a countermeasure generation network through an initial geomagnetic chart and a target geomagnetic chart acquired by a magnetometer, and further improves the accuracy of the geomagnetic chart of a target area through the countermeasure generation network after training in the subsequent actual navigation process, so that the obtained super-resolution geomagnetic chart is ensured to conform to the actual geomagnetic distribution, and the accuracy of geomagnetic navigation is further ensured.
In the present specification, an execution subject for implementing a model training method may be a designated device such as a server, and for convenience of description, the present specification uses only a server as an execution subject, and describes a model training method provided in the present specification.
The server may first obtain an open source geomagnetic field model corresponding to the designated area, where the open source geomagnetic field model may include an international geomagnetic reference field (International Geomagnetic Reference Field, IGRF), a world geomagnetic field model (World Magnetic Model, WMM), and the like, which is not specifically limited in this specification.
The server may calculate the magnetic field data of the specified region based on the geomagnetic field model, the magnetic field data may be magnetic field intensity distribution data of the specified region, and then the server may construct a low resolution initial geomagnetic map based on the magnetic field data.
Further, the server can measure the real geomagnetic field data of the specified area through a high-precision preset vector magnetometer, and further construct a high-resolution target geomagnetic chart according to the real geomagnetic field data of the specified area.
Since the target geomagnetic chart is constructed by measuring magnetic field data with a high-precision vector magnetometer, the resolution (precision) of the target geomagnetic chart is higher than that of the initial geomagnetic chart.
S103: and inputting the initial geomagnetic chart serving as a training sample into a generating network in a generating model to be trained so as to generate the super-resolution geomagnetic chart of the designated area through the generating network.
S104: inputting the super-resolution geomagnetic chart and the target geomagnetic chart into a discrimination network in the generation model, so as to determine the probability that the super-resolution geomagnetic chart is the target geomagnetic chart through the discrimination network.
The server may select a plurality of regions from the open source geomagnetic field model as designated regions, and the size and the position of each designated region may be set according to actual situations, which is not specifically limited in this specification. In addition, for each designated area, the server may use the initial geomagnetic chart of the designated area as a training sample for training the generated model, and use the target geomagnetic chart of the designated area as a label of the training sample.
In the present specification, the generation model may be a stacked countermeasure generation network (Stacked Generative Adversarial Networks, SGAN) model in which a generation network and a discrimination network may be provided.
In the present specification, the Network structure of the generation Network may include a Deep Convolutional Neural Network (DCNN), a Residual Network (ResNet), and the like, and the Network structure of the discrimination Network may include a convolutional neural Network (Convolution Neural Network, CNN), a fully connected neural Network (Fully Connected Neural Network, FCNN), and the like, which are not particularly limited in the present specification.
Specifically, the server may input the initial geomagnetic map into a generating network of the generating model to be trained, and generate a super-resolution geomagnetic map through the generating network, where the resolution of the super-resolution geomagnetic map is higher than that of the initial geomagnetic map, so as to enhance the precision (resolution) of the initial geomagnetic map.
And then the server can input the super-resolution geomagnetic chart and the target geomagnetic chart into a judging network of the to-be-trained generation model, and the probability of judging the super-resolution geomagnetic chart as the target geomagnetic chart is determined through the judging network.
S105: and training the generation model by taking the probability that the super-resolution geomagnetic chart is minimized to be the target geomagnetic chart as an optimization target and taking the minimized deviation between the super-resolution geomagnetic chart and the target geomagnetic chart as the optimization target.
The server may train the generating network with the deviation between the super-resolution geomagnetic chart output by the minimizing generating network and the target geomagnetic chart as an optimization target, and train the discriminating network with the probability that the minimizing discriminating network discriminates the super-resolution geomagnetic chart as the target geomagnetic chart as the optimization target.
The purpose of optimizing the deviation between the super-resolution geomagnetic chart and the target geomagnetic chart output by the minimized generation network is to reduce the difference between the super-resolution geomagnetic chart and the target geomagnetic chart, so that the discrimination network is cheated (the probability of discriminating the super-resolution geomagnetic chart into the target geomagnetic chart is improved), and the generation network is trained. Therefore, the countermeasure relation between the generating network and the judging network can be utilized, the super-resolution geomagnetic chart generated by the generating network gradually approaches to the real target geomagnetic chart along with model training, so that the judging network is deceived, the judging network can more accurately distinguish the super-resolution geomagnetic chart from the target geomagnetic chart along with model training, and the reality and the accuracy of the generated super-resolution geomagnetic chart are improved through countermeasure.
Of course, the above-mentioned discrimination network may also output the discrimination result of the two classifications, that is, output whether the super-resolution geomagnetic chart is the target geomagnetic chart. The server can train the discrimination network by taking the deviation between the discrimination result output by the minimum discrimination network and the actual classification of the super-resolution geomagnetic chart and the target geomagnetic chart as an optimization target.
The server may implement optimization of the model through a loss function, where the loss function may include countering loss, perceived loss, content loss, and the like, and this description is not limited in detail. For ease of understanding, the present description provides a schematic diagram of a training process for generating a model, as shown in fig. 2.
Fig. 2 is a schematic diagram of a training process for generating a model provided in the present specification.
The server inputs the initial geomagnetic chart (LR) into a generating network to obtain a super-resolution geomagnetic chart (SR), then inputs the target geomagnetic chart (HR) and the super-resolution geomagnetic chart into a judging network, when the judging network judges the super-resolution geomagnetic chart as the target geomagnetic chart, the judging result is false, and otherwise, the judging result is true.
When the training target is met, the server can deploy the generated model to execute subsequent tasks through the deployed generated model. It should be noted that, the discriminating network in the generated model is only used in the model training process, and is not required in the actual application process, so the server can delete the discriminating network from the generated model after deploying the generated model.
In the present specification, the training targets may be various, for example, the generated model converges to a preset range, or reaches a preset training number, and the preset range and the preset training number may be set according to actual situations, which is not specifically limited in the present specification.
The foregoing describes a training process for generating a model from the viewpoint of model training, and the following describes a geomagnetic chart optimization method provided in the present specification from the viewpoint of practical application of generating a model with reference to the accompanying drawings.
Fig. 3 is a schematic flow chart of a geomagnetic chart optimization method provided in the specification, which includes the following steps:
s301: acquiring an initial geomagnetic chart of a target area;
s302: inputting the initial geomagnetic chart into a pre-trained generation model to optimize the initial geomagnetic chart through the generation model, and generating a super-resolution geomagnetic chart corresponding to the target area, wherein the resolution of the super-resolution geomagnetic chart is higher than that of the initial geomagnetic chart, and the generation model is obtained through training by the method.
In the geomagnetic map navigation process for the driving device, the server may obtain an initial geomagnetic map of a target area (for example, an area where the driving device is currently located) according to an open source geomagnetic map model, where the area of the target area may be set according to actual situations, which is not specifically limited in this specification.
The driving device may include an unmanned device (such as an unmanned plane, an unmanned vehicle, an intelligent robot, etc.), and of course, may also include a common manned device (such as an aircraft, a submarine, a ship, etc.).
The server can input the initial geomagnetic chart of the target area into the generation model, and the super-resolution geomagnetic chart of the target area is generated through a generation network of the generation model, and the super-resolution geomagnetic chart generated through the generation model is higher in precision than the initial geomagnetic chart because the generation model is trained in advance, so that the optimization of the initial geomagnetic chart is realized.
In addition, since the topographic data may provide information about the ground level, this may affect the intensity and distribution of the geomagnetic field. The topography data may provide information about ground characteristics, such as mountains, rivers, lakes, etc., which may also affect the characteristics of the geomagnetic field.
Therefore, the server can acquire the geographic information data of the target area, the geographic information data can comprise the topographic data, the geomorphic data and the like, and then the server can adjust the super-resolution geomagnetic chart output by the generation model according to the geographic information data to obtain adjusted geomagnetic chart data, so that the accuracy and the reliability of the geomagnetic chart are further improved. And executing tasks according to the adjusted map data. The task may include a geomagnetic map navigation task for the driving apparatus.
In the specification, the server can adapt to geomagnetic field changes in different areas and times through a model training mode of transfer learning or increment learning so as to improve the practicability of geomagnetic map generation.
Where the transfer learning allows the application of knowledge learned from one task to another task. Geomagnetic field data for different regions or for different time periods may have some shared geomagnetic field characteristics or laws, such as basic physical characteristics and change laws of the geomagnetic field. If there is already a model trained in a certain region or period, the server may use transfer learning to apply knowledge of this model to the geomagnetic field data of a new region or period without having to train a completely new model from scratch. This can save a significant amount of training time.
Incremental learning allows the model to train on new data while retaining prior knowledge. When new geomagnetic field data (e.g., data from different times in the same region) is available, we can use the new data to update and optimize our model without forgetting the previous training results. The method can make the model better adapt to the change of geomagnetic field, thereby improving the generation quality of geomagnetic pictures.
According to the method, in the process of training the generated model, the countermeasure relation between the generated network and the discrimination network can be utilized, so that the super-resolution geomagnetic chart generated by the generated network gradually approaches to the real target geomagnetic chart along with the training of the model, the discrimination network can more accurately distinguish the super-resolution geomagnetic chart from the target geomagnetic chart along with the training of the model, and the accuracy and the authenticity of the super-resolution geomagnetic chart generated by the generated network are improved along with the improvement of the discrimination accuracy of the discrimination network, so that the discrimination network is deceived, and the accuracy of the geomagnetic chart is accurately improved on the premise of guaranteeing the distribution condition of the actual geomagnetism, and the accuracy of geomagnetic navigation is further ensured.
The foregoing describes one or more methods for implementing model training in the present specification, and provides a corresponding model training apparatus based on the same concept, as shown in fig. 4.
Fig. 4 is a schematic diagram of a model training device provided in the present specification, including:
an obtaining module 401, configured to obtain an initial geomagnetic chart of a specified area;
a measurement module 402, configured to measure magnetic field data of the specified area by using a preset vector magnetometer, and generate a target geomagnetic chart based on the magnetic field data, where a resolution of the target geomagnetic chart is higher than a resolution of the initial geomagnetic chart;
a generating module 403, configured to input the initial geomagnetic chart as a training sample into a generating network in a to-be-trained generating model, so as to generate a super-resolution geomagnetic chart of the specified area through the generating network;
a discrimination module 404, configured to input the super-resolution geomagnetic chart and the target geomagnetic chart into a discrimination network in the generation model, so as to determine a probability that the super-resolution geomagnetic chart is the target geomagnetic chart through the discrimination network;
the training module 405 is configured to train the generated model with a probability of minimizing a deviation between the super-resolution geomagnetic chart and the target geomagnetic chart, and minimizing a discrimination of the super-resolution geomagnetic chart to the target geomagnetic chart as an optimization target.
Optionally, the obtaining module 401 is specifically configured to obtain a geomagnetic field model corresponding to the specified area; calculating magnetic field data corresponding to the designated area based on the geomagnetic field model; and constructing the initial geomagnetic chart according to the calculated magnetic field data.
Optionally, the training module 405 is specifically configured to train the generating network with a deviation between the minimized super-resolution geomagnetic chart and the target geomagnetic chart as an optimization target, and train the discriminating network with a probability of minimizing the discrimination of the super-resolution geomagnetic chart as the target geomagnetic chart as an optimization target.
The specification also provides a corresponding geomagnetic chart optimizing device, as shown in fig. 5.
Fig. 5 is a schematic diagram of a geomagnetic chart optimizing apparatus provided in the present specification, including:
an obtaining module 501, configured to obtain an initial geomagnetic chart of a target area;
the optimizing module 502 is configured to input the initial geomagnetic chart into a pre-trained generating model, so as to optimize the initial geomagnetic chart through the generating model, and generate a super-resolution geomagnetic chart corresponding to the target area, where the resolution of the super-resolution geomagnetic chart is higher than that of the initial geomagnetic chart, and the generating model is obtained through training by the method.
Optionally, the optimizing module 502 is further configured to obtain geographic information data of the target area;
according to the geographic information data, the super-resolution geomagnetic chart is adjusted, and adjusted geomagnetic chart data are obtained; and executing tasks according to the adjusted geomagnetic chart data.
Optionally, the geographic information data includes: at least one of the topographic data and the topographic data of the target area.
Optionally, before acquiring the initial geomagnetic chart of the target area, the apparatus further includes:
the deployment module 503 is configured to deploy the generated model after training is completed, and delete the discrimination network in the generated model.
The present specification also provides a computer readable storage medium storing a computer program operable to perform a model training and geomagnetic chart optimization method as provided by either of the above figures 1 or 3.
The present specification also provides a schematic structural diagram of an electronic device corresponding to fig. 1 or fig. 2 shown in fig. 6. At the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile storage, as illustrated in fig. 6, although other hardware required by other services may be included. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs to implement the model training and geomagnetic chart optimization method described above with respect to fig. 1 or fig. 2. Of course, other implementations, such as logic devices or combinations of hardware and software, are not excluded from the present description, that is, the execution subject of the following processing flows is not limited to each logic unit, but may be hardware or logic devices.
Improvements to one technology can clearly distinguish between improvements in hardware (e.g., improvements to circuit structures such as diodes, transistors, switches, etc.) and software (improvements to the process flow). However, with the development of technology, many improvements of the current method flows can be regarded as direct improvements of hardware circuit structures. Designers almost always obtain corresponding hardware circuit structures by programming improved method flows into hardware circuits. Therefore, an improvement of a method flow cannot be said to be realized by a hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (e.g., field programmable gate array (Field Programmable Gate Array, FPGA)) is an integrated circuit whose logic function is determined by the programming of the device by a user. A designer programs to "integrate" a digital system onto a PLD without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, such programming is mostly implemented by using "logic compiler" software, which is similar to the software compiler used in program development and writing, and the original code before the compiling is also written in a specific programming language, which is called hardware description language (Hardware Description Language, HDL), but not just one of the hdds, but a plurality of kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), lava, lola, myHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog are currently most commonly used. It will also be apparent to those skilled in the art that a hardware circuit implementing the logic method flow can be readily obtained by merely slightly programming the method flow into an integrated circuit using several of the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable logic controllers, and embedded microcontrollers, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic of the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller in a pure computer readable program code, it is well possible to implement the same functionality by logically programming the method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Such a controller may thus be regarded as a kind of hardware component, and means for performing various functions included therein may also be regarded as structures within the hardware component. Or even means for achieving the various functions may be regarded as either software modules implementing the methods or structures within hardware components.
The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being functionally divided into various units, respectively. Of course, the functions of each element may be implemented in one or more software and/or hardware elements when implemented in the present specification.
It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The present description is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
The foregoing is merely exemplary of the present disclosure and is not intended to limit the disclosure. Various modifications and alterations to this specification will become apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, or the like, which are within the spirit and principles of the present description, are intended to be included within the scope of the claims of the present description.
Claims (10)
1. A method of model training, comprising:
acquiring an initial geomagnetic chart of a designated area;
measuring magnetic field data of the designated area through a preset vector magnetometer, and generating a target geomagnetic chart based on the magnetic field data, wherein the resolution of the target geomagnetic chart is higher than that of the initial geomagnetic chart;
inputting the initial geomagnetic chart serving as a training sample into a generating network in a generating model to be trained so as to generate a super-resolution geomagnetic chart of the designated area through the generating network;
inputting the super-resolution geomagnetic chart and the target geomagnetic chart into a judging network in the generation model to determine the probability of judging the super-resolution geomagnetic chart as the target geomagnetic chart through the judging network;
and training the generated model by taking the minimized deviation between the super-resolution geomagnetic chart and the target geomagnetic chart and the minimized probability of judging the super-resolution geomagnetic chart as the target geomagnetic chart as an optimization target.
2. The method of claim 1, wherein obtaining an initial geomagnetic map of the designated area, specifically comprises:
acquiring a geomagnetic field model corresponding to the designated area;
calculating magnetic field data corresponding to the designated area based on the geomagnetic field model;
and constructing the initial geomagnetic chart according to the calculated magnetic field data.
3. The method of claim 1, wherein training the generation model with a minimized deviation between the super-resolution geomagnetic map and the target geomagnetic map and a minimized probability of discriminating the super-resolution geomagnetic map as the target geomagnetic map as an optimization target, specifically includes:
and training the generating network by taking the deviation between the minimized super-resolution geomagnetic chart and the target geomagnetic chart as an optimization target, and training the judging network by taking the probability of the minimized super-resolution geomagnetic chart being judged to be the target geomagnetic chart as the optimization target.
4. A geomagnetic chart optimization method, comprising:
acquiring an initial geomagnetic chart of a target area;
inputting the initial geomagnetic chart into a pre-trained generation model to optimize the initial geomagnetic chart through the generation model, and generating a super-resolution geomagnetic chart corresponding to the target area, wherein the resolution of the super-resolution geomagnetic chart is higher than that of the initial geomagnetic chart, and the generation model is obtained through training by the method of any one of claims 1-3.
5. The method of claim 4, wherein the method further comprises:
obtaining geographic information data of the target area;
according to the geographic information data, the super-resolution geomagnetic chart is adjusted, and adjusted geomagnetic chart data are obtained;
and executing tasks according to the adjusted geomagnetic chart data.
6. The method of claim 5, wherein the geographic information data comprises: at least one of the topographic data and the topographic data of the target area.
7. The method of claim 4, wherein prior to obtaining the initial geomagnetic map of the target area, the method further comprises:
and deploying the generated model after training is completed, and deleting the discrimination network in the generated model.
8. A model training device, comprising:
the acquisition module acquires an initial geomagnetic chart of the designated area;
the measurement module is used for measuring magnetic field data of the designated area through a preset vector magnetometer and generating a target geomagnetic chart based on the magnetic field data, wherein the resolution of the target geomagnetic chart is higher than that of the initial geomagnetic chart;
the generation module is used for inputting the initial geomagnetic chart serving as a training sample into a generation network in a to-be-trained generation model so as to generate a super-resolution geomagnetic chart of the designated area through the generation network;
the judging module inputs the super-resolution geomagnetic chart and the target geomagnetic chart into a judging network in the generating model so as to determine the probability that the super-resolution geomagnetic chart is the target geomagnetic chart through the judging network;
and the training module is used for training the generated model by taking the probability of minimizing the deviation between the super-resolution geomagnetic chart and the target geomagnetic chart and the probability of judging the super-resolution geomagnetic chart as the target geomagnetic chart as an optimization target.
9. A computer readable storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method of any of the preceding claims 1-7.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of the preceding claims 1-7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311010106.5A CN116721316A (en) | 2023-08-11 | 2023-08-11 | Model training and geomagnetic chart optimizing method, device, medium and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311010106.5A CN116721316A (en) | 2023-08-11 | 2023-08-11 | Model training and geomagnetic chart optimizing method, device, medium and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116721316A true CN116721316A (en) | 2023-09-08 |
Family
ID=87868407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311010106.5A Pending CN116721316A (en) | 2023-08-11 | 2023-08-11 | Model training and geomagnetic chart optimizing method, device, medium and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116721316A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104714257A (en) * | 2015-01-29 | 2015-06-17 | 哈尔滨工程大学 | Local geomagnetic map construction method based on multifractal kriging interpolation of gradual interpolation correction |
| US20180075581A1 (en) * | 2016-09-15 | 2018-03-15 | Twitter, Inc. | Super resolution using a generative adversarial network |
| CN109978762A (en) * | 2019-02-27 | 2019-07-05 | 南京信息工程大学 | A kind of super resolution ratio reconstruction method generating confrontation network based on condition |
| CN110610464A (en) * | 2019-08-15 | 2019-12-24 | 天津中科智能识别产业技术研究院有限公司 | Face image super-resolution method based on dense residual error neural network |
| CN112033406A (en) * | 2020-08-19 | 2020-12-04 | 五邑大学 | Navigation method, device and storage medium based on lightweight network |
| CN113359077A (en) * | 2021-06-08 | 2021-09-07 | 苏州深透智能科技有限公司 | Magnetic resonance imaging method and related equipment |
| CN114241078A (en) * | 2022-02-23 | 2022-03-25 | 南昌睿度医疗科技有限公司 | Accelerated magnetic resonance super-resolution imaging method, device, equipment and medium |
-
2023
- 2023-08-11 CN CN202311010106.5A patent/CN116721316A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104714257A (en) * | 2015-01-29 | 2015-06-17 | 哈尔滨工程大学 | Local geomagnetic map construction method based on multifractal kriging interpolation of gradual interpolation correction |
| US20180075581A1 (en) * | 2016-09-15 | 2018-03-15 | Twitter, Inc. | Super resolution using a generative adversarial network |
| CN109978762A (en) * | 2019-02-27 | 2019-07-05 | 南京信息工程大学 | A kind of super resolution ratio reconstruction method generating confrontation network based on condition |
| CN110610464A (en) * | 2019-08-15 | 2019-12-24 | 天津中科智能识别产业技术研究院有限公司 | Face image super-resolution method based on dense residual error neural network |
| CN112033406A (en) * | 2020-08-19 | 2020-12-04 | 五邑大学 | Navigation method, device and storage medium based on lightweight network |
| CN113359077A (en) * | 2021-06-08 | 2021-09-07 | 苏州深透智能科技有限公司 | Magnetic resonance imaging method and related equipment |
| CN114241078A (en) * | 2022-02-23 | 2022-03-25 | 南昌睿度医疗科技有限公司 | Accelerated magnetic resonance super-resolution imaging method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111639682B (en) | Ground segmentation method and device based on point cloud data | |
| CN111238450B (en) | Visual positioning method and device | |
| CN116740361B (en) | Point cloud segmentation method and device, storage medium and electronic equipment | |
| CN112036462B (en) | Model training and target detection method and device | |
| CN116977525B (en) | Image rendering method and device, storage medium and electronic equipment | |
| CN111797711A (en) | Model training method and device | |
| CN115600157B (en) | Data processing method and device, storage medium and electronic equipment | |
| CN116012483A (en) | Image rendering method and device, storage medium and electronic equipment | |
| CN116309823A (en) | Pose determining method, pose determining device, pose determining equipment and storage medium | |
| CN114494381A (en) | Model training and depth estimation method and device, storage medium and electronic equipment | |
| CN117237675A (en) | Unmanned aerial vehicle autonomous positioning method and device, storage medium and electronic equipment | |
| CN116167431B (en) | Service processing method and device based on hybrid precision model acceleration | |
| CN116403097A (en) | Target detection method and device, storage medium and electronic equipment | |
| CN117093862A (en) | Model training method and device, electronic equipment and storage medium | |
| CN116721316A (en) | Model training and geomagnetic chart optimizing method, device, medium and equipment | |
| CN116385999A (en) | Parking space identification method, device and equipment | |
| CN112393723B (en) | Positioning method, positioning device, medium and unmanned equipment | |
| CN117095371A (en) | Target detection method and detection device | |
| CN116558545A (en) | Calibration method and device for sensor data | |
| CN114283200B (en) | Pose determining method and device, storage medium and electronic equipment | |
| CN114972909B (en) | Model training method, map construction method and map construction device | |
| CN117876610B (en) | Model training method, device and storage medium for three-dimensional construction model | |
| CN118603079B (en) | Geomagnetic navigation method and device, storage medium and electronic equipment | |
| CN117576522B (en) | Model training method and device based on mimicry structure dynamic defense | |
| CN116740197B (en) | External parameter calibration method and device, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230908 |
|
| RJ01 | Rejection of invention patent application after publication |