CN116958607B - Data processing method and device for target damage prediction - Google Patents
Data processing method and device for target damage prediction Download PDFInfo
- Publication number
- CN116958607B CN116958607B CN202311216311.7A CN202311216311A CN116958607B CN 116958607 B CN116958607 B CN 116958607B CN 202311216311 A CN202311216311 A CN 202311216311A CN 116958607 B CN116958607 B CN 116958607B
- Authority
- CN
- China
- Prior art keywords
- damage
- data
- target
- model
- sample
- 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.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 31
- 238000012549 training Methods 0.000 claims abstract description 101
- 238000000034 method Methods 0.000 claims abstract description 59
- 238000012545 processing Methods 0.000 claims abstract description 58
- 230000001066 destructive effect Effects 0.000 claims description 46
- 230000008569 process Effects 0.000 claims description 46
- 238000012360 testing method Methods 0.000 claims description 23
- 238000010276 construction Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 12
- 238000013136 deep learning model Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 3
- 230000003902 lesion Effects 0.000 claims description 3
- 238000009658 destructive testing Methods 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 13
- 238000004364 calculation method Methods 0.000 description 16
- 239000002360 explosive Substances 0.000 description 7
- 238000004880 explosion Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000013077 target material Substances 0.000 description 2
- 229910001294 Reinforcing steel Inorganic materials 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Classifications
-
- 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- 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
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The application discloses a data processing method and device for target damage prediction. The method comprises the steps of obtaining data to be processed, wherein the data to be processed comprises first damage data and second damage data, the first damage data are data used for representing a damage target, and the second damage data are data used for representing damage to the damage target; performing model matching processing based on a damaged target on the first damaged data to obtain a target damage prediction model, wherein the target damage prediction model corresponds to the damaged target; performing damage prediction processing on the second damage data based on the target damage prediction model to obtain target damage data, wherein the target damage data is data used for representing damage equivalent of the damaged target. The damage prediction model is obtained through training, and when the target damage effect is predicted, the target damage prediction is performed based on the damage prediction model obtained through training, so that the technical effect of improving the efficiency of target damage prediction is realized.
Description
Technical Field
The present application relates to the field of computers, and in particular, to a data processing method and apparatus for target damage prediction.
Background
The target damage effect prediction is an important basis for the commander to hit a decision, and how to accurately predict the damage of the target has important influence on the thermal resource integration and the thermal scheme optimization. For different targets, the corresponding damage calculation mechanisms are different, and for targets with complex damage calculation mechanisms, such as building targets, complex damage effect calculation is required, in the prior art, the damage calculation is carried out on the building targets by using a numerical calculation method, so that the calculation speed is low, and the temporary targets are difficult to calculate quickly.
Therefore, the target damage prediction in the prior art has the problem of low efficiency.
Disclosure of Invention
The main purpose of the application is to provide a data processing method and device for target damage prediction, so as to solve the technical problem of low efficiency in target damage prediction in the prior art, and achieve the technical effect of improving the target damage prediction efficiency.
To achieve the above object, a first aspect of the present application proposes a data processing method for target damage prediction, including:
obtaining data to be processed, wherein the data to be processed comprises first damage data and second damage data, the first damage data is data for representing a damage target, and the second damage data is data for representing damage to the damage target;
performing model matching processing based on a damaged target on the first damaged data to obtain a target damage prediction model, wherein the target damage prediction model corresponds to the damaged target;
performing damage prediction processing on the second damage data based on the target damage prediction model to obtain target damage data, wherein the target damage data is data for representing damage equivalent of a damaged target.
Further, performing a damage prediction process on the second damage data based on the target damage prediction model, to obtain target damage data includes:
performing identification processing based on damage characteristics on the second damage data to obtain damage characteristic data, wherein the damage characteristic data is characteristic data used for representing damage to a target;
performing graphical feature extraction processing on the damage feature data to obtain graphical damage feature data, wherein the graphical damage feature data is damage image feature data corresponding to a damage target;
and performing damage prediction processing on the graphic damage characteristic data to obtain the target damage data.
Further, performing model matching processing based on the damage target on the first damage data to obtain a target damage prediction model includes:
performing identification processing based on a damaged target on the first damaged data to obtain damaged target characteristic data, wherein the damaged target characteristic data is characteristic data for representing a damaged target;
and matching a damage prediction model corresponding to the damage target characteristic data in a preset damage model database to obtain the target damage prediction model.
Further, before acquiring the data to be processed, the data processing method includes:
obtaining damage training sample data, wherein the damage training sample data is sample data for training the target damage prediction model;
performing extraction processing based on a damaged target on the damaged training sample data to obtain target damaged training sample data, wherein the target damaged training sample data is model training sample data corresponding to the damaged target;
performing damage characteristic extraction processing on the target damage training sample data to obtain sample damage characteristic data, wherein the sample damage characteristic data is data for representing target damage characteristics;
matching sample damage image data corresponding to the sample damage characteristic data in the damage training sample data to obtain sample damage image data;
training a pre-constructed deep learning model according to the sample damage characteristic data, the sample damage image data and the sample damage data corresponding to the sample damage characteristic data to obtain a target damage prediction model.
Further, training a pre-constructed deep learning model according to the sample damage characteristic data, the sample damage image data and the sample damage data corresponding to the sample damage characteristic data, and obtaining a target damage prediction model comprises the following steps:
training a pre-constructed deep learning model according to the sample damage characteristic data, the sample damage image data and the sample damage data corresponding to the sample damage characteristic data to obtain a process damage prediction model;
performing test processing based on a test damage sample on the process damage prediction model to obtain process prediction accuracy;
judging the process prediction accuracy based on a preset accuracy threshold to judge whether the process damage prediction model meets a preset damage model training rule,
if the process prediction accuracy is greater than or equal to the preset accuracy threshold, obtaining the target damage prediction model, wherein the target damage prediction model is the process damage prediction model;
and if the process prediction accuracy is smaller than the preset accuracy threshold, performing iterative training on the process damage prediction model until the preset damage model training rule is met, and obtaining a target damage prediction model.
Further, obtaining the lesion training sample data comprises:
obtaining damage target data, wherein the damage target data is data for representing a damage target;
performing physical model construction processing on the damaged target data to obtain damaged target model data, wherein the damaged target model data is data of a physical model for representing a damaged target;
performing a destructive test construction process on the destructive target model data to obtain destructive training sample data, wherein the destructive training sample data is test data obtained by a destructive test of a physical model for representing the destructive target.
Further, performing a destructive test construction process on the destructive target model data to obtain destructive training sample data, wherein the destructive training sample data comprises:
performing a damage test construction process on the damage target model data to obtain first damage training sample data;
sample data expansion processing based on sample learning is carried out on the first damage training sample data, so that second damage training sample data is obtained;
determining the destructive training sample data according to the first destructive training sample data and the second destructive training sample data, wherein the destructive training sample data comprises the first destructive training sample data and the second destructive training sample data.
According to a second aspect of the present application, there is provided a data processing apparatus for target damage prediction, comprising:
the data acquisition module is used for acquiring data to be processed, wherein the data to be processed comprises first damage data and second damage data, the first damage data is data used for representing a damage target, and the second damage data is data used for representing damage to the damage target;
the model matching module is used for carrying out model matching processing based on a damaged target on the first damaged data to obtain a target damage prediction model, wherein the target damage prediction model corresponds to the damaged target;
the damage prediction module is used for performing damage prediction processing on the second damage data based on the target damage prediction model to obtain target damage data, wherein the target damage data is data used for representing damage equivalent of a damaged target.
According to a third aspect of the present application, the present application proposes a computer readable storage medium storing computer instructions for causing the computer to perform the above-mentioned data processing method for target damage prediction.
According to a fourth aspect of the present application, the present application proposes an electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to cause the at least one processor to perform the data processing method for target damage prediction described above.
The technical scheme provided by the embodiment of the application can comprise the following beneficial effects:
in the application, the data to be processed is obtained, wherein the data to be processed comprises first damage data and second damage data, the first damage data is data for representing a damage target, and the second damage data is data for representing damage to the damage target; performing model matching processing based on a damaged target on the first damaged data to obtain a target damage prediction model, wherein the target damage prediction model corresponds to the damaged target; performing damage prediction processing on the second damage data based on the target damage prediction model to obtain target damage data, wherein the target damage data is data for representing damage equivalent of a damaged target. The damage prediction model is obtained through training, and when the target damage effect is predicted, the target damage prediction is performed based on the damage prediction model obtained through training, and compared with the prior art that the damage calculation is realized through numerical calculation, the efficiency of target damage prediction is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, are included to provide a further understanding of the application and to provide a further understanding of the application with regard to the other features, objects and advantages of the application. The drawings of the illustrative embodiments of the present application and their descriptions are for the purpose of illustrating the present application and are not to be construed as unduly limiting the present application. In the drawings:
FIG. 1 is a flow chart of a data processing method for target damage prediction provided herein;
FIG. 2 is a flow chart of a data processing method for target damage prediction provided herein;
FIG. 3 is a flow chart of a data processing method for target damage prediction provided herein;
FIG. 4 is a schematic diagram of a data processing apparatus for target damage prediction according to the present application.
Detailed Description
In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the present application described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In the present application, the terms "upper", "lower", "left", "right", "front", "rear", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "lateral", "longitudinal" and the like indicate an azimuth or a positional relationship based on that shown in the drawings. These terms are used primarily to better describe the present application and its embodiments and are not intended to limit the indicated device, element or component to a particular orientation or to be constructed and operated in a particular orientation.
Also, some of the terms described above may be used to indicate other meanings in addition to orientation or positional relationships, for example, the term "upper" may also be used to indicate some sort of attachment or connection in some cases. The specific meaning of these terms in this application will be understood by those of ordinary skill in the art as appropriate.
Furthermore, the terms "mounted," "configured," "provided," "connected," "coupled," and "sleeved" are to be construed broadly. For example, "connected" may be in a fixed connection, a removable connection, or a unitary construction; may be a mechanical connection, or an electrical connection; may be directly connected, or indirectly connected through intervening media, or may be in internal communication between two devices, elements, or components. The specific meaning of the terms in this application will be understood by those of ordinary skill in the art as the case may be.
In the prior art, when the target damage effect is predicted, the method mainly adopts a numerical calculation method to perform fine calculation of the damage effect on the damaged target, but the method has the problems of low calculation speed, slow decision response caused by the fact that the temporary target and the newly-added target cannot be rapidly calculated, and therefore, the target damage effect prediction in the prior art has low efficiency.
Therefore, the application provides a data processing method for target damage prediction, a target damage training sample data set is established through a target damage test, a deep learning network is trained according to the target damage training sample data set to obtain a target damage prediction model, the target damage prediction model is used for rapidly calculating target damage of a damaged target, and the target damage prediction efficiency is improved.
In an alternative embodiment of the present application, a data processing method for target damage prediction is proposed, and fig. 1 is a flowchart of a data processing method for target damage prediction provided in the present application, and as shown in fig. 1, the method includes the following steps:
s101: acquiring data to be processed;
the data to be processed comprises first damage data and second damage data, wherein the first damage data is data for representing a damage target, and the second damage data is data for representing damage to the damage target;
the data to be processed is related data used for representing a damaged target, and comprises related data used for representing the target, such as a target type, a target size, target structure parameters and the like, and data corresponding to a target damage scheme used for carrying out target damage, such as weapon parameter data used for target damage, weapon drop point information and the like; the data to be processed can be divided into pre-damage data and post-damage data according to the target damage process, wherein the pre-damage data comprises data of a damage falling point, such as the falling point position, the speed, the intersection angle and the like, weapon information such as explosion power, warhead appearance parameters and the like, and data such as target materials, sizes, structural parameters and the like; the post-damage data comprise the post-damage target image, the position, the speed, the intersection angle and the like of the falling point, the weapon information such as the explosion power, the warhead appearance parameter and the like, and the target material, the size, the structural parameter and the like. When the acquired data to be processed is the pre-damage data, performing damage prediction processing on the pre-damage data according to the target damage prediction model degree, and predicting damage equivalent data suffered by the target; when the acquired data to be processed is the target post-damage data, performing damage evaluation processing on the post-damage data according to the target damage prediction model, and evaluating the damage equivalent data suffered by the current damaged target.
S102: performing model matching processing based on a damage target on the first damage data to obtain a target damage prediction model;
the target damage prediction model corresponds to a damage target; because the materials and the structure types corresponding to the different types of targets are different, when the damage prediction of the different types of damage targets is performed, the damage prediction model corresponding to the target type needs to be obtained according to the damage prediction model corresponding to the target type, for example, the current damage target is the first damage target, and the damage prediction model corresponding to the first damage target is matched in a preset damage prediction model database; the current target to be destroyed is a second destroyed target, and a destroy prediction model corresponding to the second destroyed target is matched in a preset destroy prediction model database to obtain a second target destroy prediction model.
In another alternative embodiment of the present application, there is provided a data processing method for target damage prediction, comprising: performing identification processing based on a damaged target on the first damaged data to obtain damaged target characteristic data, wherein the damaged target characteristic data is characteristic data for representing the damaged target; and matching a damage prediction model corresponding to the damage target characteristic data in a preset damage model database to obtain a target damage prediction model. By matching the damage prediction model corresponding to the targets, damage prediction is performed on multiple types of targets, and the efficiency of damage prediction on different types of targets is improved.
S103: performing damage prediction processing on the second damage data based on the target damage prediction model to obtain target damage data.
The target damage data is data representing the equivalent of damage to the damaged target.
In another alternative embodiment of the present application, a data processing method for target damage prediction is provided, and fig. 2 is a flowchart of a data processing method for target damage prediction as provided in the present application, and as shown in fig. 2, the method includes the following steps:
s201: performing identification processing based on the damage characteristics on the second damage data to obtain damage characteristic data;
the damage characteristic data is characteristic data used for representing damage to the target; the second damage data is data for representing the damage to the damaged target, the damage characteristics are divided into pre-damage characteristics and post-damage characteristics according to the damage process, the pre-damage characteristics are characteristics for representing the damage falling point, such as the falling point position, the speed, the intersection angle and the like, the post-damage characteristics are characteristics for representing the damage state of the target, such as the damage image of the target, and the damage equivalent data of the damaged target is obtained when the damage characteristic data is subjected to damage prediction processing.
S202: performing graphical feature extraction processing on the damage feature data to obtain graphical damage feature data;
the graph damage characteristic data is damage image characteristic data corresponding to a damage target; performing graphical feature extraction processing on the damage feature data, wherein the graphical feature extraction processing comprises the following steps: performing identification processing based on graphic features on the damage feature data, and if graphic damage feature data exists in the damage feature data, performing target damage prediction based on image identification according to the graphic damage feature data to obtain a prediction result; if the target damage characteristic data does not exist in the damage characteristic data, identifying damage point characteristics in the damage characteristic data, performing graphical characteristic extraction according to the damage characteristic data comprising the damage point characteristics, and generating predicted damage image data corresponding to the damage characteristic data to obtain graphical damage characteristic data.
S203: and performing damage prediction processing on the graphic damage characteristic data to obtain target damage data.
In another alternative embodiment of the present application, a data processing method for target damage prediction is provided, and fig. 3 is a flowchart of a data processing method for target damage prediction provided in the present application, as shown in fig. 3, and the method includes the following steps:
s301: acquiring data of a damage training sample;
the damage training sample data is sample data for training a target damage prediction model;
in another alternative embodiment of the present application, there is provided a data processing method for target damage prediction, comprising:
obtaining damage target data, wherein the damage target data is data for representing a damage target; performing physical model construction processing on the damaged target data to obtain damaged target model data, wherein the damaged target model data is the data of a physical model for representing a damaged target; performing destructive test construction processing on the destructive target model data to obtain destructive training sample data, wherein the destructive training sample data is test data obtained by subjecting a physical model for representing the destructive target to destructive test.
In an alternative embodiment of the application, the damage target is a building, a physical building model, an air model and an explosive model are constructed, the explosive model is used for carrying out building damage on the building, the air model does not participate in building damage, and the air model is used as a propagation carrier of shock waves generated by explosion. For example, a physical model of a building is constructed, the model size being 10m×20m×10m. The wall board has a thickness of 1.0m, four upright posts are arranged in the storehouse, the height of the four upright posts is 8.0m, each upright post contains 4 longitudinal reinforcing steel bars and 8 stirrups. The longitudinal ribs are 5m long, and the radius of the section is 0.01m. The radius of the single section of the stirrup is 0.01m; the explosive model is spherical explosive, the material is standard TNT, and the density is 1650kg/m 3 The device is used for simulating warhead charging of weapon equipment in experiments. Meanwhile, the size of the explosive model is not fixed, adjustment and improvement can be carried out according to the requirement of the explosive loading amount, so that calculation of other equivalent warheads is facilitated, and the damage effect of the warheads with the equivalent of 150kg on a target is simulated by adopting spherical charges with the radius of 0.13m and 0.17m as warhead charges.
The method comprises the steps that LS-DYNA software is adopted to complete grid division, concrete units of an ammunition library building model are selected to be regular hexahedral grids with preset side lengths to be subjected to division treatment, and grid division results meeting the requirements of test preset precision, grid preset size and calculation force are obtained through simulation; and for the air model, carrying out segmentation processing on grids of the air model to meet the construction requirement of a preset model, and obtaining damage target model data comprising a building model, the air model and an explosive model.
In another alternative embodiment of the present application, the present application provides a data processing method for target damage prediction, comprising:
performing a destructive test construction process on the destructive target model data to obtain first destructive training sample data; sample data expansion processing based on sample learning is carried out on the first damage training sample data, so that second damage training sample data is obtained; and determining the destructive training sample data according to the first destructive training sample data and the second destructive training sample data, wherein the destructive training sample data comprises the first destructive training sample data and the second destructive training sample data.
In an alternative embodiment of the present application, a sample learning expansion method based on deep learning is provided, including: sample expansion is carried out on sample data obtained by carrying out a damage test on a damage target physical model, and the sample expansion is realized by a small sample learning method, which comprises the following steps: filling small sample data with insufficient labels by using the pseudo data; decision boundaries learned by a sharpening classification algorithm using the dummy data to display. The small sample data is expanded through the pseudo data, the sample data obtained based on sample learning expansion and the damage test are mixed, the mixed data is used for carrying out model training on the damage prediction model, and the sample learning is adopted for carrying out data expansion, so that the problem of insufficient sample number or data missing is solved.
S302: performing extraction processing based on a damaged target on the damaged training sample data to obtain target damaged training sample data;
the target damage training sample data is model training sample data corresponding to the damage target; classifying the data obtained by performing a damage test on the same type of damage target in the training sample data, and constructing a damage prediction model by classifying the sample data of different types of damage targets so as to enable the sample data obtained by training to be suitable for damage prediction of multiple types of damage targets.
S303: performing damage characteristic extraction processing on target damage training sample data to obtain sample damage characteristic data;
the sample damage characteristic data is data for representing target damage characteristics; the target damage characteristics comprise target damage image characteristics and target damage parameter characteristics, the target damage image is used for feeding back the damage condition of the target, the target damage parameter characteristics have influence on the effect of target damage, the damage characteristic data in the training sample data are extracted, and the damage prediction model is trained according to the damage characteristic data.
S304: matching sample damage image data corresponding to the sample damage characteristic data in the damage training sample data to obtain sample damage image data;
when the damage training sample is obtained, the damage training sample comprises damage parameters, a damage target image after damage corresponds to the image after damage of the target, and the sample damage image and the damage parameters are corresponding to each other, so that the multidimensional characteristics affecting the damage effect are trained in the model training process, and the damage prediction model obtained by training has higher accuracy in the damage prediction process.
S305: training a pre-constructed deep learning model according to the sample damage characteristic data, the sample damage image data and the sample damage data corresponding to the sample damage characteristic data to obtain a target damage prediction model.
The sample damage characteristic data comprises damage equivalent data and damage falling point characteristic data, the sample damage image data and the damage falling point characteristic data are used as input of a pre-built deep learning model, the damage equivalent data received by a target are used as output of the pre-built deep learning model, and the deep learning model is trained to obtain a target damage prediction model.
In the embodiment of the application, in the process of performing network construction, a VGG16 network structure model in a deep learning model is adopted for performing experiments. The network training requires the advance setting of relevant training parameters. The process damage prediction model is obtained by carrying out prediction accuracy calculation processing based on a test sample, the process prediction accuracy is obtained, the process prediction accuracy is judged according to a preset accuracy rule, so as to judge whether the process damage prediction model meets a model training rule, and if the process prediction accuracy is greater than or equal to a preset accuracy threshold, the process damage prediction model meets a preset model training rule, so that the target damage prediction model is obtained; if the process prediction accuracy is smaller than the preset accuracy threshold, the process prediction damage model does not meet the preset model training rule, and iterative training is carried out on the process model until the process model meets the preset model training rule, so that the target damage prediction model is obtained.
In another alternative embodiment of the present application, there is provided a data processing apparatus for target damage prediction, and fig. 4 is a schematic diagram of a data processing apparatus for target damage prediction provided in the present application, as shown in fig. 4, and the method includes:
the data acquisition module 41 is configured to acquire data to be processed, where the data to be processed includes first damaged data and second damaged data, the first damaged data is data for representing a damaged target, and the second damaged data is data for representing a damaged target;
the model matching module 42 is configured to perform a model matching process based on a damaged target on the first damaged data to obtain a target damage prediction model, where the target damage prediction model corresponds to the damaged target;
the damage prediction module 43 is configured to perform a damage prediction process on the second damage data based on the target damage prediction model to obtain target damage data, where the target damage data is data representing a damage equivalent of the damaged target.
The specific manner in which the operations of the units in the above embodiments are performed has been described in detail in the embodiments related to the method, and will not be described in detail here.
In summary, in the present application, the data to be processed is obtained, where the data to be processed includes first damage data and second damage data, the first damage data is data for representing a damage target, and the second damage data is data for representing damage to the damage target; performing model matching processing based on a damaged target on the first damaged data to obtain a target damage prediction model, wherein the target damage prediction model corresponds to the damaged target; performing damage prediction processing on the second damage data based on the target damage prediction model to obtain target damage data, wherein the target damage data is data for representing damage equivalent of a damaged target. The damage prediction model is obtained through training, and when the target damage effect is predicted, the target damage prediction is performed based on the damage prediction model obtained through training, and compared with the prior art that the damage calculation is realized through numerical calculation, the efficiency of target damage prediction is improved.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
It will be apparent to those skilled in the art that the elements or steps of the application described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, or they may alternatively be implemented in program code executable by computing devices, such that they may be stored in a memory device for execution by the computing devices, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the same, but rather, various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.
Claims (9)
1. A data processing method for target damage prediction, comprising:
obtaining data to be processed, wherein the data to be processed comprises first damage data and second damage data, the first damage data is data for representing a damage target, and the second damage data is data for representing damage to the damage target;
performing model matching processing based on a damaged target on the first damaged data to obtain a target damage prediction model, wherein the target damage prediction model corresponds to the damaged target;
performing identification processing based on a damaged target on the first damaged data to obtain damaged target characteristic data, wherein the damaged target characteristic data is characteristic data for representing a damaged target;
matching a damage prediction model corresponding to the damage target characteristic data in a preset damage model database to obtain the target damage prediction model;
performing damage prediction processing on the second damage data based on the target damage prediction model to obtain target damage data, wherein the target damage data is data for representing damage equivalent of a damaged target.
2. The data processing method according to claim 1, wherein performing a damage prediction process based on the target damage prediction model on the second damage data to obtain target damage data includes:
performing identification processing based on damage characteristics on the second damage data to obtain damage characteristic data, wherein the damage characteristic data is characteristic data used for representing damage to a target;
performing graphical feature extraction processing on the damage feature data to obtain graphical damage feature data, wherein the graphical damage feature data is damage image feature data corresponding to a damage target;
and performing damage prediction processing on the graphic damage characteristic data to obtain the target damage data.
3. The data processing method according to claim 1, wherein before acquiring the data to be processed, the data processing method comprises:
obtaining damage training sample data, wherein the damage training sample data is sample data for training the target damage prediction model;
performing extraction processing based on a damaged target on the damaged training sample data to obtain target damaged training sample data, wherein the target damaged training sample data is model training sample data corresponding to the damaged target;
performing damage characteristic extraction processing on the target damage training sample data to obtain sample damage characteristic data, wherein the sample damage characteristic data is data for representing target damage characteristics;
matching sample damage image data corresponding to the sample damage characteristic data in the damage training sample data to obtain sample damage image data;
training a pre-constructed deep learning model according to the sample damage characteristic data, the sample damage image data and the sample damage data corresponding to the sample damage characteristic data to obtain a target damage prediction model.
4. The data processing method of claim 3, wherein training a pre-constructed deep learning model based on the sample damage feature data, the sample damage image data, and sample damage data corresponding to the sample damage feature data to obtain a target damage prediction model comprises:
training a pre-constructed deep learning model according to the sample damage characteristic data, the sample damage image data and the sample damage data corresponding to the sample damage characteristic data to obtain a process damage prediction model;
performing test processing based on a test damage sample on the process damage prediction model to obtain process prediction accuracy;
judging the process prediction accuracy based on a preset accuracy threshold to judge whether the process damage prediction model meets a preset damage model training rule,
if the process prediction accuracy is greater than or equal to the preset accuracy threshold, obtaining the target damage prediction model, wherein the target damage prediction model is the process damage prediction model;
and if the process prediction accuracy is smaller than the preset accuracy threshold, performing iterative training on the process damage prediction model until the preset damage model training rule is met, and obtaining a target damage prediction model.
5. A data processing method according to claim 3, wherein obtaining the destructive training sample data comprises:
obtaining damage target data, wherein the damage target data is data for representing a damage target;
performing physical model construction processing on the damaged target data to obtain damaged target model data, wherein the damaged target model data is data of a physical model for representing a damaged target;
performing a destructive test construction process on the destructive target model data to obtain destructive training sample data, wherein the destructive training sample data is test data obtained by a destructive test of a physical model for representing the destructive target.
6. The data processing method according to claim 5, wherein performing a destructive testing construction process on the destructive object model data to obtain destructive training sample data comprises:
performing a damage test construction process on the damage target model data to obtain first damage training sample data;
sample data expansion processing based on sample learning is carried out on the first damage training sample data, so that second damage training sample data is obtained;
determining the destructive training sample data according to the first destructive training sample data and the second destructive training sample data, wherein the destructive training sample data comprises the first destructive training sample data and the second destructive training sample data.
7. A data processing apparatus for target lesion prediction, comprising:
the data acquisition module is used for acquiring data to be processed, wherein the data to be processed comprises first damage data and second damage data, the first damage data is data used for representing a damage target, and the second damage data is data used for representing damage to the damage target;
the model matching module is used for carrying out model matching processing based on a damaged target on the first damaged data to obtain a target damage prediction model, wherein the target damage prediction model corresponds to the damaged target;
performing identification processing based on a damage target on the first damage data to obtain damage target characteristic data, wherein the damage target characteristic data is characteristic data for representing the damage target;
matching a damage prediction model corresponding to the damage target characteristic data in a preset damage model database to obtain the target damage prediction model;
the damage prediction module is used for performing damage prediction processing on the second damage data based on the target damage prediction model to obtain target damage data, wherein the target damage data is data used for representing damage equivalent of a damaged target.
8. A computer readable storage medium, characterized in that the computer readable storage medium stores computer instructions for causing the computer to execute the data processing method for target damage prediction according to any one of claims 1-6.
9. An electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor to cause the at least one processor to perform the data processing method for target lesion prediction according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311216311.7A CN116958607B (en) | 2023-09-20 | 2023-09-20 | Data processing method and device for target damage prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311216311.7A CN116958607B (en) | 2023-09-20 | 2023-09-20 | Data processing method and device for target damage prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116958607A CN116958607A (en) | 2023-10-27 |
| CN116958607B true CN116958607B (en) | 2023-12-22 |
Family
ID=88456853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311216311.7A Active CN116958607B (en) | 2023-09-20 | 2023-09-20 | Data processing method and device for target damage prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116958607B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020155518A1 (en) * | 2019-02-03 | 2020-08-06 | 平安科技（深圳）有限公司 | Object detection method and device, computer device and storage medium |
| CN114818471A (en) * | 2022-03-28 | 2022-07-29 | 北京航天飞腾装备技术有限责任公司 | Damage probability calculation method based on machine learning neural network |
| CN114841956A (en) * | 2022-04-29 | 2022-08-02 | 中国人民解放军军事科学院战争研究院 | Damage assessment method, system, equipment and storage medium based on image analysis |
| CN115310508A (en) * | 2022-05-25 | 2022-11-08 | 北京航天飞腾装备技术有限责任公司 | Damage probability calculation method based on machine learning classifier |
| CN115994465A (en) * | 2022-12-14 | 2023-04-21 | 中国人民解放军军事科学院国防工程研究院工程防护研究所 | Method for evaluating damage effect kNN of unexpected explosion air shock wave to bridge |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11131607B2 (en) * | 2019-05-02 | 2021-09-28 | Caterpillar Inc. | Modal analysis for damage rate measurement and prediction |
-
2023
- 2023-09-20 CN CN202311216311.7A patent/CN116958607B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020155518A1 (en) * | 2019-02-03 | 2020-08-06 | 平安科技（深圳）有限公司 | Object detection method and device, computer device and storage medium |
| CN114818471A (en) * | 2022-03-28 | 2022-07-29 | 北京航天飞腾装备技术有限责任公司 | Damage probability calculation method based on machine learning neural network |
| CN114841956A (en) * | 2022-04-29 | 2022-08-02 | 中国人民解放军军事科学院战争研究院 | Damage assessment method, system, equipment and storage medium based on image analysis |
| CN115310508A (en) * | 2022-05-25 | 2022-11-08 | 北京航天飞腾装备技术有限责任公司 | Damage probability calculation method based on machine learning classifier |
| CN115994465A (en) * | 2022-12-14 | 2023-04-21 | 中国人民解放军军事科学院国防工程研究院工程防护研究所 | Method for evaluating damage effect kNN of unexpected explosion air shock wave to bridge |
Non-Patent Citations (2)
| Title |
|---|
| 基于图像信息的打击效果评估模型研究;张选东等;舰船电子工程(第06期);全文 * |
| 目标毁伤效果预测研究;武健等;火力与指挥控制(第09期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116958607A (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7610184B1 (en) | Sector meshing and neighbor searching for object interaction simulation | |
| CN110222757A (en) | Based on insulator image pattern extending method, the system for generating confrontation network | |
| CN109977571B (en) | Simulation calculation method and device based on data and model mixing | |
| CN110457757A (en) | Instability of Rock Body stage forecast method and device based on multi-feature fusion | |
| CN115408925A (en) | Rock mass parameter prediction method and device for tunnel construction | |
| CN116958607B (en) | Data processing method and device for target damage prediction | |
| CN113343503A (en) | Blasting method and device for predicting blasting vibration damage | |
| CN112017730B (en) | Cell screening method and device based on expression quantity prediction model | |
| CN113656891A (en) | Liquid rocket dynamic characteristic modeling analysis method and terminal equipment | |
| CN112766381A (en) | Attribute-guided SAR image generation method under limited sample | |
| CN115795353B (en) | Underground metal target classification method and system based on unbalanced data set | |
| CN113196274A (en) | Computer-aided design of custom cellular lattice nuclei based on material properties | |
| CN106934729B (en) | Building detection and identification method and device | |
| CN114626267A (en) | Chip failure analysis method and device, electronic equipment and storage medium | |
| Štoller et al. | Design and assessment of shape of protective structure by usage of CFD software environment Ansys Fluent | |
| Yuan et al. | Enhanced Morris for the extraction of significant parameters in high-dimensional design optimization | |
| CN105488502A (en) | Target detection method and device | |
| US20110119292A1 (en) | Nearest neighbor search method | |
| CN113449460B (en) | Method and device for measuring fragment resistance coefficient and computer readable storage medium | |
| CN113837644B (en) | Equipment combat effectiveness and contribution rate integrated evaluation method based on grey correlation | |
| CN116310791B (en) | Rapid judgment method and electronic equipment for extremely disaster area based on building earthquake damage detection | |
| CN116957414B (en) | Village planning analysis method and device based on artificial intelligence | |
| Huang et al. | Predicting brittle fracture surface shape from a versatile database | |
| CN117765307A (en) | Knowledge distillation-based power transmission tower pole foreign matter hazard identification method and system | |
| CN114724083A (en) | Trajectory prediction system training method, trajectory prediction device and trajectory prediction system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |