CN114120122B - Disaster damage identification method, device, equipment and storage medium based on remote sensing image - Google Patents
Disaster damage identification method, device, equipment and storage medium based on remote sensing image Download PDFInfo
- Publication number
- CN114120122B CN114120122B CN202111435237.9A CN202111435237A CN114120122B CN 114120122 B CN114120122 B CN 114120122B CN 202111435237 A CN202111435237 A CN 202111435237A CN 114120122 B CN114120122 B CN 114120122B
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- sensing image
- disaster damage
- value
- preset
- 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
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012549 training Methods 0.000 claims abstract description 117
- 238000009825 accumulation Methods 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 73
- 230000006835 compression Effects 0.000 claims description 14
- 238000007906 compression Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000011176 pooling Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 6
- 238000012217 deletion Methods 0.000 claims description 4
- 230000037430 deletion Effects 0.000 claims description 4
- 238000013519 translation Methods 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to an artificial intelligence technology, and discloses a disaster damage identification method based on remote sensing images, which comprises the following steps: inputting the training remote sensing image set with the enhanced data into a disaster damage recognition model to obtain a disaster damage degree type, counting the number of pixel points in the training remote sensing image under the disaster damage degree type, calculating a first distribution value and a second distribution value corresponding to the disaster damage degree, calculating preset loss weight values of the first distribution value and the second distribution value, carrying out weighted accumulation on the preset loss weight values and the loss values to obtain a final loss value, adjusting the disaster damage recognition model according to the final loss value to obtain a standard disaster damage recognition model, and inputting the remote sensing image to be recognized into the standard disaster damage recognition model to obtain a disaster damage recognition result. In addition, the invention also relates to a block chain technology, and the first distribution value can be stored in a node of the block chain. The invention further provides a disaster damage identification device based on the remote sensing image, electronic equipment and a storage medium. The invention can improve the disaster damage identification efficiency.
Description
Technical Field
The present invention relates to the field of artificial intelligence technologies, and in particular, to a method and apparatus for identifying disaster damage based on remote sensing images, an electronic device, and a computer readable storage medium.
Background
Various natural disasters occur more and more frequently in the world, and the natural disasters occurring each year can bring serious threat to national economy and people economy property safety of various countries. Therefore, the analysis is needed according to different areas where disasters occur each time and the corresponding disaster loss degree, so that the follow-up monitoring and prevention are convenient. The existing disaster damage identification method is usually used for judging the loss condition of a target according to the field investigation of a salesman, is not intelligent enough, consumes labor, and is not high in accuracy and efficiency for disaster damage identification. Therefore, a more efficient disaster damage identification method is needed.
Disclosure of Invention
The invention provides a disaster damage identification method and device based on remote sensing images and a computer readable storage medium, and mainly aims to improve the disaster damage identification efficiency.
In order to achieve the above object, the present invention provides a disaster damage identification method based on remote sensing images, including:
acquiring an original remote sensing image set, and carrying out data enhancement on the original remote sensing image set to obtain a training remote sensing image set;
inputting the training remote sensing image set into a preset disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set;
counting the number of pixel points corresponding to the training remote sensing image set under the disaster damage degree category, and calculating a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of the pixel points;
Calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method;
Respectively calculating loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and carrying out weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain final loss values;
The disaster damage recognition model is adjusted according to the size between the final loss value and a preset loss threshold value, and a standard disaster damage recognition model is obtained;
and acquiring a remote sensing image to be identified, and inputting the remote sensing image to be identified into the standard disaster damage identification model to obtain a disaster damage identification result corresponding to the remote sensing image to be identified.
Optionally, the inputting the training remote sensing image set into a preset disaster damage recognition model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set includes:
performing convolution processing and pooling processing on the training remote sensing image set by utilizing a compression channel in the disaster damage identification model to obtain an initial pooled image set;
Performing deconvolution operation on the initial pool image set to obtain a deconvolution image set;
performing image stitching on the initial pool image set and the deconvolution image set, and performing feature extraction on the stitched image set to obtain a feature image set;
and inputting the characteristic image set to an output layer in the disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set.
Optionally, the counting the number of pixels corresponding to the training remote sensing image set under the disaster damage degree category includes:
determining the areas of disaster damage degree categories corresponding to different training remote sensing images in the training remote sensing image set;
And identifying and summarizing the pixel points in the areas with different disaster damage degree categories to obtain the number of the pixel points corresponding to the different disaster damage degree categories.
Optionally, the calculating, based on the number of the pixel points, a first distribution value and a second distribution value corresponding to the disaster damage degree includes:
Utilizing a preset first distribution value formula and a preset second distribution value formula;
The number of the pixel points and the number of categories of the disaster damage degree obtained in advance are used as the input of the first step value formula, and a first distribution value is obtained;
and taking the number of categories of the disaster damage degree as the input of the second distribution value formula to obtain a second distribution value.
Optionally, the preset first distribution value formula is:
Wherein s1 is the first distribution value, w i represents the number of the i-th type disaster damage degree pixel points, and k represents the number of the types of disaster damage degree.
Optionally, the calculating, by using a preset weight assignment method, a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value includes:
Multiplying the second distribution value with a preset first reference value to obtain a first standard value, and determining the preset loss weight value as a first combined value if the first distribution value is greater than or equal to the first standard value and less than or equal to the second distribution value; or (b)
Multiplying the second distribution value with a preset second reference value to obtain a second standard value, and determining that the preset loss weight value is a second combined value if the first distribution value is greater than or equal to the second standard value and less than or equal to the first standard value; or (b)
Multiplying the second distribution value with a preset third reference value to obtain a third standard value, and determining the preset loss weight value as a third combined value if the first distribution value is greater than or equal to the third standard value and less than or equal to the second standard value; or (b)
And if the first distribution value is larger than or equal to a preset fourth standard value and smaller than or equal to the third standard value, determining that the preset loss weight value is a fourth combined value.
Optionally, the performing data enhancement on the original remote sensing image set to obtain a training remote sensing image set includes:
detecting a missing remote sensing image in the original remote sensing image set, and executing deletion operation on the missing remote sensing image to obtain an original remote sensing image set;
and performing image rotation, image translation and image scaling on the initial remote sensing images in the initial remote sensing image set to obtain a training remote sensing image set.
In order to solve the above problems, the present invention further provides a disaster damage recognition device based on remote sensing images, the device includes:
the data enhancement module is used for acquiring an original remote sensing image set, and carrying out data enhancement on the original remote sensing image set to obtain a training remote sensing image set;
The category prediction module is used for inputting the training remote sensing image set into a preset disaster damage recognition model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set;
The model training module is used for counting the number of pixel points corresponding to the training remote sensing image set under the disaster damage degree category, calculating a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of the pixel points, calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method, respectively calculating loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and carrying out weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain a final loss value, and adjusting the disaster damage identification model according to the size between the final loss value and a preset loss threshold value to obtain a standard disaster damage identification model;
the disaster damage recognition module is used for acquiring a remote sensing image to be recognized, inputting the remote sensing image to be recognized into the standard disaster damage recognition model, and obtaining a disaster damage recognition result corresponding to the remote sensing image to be recognized.
In order to solve the above-mentioned problems, the present invention also provides an electronic apparatus including:
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 enable the at least one processor to perform the remote sensing image-based disaster identification method described above.
In order to solve the above-mentioned problems, the present invention further provides a computer readable storage medium, where at least one computer program is stored, where the at least one computer program is executed by a processor in an electronic device to implement the above-mentioned disaster damage identification method based on remote sensing images.
According to the embodiment of the invention, the obtained original remote sensing image set is subjected to data enhancement, so that the richness of the data of the obtained training remote sensing image set can be improved, and the robustness of subsequent model training is enhanced. The disaster degree category corresponding to the training remote sensing image set is identified by utilizing a disaster degree identification model, a first distribution value and a second distribution value corresponding to the disaster degree are calculated based on the number of pixels corresponding to the training remote sensing image set under the disaster degree category, the first distribution value and the second distribution value reflect the distribution condition of the disaster degree, the accuracy of a loss function and the accuracy of model training can be improved by calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value and constructing a loss function according to the preset loss weight values, the disaster recognition model is adjusted according to the size between the final loss value and a preset loss threshold value, a standard disaster recognition model is obtained, the remote sensing image to be recognized is recognized, and a disaster recognition result corresponding to the remote sensing image to be recognized is obtained. Therefore, the disaster damage identification method, the device, the electronic equipment and the computer readable storage medium based on the remote sensing image can solve the problem that the disaster damage identification efficiency is not high enough.
Drawings
FIG. 1 is a schematic flow chart of a disaster damage identification method based on remote sensing images according to an embodiment of the invention;
FIG. 2 is a functional block diagram of a disaster damage recognition device based on remote sensing images according to an embodiment of the present invention;
Fig. 3 is a schematic structural diagram of an electronic device for implementing the disaster damage identification method based on remote sensing images according to an embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The embodiment of the application provides a disaster damage identification method based on remote sensing images. The execution subject of the disaster damage identification method based on the remote sensing image comprises at least one of electronic equipment, such as a server, a terminal and the like, which can be configured to execute the method provided by the embodiment of the application. In other words, the disaster damage identification method based on the remote sensing image can be executed by software or hardware installed in the terminal equipment or the server equipment, and the software can be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (ContentDelivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.
Referring to fig. 1, a flow chart of a disaster damage identification method based on remote sensing images according to an embodiment of the invention is shown. In this embodiment, the disaster damage identification method based on remote sensing image includes:
S1, acquiring an original remote sensing image set, and carrying out data enhancement on the original remote sensing image set to obtain a training remote sensing image set.
In the embodiment of the present invention, the original Remote sensing image set includes a large number of original Remote sensing images, where the original Remote sensing image (Remote SENSING IMAGE) is a film or a photo for recording electromagnetic wave sizes of various ground objects, and is mainly divided into an aerial photo and a satellite photo. The 16 m resolution remote sensing image of the high-resolution one-number multispectral camera WFV can be obtained from the state-run assets source satellite application center.
Specifically, the performing data enhancement on the original remote sensing image set to obtain a training remote sensing image set includes:
detecting a missing remote sensing image in the original remote sensing image set, and executing deletion operation on the missing remote sensing image to obtain an original remote sensing image set;
and performing image rotation, image translation and image scaling on the initial remote sensing images in the initial remote sensing image set to obtain a training remote sensing image set.
In detail, the missing remote sensing image in the original remote sensing image set may be an image with serious missing image information.
The original remote sensing image set is subjected to data enhancement to obtain a training remote sensing image set, so that the number of images in the training remote sensing image set is enriched, and the robustness of subsequent model training is improved.
S2, inputting the training remote sensing image set into a preset disaster damage recognition model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set.
In the embodiment of the invention, the disaster damage identification model can be networks such as res-net, U-net, deeplab-v3 and the like. In the scheme, the disaster damage identification model is a U-net network. The U-net network is formed by a left compression channel and a right expansion channel to form the shape of a letter U. Wherein the compression channel is a typical convolutional neural network structure.
Specifically, the inputting the training remote sensing image set into a preset disaster damage recognition model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set includes:
performing convolution processing and pooling processing on the training remote sensing image set by utilizing a compression channel in the disaster damage identification model to obtain an initial pooled image set;
Performing deconvolution operation on the initial pool image set to obtain a deconvolution image set;
performing image stitching on the initial pool image set and the deconvolution image set, and performing feature extraction on the stitched image set to obtain a feature image set;
and inputting the characteristic image set to an output layer in the disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set.
In detail, the disaster recognition model is composed of a left compression channel (Contracting Path) and a right expansion channel (Expansive Path). The compression channel is a typical convolutional neural network structure, which repeatedly adopts a structure of 2 convolutional layers and 1 maximum pooling layer, and the dimension of the characteristic diagram is increased by 1 time after each pooling operation. And in the expansion channel, firstly, carrying out deconvolution operation for 1 time to halve the dimension of the feature map, then splicing the feature map obtained by cutting the corresponding compression channel, reconstructing a feature map with the size of 2 times, carrying out feature extraction by adopting 2 convolution layers, and repeating the structure. At the final output layer, the 64-dimensional feature map is mapped into a 2-dimensional output map with 2 convolutional layers. The output graph comprises disaster damage degree categories.
The disaster damage degree category refers to the severity degree of disaster damage and can be divided into primary disaster damage, secondary disaster damage, tertiary disaster damage and the like.
And S3, counting the number of pixels corresponding to the training remote sensing image set under the disaster damage degree category, and calculating a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of pixels.
In the embodiment of the present invention, the counting the number of pixels corresponding to the training remote sensing image set in the disaster damage degree category includes:
determining the areas of disaster damage degree categories corresponding to different training remote sensing images in the training remote sensing image set;
And identifying and summarizing the pixel points in the areas with different disaster damage degree categories to obtain the number of the pixel points corresponding to the different disaster damage degree categories.
In detail, the disaster degree category corresponding to the training remote sensing image comprises primary disaster damage, secondary disaster damage or tertiary disaster damage marked on the remote sensing image, the areas where the primary disaster damage, the secondary disaster damage and the tertiary disaster damage are located are respectively determined, and the number of pixels in the areas where different disaster damage are located is identified and counted.
Specifically, the calculating, based on the number of the pixel points, a first distribution value and a second distribution value corresponding to the disaster degree includes:
Utilizing a preset first distribution value formula and a preset second distribution value formula;
The number of the pixel points and the number of categories of the disaster damage degree obtained in advance are used as the input of the first step value formula, and a first distribution value is obtained;
and taking the number of categories of the disaster damage degree as the input of the second distribution value formula to obtain a second distribution value.
Further, the preset first distribution value formula is:
Wherein s1 is the first distribution value, w i represents the number of the i-th type disaster damage degree pixel points, and k represents the number of the types of disaster damage degree.
Specifically, the preset second distribution value formula is:
Wherein s2 is the second distribution value, and k represents the number of categories of the disaster damage degree.
S4, calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method.
In the embodiment of the present invention, the calculating, by using a preset weight assignment method, a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value includes:
Multiplying the second distribution value with a preset first reference value to obtain a first standard value, and determining the preset loss weight value as a first combined value if the first distribution value is greater than or equal to the first standard value and less than or equal to the second distribution value; or (b)
Multiplying the second distribution value with a preset second reference value to obtain a second standard value, and determining that the preset loss weight value is a second combined value if the first distribution value is greater than or equal to the second standard value and less than or equal to the first standard value; or (b)
Multiplying the second distribution value with a preset third reference value to obtain a third standard value, and determining the preset loss weight value as a third combined value if the first distribution value is greater than or equal to the third standard value and less than or equal to the second standard value; or (b)
And if the first distribution value is larger than or equal to a preset fourth standard value and smaller than or equal to the third standard value, determining that the preset loss weight value is a fourth combined value.
Preferably, the first reference value is 0.75, the second reference value is 0.5, the third reference value is 0.25, the fourth reference value is 0, the first combined value is (0.15,0.35,0.15,0.35), the second combined value is (0.25,0.25,0.25,0.25), the third combined value is (0.35,0.15,0.35,0.15), and the fourth combined value is (0.5,0,0.5,0).
S5, respectively calculating loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and carrying out weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain final loss values.
In the embodiment of the present invention, the preset number of Loss functions may be four different Loss functions, which are CE Loss functions, focal Loss functions, dice Loss functions, and lovassz Loss functions, respectively.
In an embodiment of the present invention, the calculating, based on the disaster damage degree category and a preset real disaster damage category, a loss value corresponding to a preset number of loss functions includes:
The CE loss functions in the preset number of loss functions are as follows:
Wherein M represents the number of disaster damage degree categories, p c represents the disaster damage degree category, and y c represents the real disaster damage category.
In an embodiment of the present invention, the calculating, based on the disaster damage degree category and a preset real disaster damage category, a loss value corresponding to a preset number of loss functions includes:
the Focal loss function in the loss functions of the preset number is as follows:
wherein p is the disaster damage degree category, and alpha and gamma are fixed parameters.
In an embodiment of the present invention, the calculating, based on the disaster damage degree category and a preset real disaster damage category, a loss value corresponding to a preset number of loss functions includes:
The Dice Loss function in the Loss function of the preset number is:
And TP, FP and FN respectively represent the numbers of pixels with true positives, false positives and false negatives in the damage degree category.
In an embodiment of the present invention, the calculating, based on the disaster damage degree category and a preset real disaster damage category, a loss value corresponding to a preset number of loss functions includes:
the Lovasz loss function in the loss functions with the preset number is as follows:
wherein y * is the damage degree category, And c is a fixed parameter for the real disaster damage category.
Specifically, the step of performing weighted accumulation on the preset loss weight value and the loss value corresponding to the loss function to obtain a final loss value includes:
and calculating to obtain a final loss value based on a preset final loss formula.
Further, the final loss formula is:
Wherein L F is the final loss value, L, FL, s and Respectively, the loss values delta, epsilon, zeta and/>, corresponding to the loss functions of the preset numberAnd respectively presetting a plurality of loss weight values.
And S6, adjusting the disaster damage identification model according to the size between the final loss value and a preset loss threshold value to obtain a standard disaster damage identification model.
In the embodiment of the invention, the disaster damage identification model is adjusted according to the magnitude of the final loss value and the magnitude of the preset loss threshold, when the final loss value is smaller than the loss threshold, the disaster damage identification model is output as a standard disaster damage identification model, and when the final loss value is larger than or equal to the loss threshold, the model parameters of the disaster damage identification model are adjusted until the final loss value is smaller than the loss threshold, and the disaster damage identification model with the model parameters adjusted is output as the standard disaster damage identification model.
The model parameters may be model weights or model gradients.
S7, acquiring a remote sensing image to be identified, and inputting the remote sensing image to be identified into the standard disaster damage identification model to obtain a disaster damage identification result corresponding to the remote sensing image to be identified.
In the embodiment of the invention, the remote sensing image to be identified is an image which needs disaster damage identification after the detected disaster occurs, and the remote sensing image to be identified is input into the standard disaster damage identification model, so that a disaster damage identification result corresponding to the remote sensing image to be identified can be obtained.
In detail, the disaster damage recognition result obtained by recognition can identify the place where the disaster occurs, and thus, the area is focused on.
According to the embodiment of the invention, the obtained original remote sensing image set is subjected to data enhancement, so that the richness of the data of the obtained training remote sensing image set can be improved, and the robustness of subsequent model training is enhanced. The disaster degree category corresponding to the training remote sensing image set is identified by utilizing a disaster degree identification model, a first distribution value and a second distribution value corresponding to the disaster degree are calculated based on the number of pixels corresponding to the training remote sensing image set under the disaster degree category, the first distribution value and the second distribution value reflect the distribution condition of the disaster degree, the accuracy of a loss function and the accuracy of model training can be improved by calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value and constructing a loss function according to the preset loss weight values, the disaster recognition model is adjusted according to the size between the final loss value and a preset loss threshold value, a standard disaster recognition model is obtained, the remote sensing image to be recognized is recognized, and a disaster recognition result corresponding to the remote sensing image to be recognized is obtained. Therefore, the disaster damage identification method based on the remote sensing image can solve the problem that the disaster damage identification efficiency is not high enough.
Fig. 2 is a functional block diagram of a disaster damage recognition device based on remote sensing images according to an embodiment of the present invention.
The disaster damage recognition device 100 based on remote sensing images can be installed in electronic equipment. Depending on the implementation function, the remote sensing image-based disaster damage recognition device 100 may include a data enhancement module 101, a category prediction module 102, a model training module 103, and a disaster damage recognition module 104. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
The data enhancement module 101 is configured to obtain an original remote sensing image set, and perform data enhancement on the original remote sensing image set to obtain a training remote sensing image set;
The category prediction module 102 is configured to input the training remote sensing image set into a preset disaster damage recognition model, so as to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set;
The model training module 103 is configured to count the number of pixels corresponding to the training remote sensing image set under the disaster damage degree category, calculate a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of pixels, calculate a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method, calculate loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and perform weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain a final loss value, and adjust the disaster damage recognition model according to the size between the final loss value and a preset loss threshold to obtain a standard disaster damage recognition model;
the disaster damage recognition module 104 is configured to obtain a remote sensing image to be recognized, input the remote sensing image to be recognized into the standard disaster damage recognition model, and obtain a disaster damage recognition result corresponding to the remote sensing image to be recognized.
In detail, the specific embodiments of each module of the disaster damage recognition device 100 based on the remote sensing image are as follows:
step one, acquiring an original remote sensing image set, and carrying out data enhancement on the original remote sensing image set to obtain a training remote sensing image set.
In the embodiment of the present invention, the original Remote sensing image set includes a large number of original Remote sensing images, where the original Remote sensing image (Remote SENSING IMAGE) is a film or a photo for recording electromagnetic wave sizes of various ground objects, and is mainly divided into an aerial photo and a satellite photo. The 16 m resolution remote sensing image of the high-resolution one-number multispectral camera WFV can be obtained from the state-run assets source satellite application center.
Specifically, the performing data enhancement on the original remote sensing image set to obtain a training remote sensing image set includes:
detecting a missing remote sensing image in the original remote sensing image set, and executing deletion operation on the missing remote sensing image to obtain an original remote sensing image set;
and performing image rotation, image translation and image scaling on the initial remote sensing images in the initial remote sensing image set to obtain a training remote sensing image set.
In detail, the missing remote sensing image in the original remote sensing image set may be an image with serious missing image information.
The original remote sensing image set is subjected to data enhancement to obtain a training remote sensing image set, so that the number of images in the training remote sensing image set is enriched, and the robustness of subsequent model training is improved.
And secondly, inputting the training remote sensing image set into a preset disaster damage recognition model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set.
In the embodiment of the invention, the disaster damage identification model can be networks such as res-net, U-net, deeplab-v3 and the like. In the scheme, the disaster damage identification model is a U-net network. The U-net network is formed by a left compression channel and a right expansion channel to form the shape of a letter U. Wherein the compression channel is a typical convolutional neural network structure.
Specifically, the inputting the training remote sensing image set into a preset disaster damage recognition model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set includes:
performing convolution processing and pooling processing on the training remote sensing image set by utilizing a compression channel in the disaster damage identification model to obtain an initial pooled image set;
Performing deconvolution operation on the initial pool image set to obtain a deconvolution image set;
performing image stitching on the initial pool image set and the deconvolution image set, and performing feature extraction on the stitched image set to obtain a feature image set;
and inputting the characteristic image set to an output layer in the disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set.
In detail, the disaster recognition model is composed of a left compression channel (Contracting Path) and a right expansion channel (Expansive Path). The compression channel is a typical convolutional neural network structure, which repeatedly adopts a structure of 2 convolutional layers and 1 maximum pooling layer, and the dimension of the characteristic diagram is increased by 1 time after each pooling operation. And in the expansion channel, firstly, carrying out deconvolution operation for 1 time to halve the dimension of the feature map, then splicing the feature map obtained by cutting the corresponding compression channel, reconstructing a feature map with the size of 2 times, carrying out feature extraction by adopting 2 convolution layers, and repeating the structure. At the final output layer, the 64-dimensional feature map is mapped into a 2-dimensional output map with 2 convolutional layers. The output graph comprises disaster damage degree categories.
The disaster damage degree category refers to the severity degree of disaster damage and can be divided into primary disaster damage, secondary disaster damage, tertiary disaster damage and the like.
And thirdly, counting the number of pixels corresponding to the training remote sensing image set under the disaster damage degree category, and calculating a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of pixels.
In the embodiment of the present invention, the counting the number of pixels corresponding to the training remote sensing image set in the disaster damage degree category includes:
determining the areas of disaster damage degree categories corresponding to different training remote sensing images in the training remote sensing image set;
And identifying and summarizing the pixel points in the areas with different disaster damage degree categories to obtain the number of the pixel points corresponding to the different disaster damage degree categories.
In detail, the disaster degree category corresponding to the training remote sensing image comprises primary disaster damage, secondary disaster damage or tertiary disaster damage marked on the remote sensing image, the areas where the primary disaster damage, the secondary disaster damage and the tertiary disaster damage are located are respectively determined, and the number of pixels in the areas where different disaster damage are located is identified and counted.
Specifically, the calculating, based on the number of the pixel points, a first distribution value and a second distribution value corresponding to the disaster degree includes:
Utilizing a preset first distribution value formula and a preset second distribution value formula;
The number of the pixel points and the number of categories of the disaster damage degree obtained in advance are used as the input of the first step value formula, and a first distribution value is obtained;
and taking the number of categories of the disaster damage degree as the input of the second distribution value formula to obtain a second distribution value.
Further, the preset first distribution value formula is:
Wherein s1 is the first distribution value, w i represents the number of the i-th type disaster damage degree pixel points, and k represents the number of the types of disaster damage degree.
Specifically, the preset second distribution value formula is:
Wherein s2 is the second distribution value, and k represents the number of categories of the disaster damage degree.
And step four, calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method.
In the embodiment of the present invention, the calculating, by using a preset weight assignment method, a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value includes:
Multiplying the second distribution value with a preset first reference value to obtain a first standard value, and determining the preset loss weight value as a first combined value if the first distribution value is greater than or equal to the first standard value and less than or equal to the second distribution value; or (b)
Multiplying the second distribution value with a preset second reference value to obtain a second standard value, and determining that the preset loss weight value is a second combined value if the first distribution value is greater than or equal to the second standard value and less than or equal to the first standard value; or (b)
Multiplying the second distribution value with a preset third reference value to obtain a third standard value, and determining the preset loss weight value as a third combined value if the first distribution value is greater than or equal to the third standard value and less than or equal to the second standard value; or (b)
And if the first distribution value is larger than or equal to a preset fourth standard value and smaller than or equal to the third standard value, determining that the preset loss weight value is a fourth combined value.
Preferably, the first reference value is 0.75, the second reference value is 0.5, the third reference value is 0.25, the fourth reference value is 0, the first combined value is (0.15,0.35,0.15,0.35), the second combined value is (0.25,0.25,0.25,0.25), the third combined value is (0.35,0.15,0.35,0.15), and the fourth combined value is (0.5,0,0.5,0).
And fifthly, respectively calculating loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and carrying out weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain final loss values.
In the embodiment of the present invention, the preset number of Loss functions may be four different Loss functions, which are CE Loss functions, focal Loss functions, dice Loss functions, and lovassz Loss functions, respectively.
In an embodiment of the present invention, the calculating, based on the disaster damage degree category and a preset real disaster damage category, a loss value corresponding to a preset number of loss functions includes:
The CE loss functions in the preset number of loss functions are as follows:
Wherein M represents the number of disaster damage degree categories, p c represents the disaster damage degree category, and y c represents the real disaster damage category.
In an embodiment of the present invention, the calculating, based on the disaster damage degree category and a preset real disaster damage category, a loss value corresponding to a preset number of loss functions includes:
the Focal loss function in the loss functions of the preset number is as follows:
wherein p is the disaster damage degree category, and alpha and gamma are fixed parameters.
In an embodiment of the present invention, the calculating, based on the disaster damage degree category and a preset real disaster damage category, a loss value corresponding to a preset number of loss functions includes:
The Dice Loss function in the Loss function of the preset number is:
And TP, FP and FN respectively represent the numbers of pixels with true positives, false positives and false negatives in the damage degree category.
In an embodiment of the present invention, the calculating, based on the disaster damage degree category and a preset real disaster damage category, a loss value corresponding to a preset number of loss functions includes:
the Lovasz loss function in the loss functions with the preset number is as follows:
wherein y * is the damage degree category, And c is a fixed parameter for the real disaster damage category.
Specifically, the step of performing weighted accumulation on the preset loss weight value and the loss value corresponding to the loss function to obtain a final loss value includes:
and calculating to obtain a final loss value based on a preset final loss formula.
Further, the final loss formula is:
Wherein L F is the final loss value, L, FL, s and Respectively, the loss values delta, epsilon, zeta and/>, corresponding to the loss functions of the preset numberAnd respectively presetting a plurality of loss weight values.
And step six, adjusting the disaster damage identification model according to the size between the final loss value and a preset loss threshold value to obtain a standard disaster damage identification model.
In the embodiment of the invention, the disaster damage identification model is adjusted according to the magnitude of the final loss value and the magnitude of the preset loss threshold, when the final loss value is smaller than the loss threshold, the disaster damage identification model is output as a standard disaster damage identification model, and when the final loss value is larger than or equal to the loss threshold, the model parameters of the disaster damage identification model are adjusted until the final loss value is smaller than the loss threshold, and the disaster damage identification model with the model parameters adjusted is output as the standard disaster damage identification model.
The model parameters may be model weights or model gradients.
And step seven, acquiring a remote sensing image to be identified, and inputting the remote sensing image to be identified into the standard disaster damage identification model to obtain a disaster damage identification result corresponding to the remote sensing image to be identified.
In the embodiment of the invention, the remote sensing image to be identified is an image which needs disaster damage identification after the detected disaster occurs, and the remote sensing image to be identified is input into the standard disaster damage identification model, so that a disaster damage identification result corresponding to the remote sensing image to be identified can be obtained.
In detail, the disaster damage recognition result obtained by recognition can identify the place where the disaster occurs, and thus, the area is focused on.
According to the embodiment of the invention, the obtained original remote sensing image set is subjected to data enhancement, so that the richness of the data of the obtained training remote sensing image set can be improved, and the robustness of subsequent model training is enhanced. The disaster degree category corresponding to the training remote sensing image set is identified by utilizing a disaster degree identification model, a first distribution value and a second distribution value corresponding to the disaster degree are calculated based on the number of pixels corresponding to the training remote sensing image set under the disaster degree category, the first distribution value and the second distribution value reflect the distribution condition of the disaster degree, the accuracy of a loss function and the accuracy of model training can be improved by calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value and constructing a loss function according to the preset loss weight values, the disaster recognition model is adjusted according to the size between the final loss value and a preset loss threshold value, a standard disaster recognition model is obtained, the remote sensing image to be recognized is recognized, and a disaster recognition result corresponding to the remote sensing image to be recognized is obtained. Therefore, the disaster damage recognition device based on the remote sensing image can solve the problem that the disaster damage recognition efficiency is not high enough.
Fig. 3 is a schematic structural diagram of an electronic device for implementing a disaster damage identification method based on remote sensing images according to an embodiment of the present invention.
The electronic device 1 may comprise a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may further comprise a computer program stored in the memory 11 and executable on the processor 10, such as a damage identification program based on telemetry images.
The processor 10 may be formed by an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be formed by a plurality of integrated circuits packaged with the same function or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, and combinations of various control chips. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the entire electronic device using various interfaces and lines, executes or executes programs or modules stored in the memory 11 (for example, executes a damage recognition program based on a remote sensing image, etc.), and invokes data stored in the memory 11 to perform various functions of the electronic device and process data.
The memory 11 includes at least one type of readable storage medium including flash memory, a removable hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 11 may also be an external storage device of the electronic device in other embodiments, such as a plug-in mobile hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD) or the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used to store not only application software installed in an electronic device and various data, such as codes of a disaster damage recognition program based on a remote sensing image, but also temporarily store data that has been output or is to be output.
The communication bus 12 may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
The communication interface 13 is used for communication between the electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), or alternatively a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.
Fig. 3 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 3 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device may further include a power source (such as a battery) for supplying power to the respective components, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device may further include various sensors, bluetooth modules, wi-Fi modules, etc., which are not described herein.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The remote sensing image-based disaster damage recognition program stored in the memory 11 of the electronic device 1 is a combination of a plurality of instructions, and when running in the processor 10, it can be implemented:
acquiring an original remote sensing image set, and carrying out data enhancement on the original remote sensing image set to obtain a training remote sensing image set;
inputting the training remote sensing image set into a preset disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set;
counting the number of pixel points corresponding to the training remote sensing image set under the disaster damage degree category, and calculating a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of the pixel points;
Calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method;
Respectively calculating loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and carrying out weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain final loss values;
The disaster damage recognition model is adjusted according to the size between the final loss value and a preset loss threshold value, and a standard disaster damage recognition model is obtained;
and acquiring a remote sensing image to be identified, and inputting the remote sensing image to be identified into the standard disaster damage identification model to obtain a disaster damage identification result corresponding to the remote sensing image to be identified.
In particular, the specific implementation method of the above instructions by the processor 10 may refer to the description of the relevant steps in the corresponding embodiment of the drawings, which is not repeated herein.
Further, the modules/units integrated in the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. The computer readable storage medium may be volatile or nonvolatile. For example, the computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
The present invention also provides a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, can implement:
acquiring an original remote sensing image set, and carrying out data enhancement on the original remote sensing image set to obtain a training remote sensing image set;
inputting the training remote sensing image set into a preset disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set;
counting the number of pixel points corresponding to the training remote sensing image set under the disaster damage degree category, and calculating a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of the pixel points;
Calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method;
Respectively calculating loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and carrying out weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain final loss values;
The disaster damage recognition model is adjusted according to the size between the final loss value and a preset loss threshold value, and a standard disaster damage recognition model is obtained;
and acquiring a remote sensing image to be identified, and inputting the remote sensing image to be identified into the standard disaster damage identification model to obtain a disaster damage identification result corresponding to the remote sensing image to be identified.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The blockchain (Blockchain), essentially a de-centralized database, is a string of data blocks that are generated in association using cryptographic methods, each of which contains information from a batch of network transactions for verifying the validity (anti-counterfeit) of its information and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.
The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Wherein artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) is the theory, method, technique, and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend, and expand human intelligence, sense the environment, acquire knowledge, and use knowledge to obtain optimal results.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (7)
1. The disaster damage identification method based on the remote sensing image is characterized by comprising the following steps:
acquiring an original remote sensing image set, and carrying out data enhancement on the original remote sensing image set to obtain a training remote sensing image set;
inputting the training remote sensing image set into a preset disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set;
counting the number of pixel points corresponding to the training remote sensing image set under the disaster damage degree category, and calculating a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of the pixel points;
Calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method;
Respectively calculating loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and carrying out weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain final loss values;
The disaster damage recognition model is adjusted according to the size between the final loss value and a preset loss threshold value, and a standard disaster damage recognition model is obtained;
Acquiring a remote sensing image to be identified, and inputting the remote sensing image to be identified into the standard disaster damage identification model to obtain a disaster damage identification result corresponding to the remote sensing image to be identified;
The calculating, based on the number of the pixel points, a first distribution value and a second distribution value corresponding to the disaster degree includes: utilizing a preset first distribution value formula and a preset second distribution value formula; the number of the pixel points and the number of categories of the disaster damage degree obtained in advance are used as the input of the first distribution value formula, and a first distribution value is obtained; the number of categories of the disaster damage degree is used as the input of the second distribution value formula, and a second distribution value is obtained;
the preset first distribution value formula is as follows:
Wherein, For the first distribution value,/>Represents the/>Weight value corresponding to number of disaster damage degree-like pixel points,/>The number of categories representing the disaster damage degree;
the preset second distribution value formula is as follows:
Wherein, For the second distribution value,/>The number of categories representing the disaster damage degree;
The calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method includes: multiplying the second distribution value with a preset first reference value to obtain a first standard value, and determining the preset loss weight value as a first combined value if the first distribution value is greater than or equal to the first standard value and less than or equal to the second distribution value; or, multiplying the second distribution value with a preset second reference value to obtain a second standard value, and if the first distribution value is greater than or equal to the second standard value and less than or equal to the first standard value, determining that the preset loss weight value is a second combined value; or, multiplying the second distribution value with a preset third reference value to obtain a third standard value, and if the first distribution value is greater than or equal to the third standard value and less than or equal to the second standard value, determining that the preset loss weight value is a third combined value; or if the first distribution value is greater than or equal to a preset fourth standard value and less than or equal to the third standard value, determining that the preset loss weight value is a fourth combined value.
2. The method for identifying disaster damage based on remote sensing image as set forth in claim 1, wherein said inputting the training remote sensing image set into a preset disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image set includes:
performing convolution processing and pooling processing on the training remote sensing image set by utilizing a compression channel in the disaster damage identification model to obtain an initial pooled image set;
Performing deconvolution operation on the initial pool image set to obtain a deconvolution image set;
performing image stitching on the initial pool image set and the deconvolution image set, and performing feature extraction on the stitched image set to obtain a feature image set;
and inputting the characteristic image set to an output layer in the disaster damage identification model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set.
3. The method for identifying disaster damage based on remote sensing images according to claim 1, wherein said counting the number of pixels corresponding to the training remote sensing image set in the disaster damage degree category comprises:
determining the areas of disaster damage degree categories corresponding to different training remote sensing images in the training remote sensing image set;
And identifying and summarizing the pixel points in the areas with different disaster damage degree categories to obtain the number of the pixel points corresponding to the different disaster damage degree categories.
4. The method for identifying disaster damage based on remote sensing images according to any one of claims 1 to 3, wherein the performing data enhancement on the original remote sensing image set to obtain a training remote sensing image set includes:
detecting a missing remote sensing image in the original remote sensing image set, and executing deletion operation on the missing remote sensing image to obtain an original remote sensing image set;
and performing image rotation, image translation and image scaling on the initial remote sensing images in the initial remote sensing image set to obtain a training remote sensing image set.
5. A remote sensing image-based disaster damage recognition device for implementing the remote sensing image-based disaster damage recognition method according to any one of claims 1 to 4, wherein the device comprises:
the data enhancement module is used for acquiring an original remote sensing image set, and carrying out data enhancement on the original remote sensing image set to obtain a training remote sensing image set;
The category prediction module is used for inputting the training remote sensing image set into a preset disaster damage recognition model to obtain a disaster damage degree category corresponding to the training remote sensing image in the training remote sensing image set;
The model training module is used for counting the number of pixel points corresponding to the training remote sensing image set under the disaster damage degree category, calculating a first distribution value and a second distribution value corresponding to the disaster damage degree based on the number of the pixel points, calculating a plurality of preset loss weight values corresponding to the first distribution value and the second distribution value by using a preset weight assignment method, respectively calculating loss values corresponding to a preset number of loss functions based on the disaster damage degree category and a preset real disaster damage category, and carrying out weighted accumulation on the preset loss weight values and the loss values corresponding to the loss functions to obtain a final loss value, and adjusting the disaster damage identification model according to the size between the final loss value and a preset loss threshold value to obtain a standard disaster damage identification model;
the disaster damage recognition module is used for acquiring a remote sensing image to be recognized, inputting the remote sensing image to be recognized into the standard disaster damage recognition model, and obtaining a disaster damage recognition result corresponding to the remote sensing image to be recognized.
6. An electronic device, the 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 enable the at least one processor to perform the remote sensing image-based disaster identification method according to any one of claims 1 to 4.
7. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the remote sensing image-based disaster identification method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111435237.9A CN114120122B (en) | 2021-11-29 | 2021-11-29 | Disaster damage identification method, device, equipment and storage medium based on remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111435237.9A CN114120122B (en) | 2021-11-29 | 2021-11-29 | Disaster damage identification method, device, equipment and storage medium based on remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114120122A CN114120122A (en) | 2022-03-01 |
| CN114120122B true CN114120122B (en) | 2024-04-30 |
Family
ID=80371424
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111435237.9A Active CN114120122B (en) | 2021-11-29 | 2021-11-29 | Disaster damage identification method, device, equipment and storage medium based on remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114120122B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111695609A (en) * | 2020-05-26 | 2020-09-22 | 平安科技（深圳）有限公司 | Target damage degree determination method, target damage degree determination device, electronic device, and storage medium |
| CN112329550A (en) * | 2020-10-16 | 2021-02-05 | 中国科学院空间应用工程与技术中心 | Weak supervision learning-based disaster-stricken building rapid positioning evaluation method and device |
| CN113591714A (en) * | 2021-07-30 | 2021-11-02 | 金陵科技学院 | Flood detection method based on satellite remote sensing image |
| CN113610070A (en) * | 2021-10-11 | 2021-11-05 | 中国地质环境监测院(自然资源部地质灾害技术指导中心) | Landslide disaster identification method based on multi-source data fusion |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10977923B2 (en) * | 2017-12-01 | 2021-04-13 | Nec Corporation | River risk level determining device, river risk level determining method, and storage medium |
-
2021
- 2021-11-29 CN CN202111435237.9A patent/CN114120122B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111695609A (en) * | 2020-05-26 | 2020-09-22 | 平安科技（深圳）有限公司 | Target damage degree determination method, target damage degree determination device, electronic device, and storage medium |
| CN112329550A (en) * | 2020-10-16 | 2021-02-05 | 中国科学院空间应用工程与技术中心 | Weak supervision learning-based disaster-stricken building rapid positioning evaluation method and device |
| CN113591714A (en) * | 2021-07-30 | 2021-11-02 | 金陵科技学院 | Flood detection method based on satellite remote sensing image |
| CN113610070A (en) * | 2021-10-11 | 2021-11-05 | 中国地质环境监测院(自然资源部地质灾害技术指导中心) | Landslide disaster identification method based on multi-source data fusion |
Non-Patent Citations (3)
| Title |
|---|
| Diagnosis of Plant Cold Damage Based on Hyperspectral Imaging and Convolutional Neural Network;WEI YANG et al.;《IEEE Acess》;20190822;第第7卷卷;第118239-118248页 * |
| Study on Evaluation Methods of Flood Disaster Grade Attribute Recognition Analysis and BP Neural Network;Xiaoling Yang et al.;《2009 Sixth International Conference on Fuzzy Systems and Knowledge Discovery》;20091228;第386-390页 * |
| 基于高光谱数据的马尾松病虫害等级分类研究;王霜;《中国优秀硕士学位论文全文数据库农业科技辑》;20200715;第34-82页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114120122A (en) | 2022-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113283446B (en) | Method and device for identifying object in image, electronic equipment and storage medium | |
| CN113705462B (en) | Face recognition method, device, electronic equipment and computer readable storage medium | |
| CN112396005A (en) | Biological characteristic image recognition method and device, electronic equipment and readable storage medium | |
| CN113887438B (en) | Watermark detection method, device, equipment and medium for face image | |
| CN112668575B (en) | Key information extraction method and device, electronic equipment and storage medium | |
| CN112016617B (en) | Fine granularity classification method, apparatus and computer readable storage medium | |
| CN114049568A (en) | Object shape change detection method, device, equipment and medium based on image comparison | |
| CN115063589A (en) | Knowledge distillation-based vehicle component segmentation method and related equipment | |
| CN114550076A (en) | Method, device and equipment for monitoring area abnormal behaviors and storage medium | |
| CN113240585B (en) | Image processing method, device and storage medium based on generation countermeasure network | |
| CN112906671B (en) | Method and device for identifying false face-examination picture, electronic equipment and storage medium | |
| CN114120122B (en) | Disaster damage identification method, device, equipment and storage medium based on remote sensing image | |
| CN113887408B (en) | Method, device, equipment and storage medium for detecting activated face video | |
| CN115049836B (en) | Image segmentation method, device, equipment and storage medium | |
| CN113792801B (en) | Method, device, equipment and storage medium for detecting face dazzling degree | |
| CN115798004A (en) | Face card punching method and device based on local area, electronic equipment and medium | |
| CN112541899B (en) | Incomplete detection method and device of certificate, electronic equipment and computer storage medium | |
| CN113792671A (en) | Method and device for detecting face synthetic image, electronic equipment and medium | |
| CN113705459A (en) | Face snapshot method and device, electronic equipment and storage medium | |
| CN113255456A (en) | Non-active living body detection method, device, electronic equipment and storage medium | |
| CN113869385A (en) | Poster comparison method, device and equipment based on target detection and storage medium | |
| CN113128440A (en) | Target object identification method, device, equipment and storage medium based on edge equipment | |
| CN113936313B (en) | Method, device, equipment and storage medium for detecting out-of-account borrowing of website employee account | |
| CN114627535B (en) | Coordinate matching method, device, equipment and medium based on binocular camera | |
| CN113408424B (en) | Article identification method, apparatus, electronic device and storage medium |
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 |