CN115761346A - Remote sensing image classification method based on multi-model fusion - Google Patents
Remote sensing image classification method based on multi-model fusion Download PDFInfo
- Publication number
- CN115761346A CN115761346A CN202211465646.8A CN202211465646A CN115761346A CN 115761346 A CN115761346 A CN 115761346A CN 202211465646 A CN202211465646 A CN 202211465646A CN 115761346 A CN115761346 A CN 115761346A
- Authority
- CN
- China
- Prior art keywords
- classification
- module
- image
- remote sensing
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 230000004927 fusion Effects 0.000 title claims abstract description 36
- 238000013145 classification model Methods 0.000 claims abstract description 98
- 238000012545 processing Methods 0.000 claims abstract description 71
- 238000000605 extraction Methods 0.000 claims abstract description 38
- 238000001514 detection method Methods 0.000 claims abstract description 33
- 238000010276 construction Methods 0.000 claims abstract description 25
- 238000013500 data storage Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 29
- 238000005457 optimization Methods 0.000 claims description 11
- 230000009466 transformation Effects 0.000 claims description 11
- 238000012937 correction Methods 0.000 claims description 10
- 238000007689 inspection Methods 0.000 claims description 9
- 238000007635 classification algorithm Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 5
- 230000005855 radiation Effects 0.000 claims description 5
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000005192 partition Methods 0.000 abstract description 5
- 238000012795 verification Methods 0.000 abstract description 3
- 239000000284 extract Substances 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a remote sensing image classification method based on multi-model fusion, which relates to the technical field of remote sensing image classification and comprises a remote sensing platform, an image classification system and an image classification model construction module, wherein the remote sensing platform is connected with a central processing unit, the central processing unit is electrically connected with a data storage module and the image classification system, and the image classification system comprises an image feature extraction module, an image feature division module, an image classification model construction module, a classification precision detection module, a classification result detection module and an image classification execution module. According to the remote sensing image classification method based on multi-model fusion, the image classification model is divided into the main body model and the plurality of support rod models, images can be intelligently classified in the image classification model mode, classification of a plurality of global images and partition classification of a single image can be achieved, precision detection and verification can be carried out on the image classification model, and classification reliability is guaranteed.
Description
Technical Field
The invention relates to the technical field of remote sensing image classification, in particular to a remote sensing image classification method based on multi-model fusion.
Background
The remote sensing image, or remote sensing photo, is the product of the information obtained by various sensors, and is the information carrier of the remote sensing detection target. The remote sensing image can distinguish a lot of information, the application of the remote sensing technology is very wide due to the fact that a large amount of information is extracted, and the remote sensing image needs to be classified in the utilization process of the remote sensing image, so that a remote sensing image classification method is needed.
The invention with the application number of CN201710213577.4 discloses an optical remote sensing image classification method based on a multi-branch network fusion model. The method for classifying the chemical remote sensing images similar to the application has the following defects: in the image classification process, an image classification model is constructed in a model training mode, the characteristics of the image cannot be accurately divided, the classification category is fuzzy, the classification result is not subjected to precision detection and verification, and the classification efficiency is high but the reliability is not enough.
In view of the above, a remote sensing image classification method based on multi-model fusion is proposed by improving the existing structure and deficiency.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a remote sensing image classification method based on multi-model fusion, which solves the problems in the background technology.
In order to realize the purpose, the invention is realized by the following technical scheme: the utility model provides a remote sensing image classification method based on multi-model fusion, includes remote sensing platform, image classification system and image classification model construction module, the remote sensing platform is connected with central processing unit, and central processing unit electric connection has data storage module and image classification system, the image classification system includes image characteristic extraction module, image characteristic division module, image classification model construction module, classification precision detection module, classification result detection module and image classification execution module, the output electric connection of image characteristic extraction module has image characteristic division module, and the output electric connection of image characteristic division module has image classification model construction module, the output electric connection of image classification model construction module has classification precision detection module, and the output electric connection of classification precision detection module has classification result detection module, the output electric connection of classification result detection module has image classification execution module.
Furthermore, the image feature extraction module comprises a statistical feature value extraction module and an image operation processing module, and the output end of the statistical feature value extraction module is electrically connected with the image operation processing module.
Furthermore, the statistical characteristic value extraction module is used for extracting a statistical characteristic value of the remote sensing image so as to identify the remote sensing image, and the image operation processing module is used for carrying out operation processing on the digital image, including but not limited to ratio processing, difference processing, principal component transformation and K-T transformation, so as to extract an image characteristic variable.
Furthermore, the feature variables extracted by the image operation processing module include global statistical feature variables and local statistical feature variables, the global statistical feature variables are feature variables obtained by transforming the whole image, and the local statistical feature variables are feature variables obtained by dividing the digital image into different recognition units.
Furthermore, the image feature division module is used for dividing each feature according to the image features extracted by the image feature extraction module, representative and discriminative image features are selected as classification bases, the image classification model construction module is used for constructing a classification model on the basis of image feature classification, and the classification model constructed by the image classification model construction module comprises a main body model and a plurality of support rod models.
Further, the image classification model building module comprises a classification mode making module, a classification category determining module and a classification model building module, wherein the output end of the classification mode making module is electrically connected with the classification category determining module, and the output end of the classification category determining module is electrically connected with the classification model building module.
Furthermore, the classification mode making module is used for selecting a proper image classification algorithm according to the classification requirement and the characteristics of the image data, the classification category determining module is used for determining the category of the image classification, and the classification model building module is used for building an image classification model according to different classification categories on the basis of determining the category of the image classification.
Furthermore, the image classification model building module further comprises a model training module and a model optimization module, the output end of the model training module is electrically connected with the model optimization module, the model training module is used for carrying out model training through remote sensing image data, and the model optimization module evaluates the classification result of the built image classification model on the basis of the model training, further optimizes and modifies the classification result, and obtains a relatively accurate classification model after repeated operation.
Furthermore, the classification precision detection module is used for comparing known training data and classification categories with classification results to confirm classification precision and reliability, the classification result inspection module is used for carrying out statistical inspection on statistical results of distinguishing classification, and the image classification execution module is used for classifying remote sensing images through an image classification model.
Furthermore, the image classification execution module comprises a remote sensing image acquisition module, a remote sensing image processing module, a remote sensing image input module and a classification result output module, wherein the output end of the remote sensing image acquisition module is electrically connected with the remote sensing image processing module, the output end of the remote sensing image processing module is electrically connected with the remote sensing image input module, and the output end of the remote sensing image input module is electrically connected with the classification result output module.
Furthermore, the remote sensing image acquisition module is used for collecting data of remote sensing images to be classified, the remote sensing image processing module is used for processing and analyzing the remote sensing images, the processing comprises image radiation correction, geometric correction and filtering processing, high-precision remote sensing images are obtained, the remote sensing image input module is used for inputting the remote sensing images to the image classification model, and the classification result output module is used for outputting image classification results.
The invention provides a remote sensing image classification method based on multi-model fusion, which has the following beneficial effects:
according to the remote sensing image classification method based on multi-model fusion, the image classification model is divided into the main body model and the plurality of support rod models, images can be intelligently classified in the image classification model mode, classification of a plurality of global images and partition classification of a single image can be achieved, precision detection and verification can be conducted on the image classification model, and classification reliability is guaranteed.
1. The remote sensing image classification method based on multi-model fusion is provided with an image feature extraction module, the statistical feature extraction module extracts statistical feature values of remote sensing images so as to identify the remote sensing images, the image operation processing module performs operation processing on the digital images, including but not limited to ratio processing, difference processing, principal component transformation and K-T transformation, and further extracts image feature variables, so that the feature extraction mode of the image feature extraction module is used as the feature extraction mode of a main model, the feature variables extracted by the image operation processing module include global statistical feature variables and local statistical feature variables, the global statistical feature variables are obtained by performing transformation processing on the whole images, the local statistical feature variables are obtained by dividing the digital images into different identification units, and the feature variables of each unit are obtained so as to facilitate subsequent classification of a plurality of global images and partition classification of a single image.
2. The remote sensing image classification method based on multi-model fusion is provided with an image feature division module, the image feature division module can divide all features according to the image features extracted by the image feature extraction module, representative and discriminative image features are selected as classification bases, the image classification model construction module is used for constructing a classification model on the basis of image feature classification, and the classification model constructed by the image classification model construction module comprises a main body model and a plurality of support rod models, so that multi-model fusion is convenient to realize.
3. The remote sensing image classification method based on multi-model fusion is provided with an image classification model establishing module, a classification mode establishing module selects a proper image classification algorithm according to classification requirements and the characteristics of image data, a classification category determining module determines the category of image classification, the classification model establishing module establishes an image classification model according to different classification categories on the basis of determining the category of the image classification, the image classification model comprises a main body model and a support rod model, a model training module carries out model training through the remote sensing image data, a model optimizing module evaluates the classification result of the established image classification model on the basis of the model training, further optimizes and modifies the classification model, and a more accurate classification model is obtained after repeated times.
4. The remote sensing image classification method based on multi-model fusion is provided with a classification precision detection module and a classification result inspection module, wherein the classification precision detection module compares known training data and classification categories with classification results to confirm classification precision and reliability, the classification result inspection module performs statistical inspection on statistical results for distinguishing the classifications to ensure the reliability of the classification results, an image classification execution module classifies remote sensing images through an image classification model, a main body model of the image classification model extracts image features, and all strut models of the image classification model divide the images according to image features such as the wave-front characteristics and the geometric shapes of the remote sensing images.
5. The remote sensing image classification method based on multi-model fusion is provided with an image classification execution module, a remote sensing image acquisition module collects data of remote sensing images to be classified, a remote sensing image processing module processes and analyzes the remote sensing images, the remote sensing images comprise image radiation correction, geometric correction and filtering processing, high-precision remote sensing images are obtained, a remote sensing image input module inputs the remote sensing images into an image classification model, and a classification result output module outputs image classification results.
Drawings
FIG. 1 is a schematic diagram of a system framework structure of a remote sensing image classification method based on multi-model fusion according to the present invention;
FIG. 2 is a schematic structural diagram of an image classification system of a remote sensing image classification method based on multi-model fusion according to the present invention;
FIG. 3 is a schematic structural diagram of an image feature extraction module of the remote sensing image classification method based on multi-model fusion according to the present invention;
FIG. 4 is a schematic structural diagram of an image classification model construction module of the remote sensing image classification method based on multi-model fusion, according to the present invention;
FIG. 5 is a schematic structural diagram of an image classification execution module of the remote sensing image classification method based on multi-model fusion, according to the present invention;
FIG. 6 is a schematic diagram of an image classification model structure of a remote sensing image classification method based on multi-model fusion.
In the figure: 1. a remote sensing platform; 2. a central processing unit; 3. a data storage module; 4. an image classification system; 5. an image feature extraction module; 501. a statistical eigenvalue extraction module; 502. an image operation processing module; 6. an image feature division module; 7. an image classification model construction module; 701. a classification mode formulation module; 702. a classification category determination module; 703. a classification model construction module; 704. a model training module; 705. a model optimization module; 8. a classification precision detection module; 9. a classification result checking module; 10. an image classification execution module; 1001. a remote sensing image acquisition module; 1002. a remote sensing image processing module; 1003. a remote sensing image input module; 1004. and a classification result output module.
Detailed Description
The embodiments of the present invention will be described in further detail with reference to the drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
Referring to fig. 1 to 6, the present invention provides a technical solution: a remote sensing image classification method based on multi-model fusion comprises a remote sensing platform 1, an image classification system 4 and an image classification model building module 7, wherein the remote sensing platform 1 is connected with a central processing unit 2, the central processing unit 2 is electrically connected with a data storage module 3 and the image classification system 4, the image classification system 4 comprises an image feature extraction module 5, an image feature division module 6, an image classification model building module 7, a classification precision detection module 8, a classification result detection module 9 and an image classification execution module 10, the output end of the image feature extraction module 5 is electrically connected with the image feature division module 6, the output end of the image feature division module 6 is electrically connected with the image classification model building module 7, the output end of the image classification model building module 7 is electrically connected with the classification precision detection module 8, the output end of the classification precision detection module 8 is electrically connected with the classification result detection module 9, and the output end of the classification result detection module 9 is electrically connected with the image classification execution module 10;
the image classification method comprises the following specific operations that an image feature division module 6 divides all features according to the image features extracted by an image feature extraction module 5, representative and discriminative image features are selected as classification bases, an image classification model construction module 7 constructs a classification model on the basis of image feature classification, the classification model constructed by the image classification model construction module 7 comprises a main body model and a plurality of support rod models, the main body model of the image classification model extracts the image features, all the support rod models of the image classification model divide images according to the image features of the remote sensing images, such as the Bopu characteristics and the geometric shapes, a classification precision detection module 8 compares known training data and classification categories with classification results to confirm the precision and the reliability of classification, a classification result detection module 9 carries out statistical detection on the statistical results of distinguishing classification to ensure the reliability of the classification results, and an image classification execution module 10 classifies the remote sensing images through the image classification model.
Referring to fig. 3, the image feature extraction module 5 includes a statistical feature value extraction module 501 and an image operation processing module 502, and an output end of the statistical feature value extraction module 501 is electrically connected to the image operation processing module 502;
the specific operation is as follows, the statistical characteristic value extraction module 501 extracts the statistical characteristic value of the remote sensing image so as to identify the remote sensing image, and the image operation processing module 502 performs operation processing on the digital image, including but not limited to ratio processing, difference processing, principal component transformation and K-T transformation, so as to extract the image characteristic variable; the feature variables extracted by the image operation processing module 502 include global statistical feature variables, which are obtained by transforming the entire image, and local statistical feature variables, which are obtained by dividing the digital image into different recognition units.
Referring to fig. 4, the image classification model building module 7 includes a classification mode making module 701, a classification category determining module 702, and a classification model building module 703, an output end of the classification mode making module 701 is electrically connected to the classification category determining module 702, and an output end of the classification category determining module 702 is electrically connected to the classification model building module 703; the image classification model construction module 7 further comprises a model training module 704 and a model optimization module 705, and the output end of the model training module 704 is electrically connected with the model optimization module 705;
the specific operation is as follows, a classification mode making module 701 selects a proper image classification algorithm according to classification requirements and characteristics of image data, a classification category determining module 702 determines categories of image classification, a classification model building module 703 builds image classification models according to different classification categories on the basis of determining the categories of the image classification, a model training module 704 performs model training through remote sensing image data, a model optimizing module 705 evaluates classification results of the built image classification models on the basis of the model training and further optimizes and modifies the classification models, and a more accurate classification model is obtained after repeated times.
Referring to fig. 5, the image classification execution module 10 includes a remote sensing image acquisition module 1001, a remote sensing image processing module 1002, a remote sensing image input module 1003 and a classification result output module 1004, wherein an output end of the remote sensing image acquisition module 1001 is electrically connected to the remote sensing image processing module 1002, an output end of the remote sensing image processing module 1002 is electrically connected to the remote sensing image input module 1003, and an output end of the remote sensing image input module 1003 is electrically connected to the classification result output module 1004;
the remote sensing image classification method comprises the following specific operations that a remote sensing image acquisition module 1001 collects data of a remote sensing image to be classified, a remote sensing image processing module 1002 processes and analyzes the remote sensing image, the remote sensing image processing module comprises image radiation correction, geometric correction and filtering processing, a high-precision remote sensing image is obtained, a remote sensing image input module 1003 inputs the remote sensing image into an image classification model, and a classification result output module 1004 outputs an image classification result.
In summary, the remote sensing image classification method based on multi-model fusion is used, firstly, a remote sensing platform 1 transmits remote sensing image data to a central processing unit 2 and stores the data through a data storage module 3, then an image classification system 4 carries out image classification, in the process, firstly, a statistical characteristic value extraction module 501 extracts a statistical characteristic value of the remote sensing image so as to identify the remote sensing image, then, an image operation processing module 502 carries out operation processing on a digital image, wherein the operation processing includes but is not limited to ratio processing, difference processing, principal component transformation and K-T transformation, and further, image characteristic variables are extracted, so that a characteristic extraction mode of an image characteristic extraction module 5 is used as a characteristic extraction mode of a main body model, the characteristic variables extracted by the image operation processing module 502 comprise global statistical characteristic variables and local statistical characteristic variables, the global statistical characteristic variables are obtained by carrying out conversion processing on the whole image, and the local statistical characteristic variables are obtained by dividing the digital image into different identification units, and characteristic variables of each unit are obtained;
then the image feature division module 6 can divide each feature according to the image features extracted by the image feature extraction module 5, representative and discriminative image features are selected as a classification basis, the image classification model construction module 7 constructs a classification model on the basis of image feature classification, in the image construction process, firstly, the classification mode formulation module 701 selects a proper image classification algorithm according to classification requirements and the features of image data, then, the classification category determination module 702 determines the classification of image classification, finally, the classification model construction module 703 constructs an image classification model according to different classification categories and sets model parameters on the basis of determining the classification category of the image, wherein the image classification model comprises a main body model and a plurality of support rod models, then, the model training module 704 performs model training through remote sensing image data, the model optimization module 705 evaluates the classification result of the constructed image classification model on the basis of the model training, further optimizes and modifies the classification result, and obtains a more accurate classification model after repeated times;
after the image classification model is constructed, known training data and classification categories are compared with classification results through a classification precision detection module 8, the classification precision and reliability are confirmed, then a classification result inspection module 9 carries out statistical inspection on the statistical results of the classification discrimination to ensure the reliability of the classification results, otherwise, model training optimization is carried out again, when the classification models meet the requirements, an image classification execution module 10 classifies the remote sensing images through an image classification model, in the process, a remote sensing image acquisition module 1001 carries out data collection on the remote sensing images to be classified, and then a remote sensing image processing module 1002 carries out processing analysis on the remote sensing images, wherein the processing comprises image radiation correction, geometric correction and filtering processing to obtain high-precision remote sensing images;
then the remote sensing image input module 1003 inputs the remote sensing image to the image classification model for model classification, at this time, the main model of the image classification model extracts image features, and can also perform classification of a plurality of global images and partition classification of a single image according to requirements, when the classification of the plurality of global images is performed, global statistical feature variable extraction is performed in the process of extracting image features by the main model, when the partition classification of the single image is performed, local statistical feature variable extraction is performed in the process of extracting image features by the main model, then each strut model of the image classification model divides the image according to the image features of the remote sensing image, such as the color classification strut model, the gray classification strut model, the shape classification strut model and the like, and finally the classification result is output through the classification result output module 1004, thus completing the using process of the whole remote sensing image classification method based on multi-model fusion.
The embodiments of the present invention have been presented for purposes of illustration and description, and are not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiment was chosen and described in order to best explain the principles of the invention and the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
Claims (10)
1. The utility model provides a remote sensing image classification method based on multi-model fusion, its characterized in that, including remote sensing platform (1), image classification system (4) and image classification model construction module (7), remote sensing platform (1) is connected with central processing unit (2), and central processing unit (2) electric connection has data storage module (3) and image classification system (4), image classification system (4) are including image characteristic extraction module (5), image characteristic division module (6), image classification model construction module (7), classification precision detection module (8), classification result detection module (9) and image classification execution module (10), the output electric connection of image characteristic extraction module (5) has image characteristic division module (6), and the output electric connection of image characteristic division module (6) has image classification model construction module (7), the output electric connection of image classification model construction module (7) has classification precision detection module (8), and the output electric connection of classification precision detection module (8) has classification result detection module (9), the output electric connection of classification result detection module (9) has image classification execution module (10).
2. The remote sensing image classification method based on multi-model fusion as claimed in claim 1, wherein the image feature extraction module (5) comprises a statistical feature value extraction module (501) and an image operation processing module (502), and the output end of the statistical feature value extraction module (501) is electrically connected with the image operation processing module (502).
3. The remote sensing image classification method based on multi-model fusion as claimed in claim 2, characterized in that the statistical characteristic value extraction module (501) is used for extracting the statistical characteristic value of the remote sensing image so as to identify the remote sensing image, and the image operation processing module (502) is used for performing operation processing on the digital image, including but not limited to ratio processing, difference processing, principal component transformation and K-T transformation, so as to extract the image characteristic variables.
4. The remote sensing image classification method based on multi-model fusion as claimed in claim 2, characterized in that the feature variables extracted by the image operation processing module (502) include global statistical feature variables and local statistical feature variables, the global statistical feature variables are feature variables obtained by transforming the whole image, and the local statistical feature variables are feature variables obtained by dividing the digital image into different recognition units.
5. The remote sensing image classification method based on multi-model fusion as claimed in claim 1, wherein the image feature classification module (6) is configured to classify each feature according to the image features extracted by the image feature extraction module (5), select representative and discriminative image features as classification bases, the image classification model construction module (7) is configured to construct a classification model based on image feature classification, and the classification model constructed by the image classification model construction module (7) includes a main body model and a plurality of strut models.
6. The remote sensing image classification method based on multi-model fusion of claim 1, wherein the image classification model building module (7) comprises a classification mode making module (701), a classification category determining module (702) and a classification model building module (703), an output end of the classification mode making module (701) is electrically connected with the classification category determining module (702), and an output end of the classification category determining module (702) is electrically connected with the classification model building module (703).
7. The remote sensing image classification method based on multi-model fusion as claimed in claim 6, characterized in that the classification mode making module (701) is configured to select a suitable image classification algorithm according to classification requirements and characteristics of image data, the classification category determining module (702) is configured to determine categories of image classification, and the classification model building module (703) is configured to build an image classification model according to different classification categories on the basis of determining the image classification categories.
8. The remote sensing image classification method based on multi-model fusion as claimed in claim 7, wherein the image classification model building module (7) further comprises a model training module (704) and a model optimization module (705), an output end of the model training module (704) is electrically connected with the model optimization module (705), the model training module (704) is used for performing model training through remote sensing image data, and the model optimization module (705) evaluates classification results of the built image classification model on the basis of the model training, further optimizes and modifies the classification results, and obtains a more accurate classification model after repeated cycles.
9. The remote sensing image classification method based on multi-model fusion as claimed in claim 1, wherein the classification precision detection module (8) is used for comparing known training data and classification classes with classification results to confirm classification precision and reliability, the classification result inspection module (9) is used for performing statistical inspection on statistical results of discriminant classification, and the image classification execution module (10) is used for classifying remote sensing images through an image classification model.
10. The remote sensing image classification method based on multi-model fusion as claimed in claim 1, wherein the image classification execution module (10) comprises a remote sensing image acquisition module (1001), a remote sensing image processing module (1002), a remote sensing image input module (1003) and a classification result output module (1004), the output end of the remote sensing image acquisition module (1001) is electrically connected with the remote sensing image processing module (1002), the output end of the remote sensing image processing module (1002) is electrically connected with the remote sensing image input module (1003), the output end of the remote sensing image input module (1003) is electrically connected with the classification result output module (1004), the remote sensing image acquisition module (1001) is used for collecting data of a to-be-classified remote sensing image, the remote sensing image processing module (1002) is used for processing and analyzing the remote sensing image, and comprises image radiation correction, geometric correction and filtering processing to obtain a high-precision remote sensing image, the remote sensing image input module (1003) is used for inputting the remote sensing image to the image classification model, and the classification result output module (1004) is used for outputting an image classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211465646.8A CN115761346A (en) | 2022-11-22 | 2022-11-22 | Remote sensing image classification method based on multi-model fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211465646.8A CN115761346A (en) | 2022-11-22 | 2022-11-22 | Remote sensing image classification method based on multi-model fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115761346A true CN115761346A (en) | 2023-03-07 |
Family
ID=85334970
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211465646.8A Pending CN115761346A (en) | 2022-11-22 | 2022-11-22 | Remote sensing image classification method based on multi-model fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115761346A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263705A (en) * | 2019-06-19 | 2019-09-20 | 上海交通大学 | Towards two phase of remote sensing technology field high-resolution remote sensing image change detecting method |
| CN113392748A (en) * | 2021-06-07 | 2021-09-14 | 中国煤炭地质总局勘查研究总院 | Remote sensing image farmland information extraction method based on convolutional neural network |
| WO2021184891A1 (en) * | 2020-03-20 | 2021-09-23 | 中国科学院深圳先进技术研究院 | Remotely-sensed image-based terrain classification method, and system |
| CN114170462A (en) * | 2021-12-06 | 2022-03-11 | 哈尔滨理工大学 | Fine-grained remote sensing ship open set identification method based on convolutional neural network |
| CN114255403A (en) * | 2021-12-21 | 2022-03-29 | 中国人民解放军战略支援部队信息工程大学 | Optical remote sensing image data processing method and system based on deep learning |
| CN114511787A (en) * | 2020-11-17 | 2022-05-17 | 上海湃星信息科技有限公司 | Neural network-based remote sensing image ground feature information generation method and system |
| CN115205649A (en) * | 2022-08-02 | 2022-10-18 | 山东交通学院 | Convolution neural network remote sensing target matching method based on fusion local features |
-
2022
- 2022-11-22 CN CN202211465646.8A patent/CN115761346A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263705A (en) * | 2019-06-19 | 2019-09-20 | 上海交通大学 | Towards two phase of remote sensing technology field high-resolution remote sensing image change detecting method |
| WO2021184891A1 (en) * | 2020-03-20 | 2021-09-23 | 中国科学院深圳先进技术研究院 | Remotely-sensed image-based terrain classification method, and system |
| CN114511787A (en) * | 2020-11-17 | 2022-05-17 | 上海湃星信息科技有限公司 | Neural network-based remote sensing image ground feature information generation method and system |
| CN113392748A (en) * | 2021-06-07 | 2021-09-14 | 中国煤炭地质总局勘查研究总院 | Remote sensing image farmland information extraction method based on convolutional neural network |
| CN114170462A (en) * | 2021-12-06 | 2022-03-11 | 哈尔滨理工大学 | Fine-grained remote sensing ship open set identification method based on convolutional neural network |
| CN114255403A (en) * | 2021-12-21 | 2022-03-29 | 中国人民解放军战略支援部队信息工程大学 | Optical remote sensing image data processing method and system based on deep learning |
| CN115205649A (en) * | 2022-08-02 | 2022-10-18 | 山东交通学院 | Convolution neural network remote sensing target matching method based on fusion local features |
Non-Patent Citations (2)
| Title |
|---|
| 张超 等: "大沽河流域北部山区土壤侵蚀状况研究", 《长春大学学报》 * |
| 龚希 等: "融合全局和局部深度特征的高分辨率遥感影像场景分类方法", 《光明学报》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109165623B (en) | Rice disease spot detection method and system based on deep learning | |
| CN110598736A (en) | Power equipment infrared image fault positioning, identifying and predicting method | |
| CN108711148B (en) | Tire defect intelligent detection method based on deep learning | |
| CN110222767B (en) | Three-dimensional point cloud classification method based on nested neural network and grid map | |
| CN114998220B (en) | Tongue image detection and positioning method based on improved Tiny-YOLO v4 natural environment | |
| CN112149758B (en) | Hyperspectral open set classification method based on Euclidean distance and deep learning | |
| CN114564982A (en) | Automatic identification method for radar signal modulation type | |
| CN112633382A (en) | Mutual-neighbor-based few-sample image classification method and system | |
| CN110059765B (en) | Intelligent mineral identification and classification system and method | |
| CN110731773A (en) | abnormal electrocardiogram screening method based on fusion of global and local depth features of electrocardiogram | |
| CN111079861A (en) | Power distribution network voltage abnormity diagnosis method based on image rapid processing technology | |
| CN111896495A (en) | Method and system for discriminating Taiping Houkui production places based on deep learning and near infrared spectrum | |
| CN107895136A (en) | A kind of colliery area recognizing method and system | |
| CN111160389A (en) | Lithology identification method based on fusion of VGG | |
| CN115147363A (en) | Image defect detection and classification method and system based on deep learning algorithm | |
| CN111967361A (en) | Emotion detection method based on baby expression recognition and crying | |
| CN109447009B (en) | Hyperspectral image classification method based on subspace nuclear norm regularization regression model | |
| CN115761346A (en) | Remote sensing image classification method based on multi-model fusion | |
| CN114662594B (en) | Target feature recognition analysis system | |
| CN115661042A (en) | Hierarchical classification defect detection method based on attention mechanism guidance | |
| CN115830302A (en) | Multi-scale feature extraction and fusion power distribution network equipment positioning identification method | |
| CN115880472A (en) | Intelligent diagnosis and analysis system for electric power infrared image data | |
| CN115757365A (en) | Multi-dimensional time sequence data anomaly detection method, model training method and device | |
| CN111259981B (en) | Automatic classification system after remote sensing image processing | |
| CN112466389A (en) | Method and system for obtaining tumor marker based on machine learning algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230307 |
|
| RJ01 | Rejection of invention patent application after publication |