CN113807278A - Deep learning-based land use classification and change prediction method - Google Patents
Deep learning-based land use classification and change prediction method Download PDFInfo
- Publication number
- CN113807278A CN113807278A CN202111115256.3A CN202111115256A CN113807278A CN 113807278 A CN113807278 A CN 113807278A CN 202111115256 A CN202111115256 A CN 202111115256A CN 113807278 A CN113807278 A CN 113807278A
- Authority
- CN
- China
- Prior art keywords
- land
- land use
- classification
- change
- 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
- 230000008859 change Effects 0.000 title claims abstract description 105
- 238000013135 deep learning Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 62
- 238000006243 chemical reaction Methods 0.000 claims abstract description 38
- 230000011218 segmentation Effects 0.000 claims abstract description 29
- 238000013145 classification model Methods 0.000 claims abstract description 24
- 230000007246 mechanism Effects 0.000 claims abstract description 14
- 230000002123 temporal effect Effects 0.000 claims abstract 4
- 238000012795 verification Methods 0.000 claims description 28
- 238000010586 diagram Methods 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 17
- 238000002372 labelling Methods 0.000 claims description 16
- 230000001413 cellular effect Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000001939 inductive effect Effects 0.000 description 4
- 230000010354 integration Effects 0.000 description 4
- 230000008520 organization Effects 0.000 description 3
- 230000002265 prevention Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
- G06F18/295—Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Computing Systems (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Biophysics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
本发明提供了一种基于深度学习的土地利用分类及变化预测方法，该方法包括：制作土地覆盖类型训练样本及对应的土地利用类型训练样本，构建深度学习语义分割网络模型，基于门控机制构建类型转换网络模型，并对其进行训练，根据深度学习语义分割网络模型及类型转换网络模型生成土地利用分类模型，根据土地利用分类模型得到土地利用分类图像，选取驱动因子构建预测模型，根据土地利用分类图像，分析归纳不同时期的土地利用时空变化特征和规律，并利用预测模型，对未来土地利用变化进行预测。本发明提供的基于深度学习的土地利用分类及变化预测方法，能够准确的实现土地利用分类及变化预测，能够为开展土地利用动态变化预测等一系列工作提供基础。
The present invention provides a land use classification and change prediction method based on deep learning. The method includes: making land cover type training samples and corresponding land use type training samples, constructing a deep learning semantic segmentation network model, and constructing based on a gate control mechanism. Type conversion network model, and train it, generate land use classification model according to deep learning semantic segmentation network model and type conversion network model, obtain land use classification image according to land use classification model, select driving factors to build prediction model, according to land use classification model Classify images, analyze and summarize the temporal and spatial change characteristics and laws of land use in different periods, and use prediction models to predict future land use changes. The deep learning-based land use classification and change prediction method provided by the invention can accurately realize land use classification and change prediction, and can provide a basis for a series of works such as land use dynamic change prediction.
Description
Technical Field
The invention relates to the technical field of remote sensing image information, in particular to a land use classification and change prediction method based on deep learning.
Background
The land use classification technology based on the high-resolution remote sensing image is widely applied to land use investigation work in a large-scale range, but the existing land use classification technology still has the problems of difficult extraction of land feature classification characteristics, more noise interference factors, insufficiently fine classification results and the like; in addition, the land use types are various and the division basis is complex, and part of the categories are composed of various different land features, so that the internal structure is complex, and the land use types with complex composition cannot be accurately classified by a classification method generally depending on the characteristics of remote sensing images. The existing land use classification and change prediction are not combined, and when the land use change prediction is carried out, a common classification technology is often adopted for classification, so that the prediction precision is low, and a series of work such as dynamic change prediction of land use, crop yield prediction, natural disaster prevention and control, reasonable organization of land use and the like cannot be carried out. Therefore, it is necessary to design a land use classification and change prediction method based on deep learning.
Disclosure of Invention
The invention aims to provide a land use classification and change prediction method based on deep learning, which can accurately realize land use classification and change prediction and provide a basis for developing a series of work such as dynamic change prediction of land use, crop yield prediction, natural disaster prevention and control, reasonable organization of land use and the like.
In order to achieve the purpose, the invention provides the following scheme:
a land use classification and change prediction method based on deep learning comprises the following steps:
step 1: acquiring historical remote sensing image data and corresponding land vector data, and making a land cover type training sample and a corresponding land utilization type training sample;
step 2: constructing a deep learning semantic segmentation network model, training the deep learning semantic segmentation network model through a land cover type training sample, constructing a type conversion network model based on a gating mechanism, and training the type conversion network model through a land utilization type training sample;
and step 3: performing series integration on the trained deep learning semantic segmentation network model and the type conversion network model to generate a land utilization classification model;
and 4, step 4: inputting the remote sensing images of the land to be detected in different periods into a land utilization classification model to obtain a land utilization classification image;
and 5: drawing a classification vector diagram according to the land utilization classification image, calculating land change amplitude, dynamic degree and space change according to the classification vector diagram, analyzing the dynamic change of the land in different periods according to the land change amplitude, the dynamic degree and the space change, and inducing the space-time change characteristics and rules of the land utilization in different periods;
step 6: selecting a driving factor to construct a CA-Markov prediction model, acquiring verification land remote sensing image data, and verifying the precision and the applicability of the CA-Markov prediction model;
and 7: and if the verification is passed, predicting the future land utilization change by using a CA-Markov prediction model according to the characteristics and rules of the land utilization space-time change in different periods.
Optionally, in step 1, obtaining historical remote sensing image data and corresponding land vector data, and making a land cover type training sample and a corresponding land use type training sample, specifically:
collecting historical remote sensing image data, corresponding land utilization vector data and historical land cover vector data, obtaining boundary areas of the vector data through a scanning line algorithm, carrying out vector rasterization processing on the vector data to generate an initial land utilization vector marking base map and an initial historical land cover vector marking base map, carrying out individual marking on representative artificial land objects in the initial land utilization vector marking base map and the initial historical land cover vector marking base map through artificial marking to obtain a land utilization vector marking base map and a historical land cover vector marking base map, carrying out slicing processing on the historical remote sensing image data and the corresponding historical land cover vector marking base map to generate land cover type training samples with standard sizes, and carrying out slicing processing on the historical land cover vector marking base map and the corresponding land utilization vector marking base map, standard land use type training samples are generated.
Optionally, in step 2, constructing a deep learning semantic segmentation network model specifically includes:
and (3) constructing a deep learning semantic segmentation model based on deep learning by using the parallel high-resolution feature extraction network model as a basic network.
Optionally, in step 2, a type conversion network model is constructed based on a gating mechanism, and the type conversion network model is trained through a land use type training sample, specifically:
the method comprises the steps of constructing a geographic space incidence relation feature extraction unit based on a gating mechanism, taking the geographic space incidence relation feature unit as a basic unit, introducing an attention module to establish a type conversion network model, dividing land utilization type training samples into lines, arranging the lines in sequence from top to bottom to form a first group of image sequences, forming a second group of image sequences in sequence from bottom to top, dividing the land utilization type training samples into lines, arranging the lines in sequence from left to right to form a third group of image sequences, forming a fourth group of image sequences in sequence from right to left, and inputting the four groups of image sequences into the type conversion network model for training.
Optionally, in step 4, inputting the remote sensing images of the land to be detected in different periods into the land use classification model to obtain a land use classification image, which specifically comprises:
and acquiring remote sensing images of the land to be detected in different periods, and inputting the remote sensing images into the land utilization classification model to obtain a land utilization classification image.
Optionally, in step 6, a driving factor is selected to construct a CA-Markov prediction model, which specifically comprises:
selecting a plurality of driving factors, generating a distribution probability schematic diagram of a land use type according to the driving factors, obtaining a conversion rule of a cellular automaton CA model according to the distribution probability schematic diagram of the land use type, and constructing a CA-Markov prediction model according to the conversion rules of the Markov model and the CA model.
Optionally, in step 6, obtaining verification land remote sensing image data, and verifying the accuracy and the applicability of the CA-Markov prediction model, specifically:
obtaining a verified land remote sensing image and actual land utilization change data, inputting the verified land remote sensing image into a land utilization classification model to obtain a verified land utilization classification image, drawing a verification classification vector diagram according to the verification land utilization classification image, calculating land variation amplitude, dynamic degree and spatial variation according to the verification classification vector diagram, and analyzes the dynamic change of the land in different periods through the land change amplitude, the dynamic degree and the space change, induces the space-time change characteristics and rules of the land utilization in different periods, and predicting land use change by using a CA-Markov prediction model according to the characteristics and rules of the land use time-space change to obtain verification land use change data, comparing the verification land use change data with the actual land use change data, and judging that the verification is passed if the verification is within a reasonable error range.
Optionally, the driving factors include at least elevation, slope, direction of slope, distance from highway and distance from administrative center.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: the invention provides a land use classification and change prediction method based on deep learning, which comprises the steps of obtaining historical remote sensing image data and corresponding land vector data, making a land cover type training sample and a corresponding land use type training sample, constructing a deep learning semantic segmentation network model, training the deep learning semantic segmentation network model through the land cover type training sample, constructing a type conversion network model based on a gating mechanism, training the type conversion network model through the land use type training sample, performing series integration on the trained deep learning semantic segmentation network model and the type conversion network model to generate the land use classification model, inputting land remote sensing images to be detected in different periods into the land use classification model to obtain land use classification images, drawing a classification vector diagram according to the land use classification images, calculating land change amplitude, dynamic degree and space change according to the classification vector diagram, analyzing dynamic change of land in different periods through the land change amplitude, the dynamic degree and the space change, inducing land utilization space-time change characteristics and rules in different periods, selecting a driving factor to construct a CA-Markov prediction model, and predicting future land utilization change by using the CA-Markov prediction model according to the land utilization space-time change characteristics and rules in different periods; the classification of land utilization is realized based on deep learning, a deep learning semantic segmentation technology is adopted, classification model training is carried out based on historical remote sensing image data and corresponding historical land cover vector labeling base maps to obtain pixel-level historical land cover vector labeling base maps which are accurately classified, a geographic space incidence relation feature extraction unit is constructed based on a gating mechanism to extract geographic space incidence relations of various land utilization types, the segmentation results of the land cover types can be merged into the same land utilization type, and the conversion from the land cover ground object type classification base maps to the land utilization classification labeling maps is completed; and a driving factor is introduced, a CA-Markov prediction model is constructed based on a Markov model and a CA model, the land utilization change is accurately analyzed and predicted, and the CA-Markov prediction model is verified by verifying land remote sensing image data, so that the precision and the applicability of the CA-Markov prediction model are ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
Fig. 1 is a schematic flow chart of a land use classification and change prediction method based on deep learning according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a land use classification and change prediction method based on deep learning, which can accurately realize land use classification and change prediction and provide a basis for developing a series of work such as dynamic change prediction of land use, crop yield prediction, natural disaster prevention and control, reasonable organization of land use and the like.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a schematic flow chart of a deep learning-based land use classification and change prediction method according to an embodiment of the present invention, and as shown in fig. 1, the deep learning-based land use classification and change prediction method according to the embodiment of the present invention includes the following steps:
step 1: acquiring historical remote sensing image data and corresponding land vector data, and making a land cover type training sample and a corresponding land utilization type training sample;
step 2: constructing a deep learning semantic segmentation network model, training the deep learning semantic segmentation network model through a land cover type training sample, constructing a type conversion network model based on a gating mechanism, and training the type conversion network model through a land utilization type training sample;
and step 3: performing series integration on the trained deep learning semantic segmentation network model and the type conversion network model to generate a land utilization classification model;
and 4, step 4: inputting the remote sensing images of the land to be detected in different periods into a land utilization classification model to obtain a land utilization classification image;
and 5: drawing a classification vector diagram according to the land utilization classification image, calculating land change amplitude, dynamic degree and space change according to the classification vector diagram, analyzing the dynamic change of the land in different periods according to the land change amplitude, the dynamic degree and the space change, and inducing the space-time change characteristics and rules of the land utilization in different periods;
step 6: selecting a driving factor to construct a CA-Markov prediction model, acquiring verification land remote sensing image data, and verifying the precision and the applicability of the CA-Markov prediction model;
and 7: and if the verification is passed, predicting the future land utilization change by using a CA-Markov prediction model according to the characteristics and rules of the land utilization space-time change in different periods.
In the step 1, obtaining historical remote sensing image data and corresponding land vector data, and making a land cover type training sample and a corresponding land utilization type training sample, specifically:
collecting historical remote sensing image data, corresponding land utilization vector data and historical land cover vector data, obtaining boundary areas of the vector data through a scanning line algorithm, carrying out vector rasterization processing on the vector data to generate an initial land utilization vector annotation base map and an initial historical land cover vector annotation base map, carrying out individual annotation on representative artificial ground objects in the initial land utilization vector annotation base map and the initial historical land cover vector annotation base map through artificial annotation, wherein the representative artificial ground objects are, for example, a playground, a stadium, a square and the like to obtain the land utilization vector annotation base map and the historical land cover vector annotation base map, carrying out slicing processing on the historical remote sensing image data and the corresponding historical land cover vector annotation base map to generate a land cover type training sample with standard size, and slicing the historical land cover vector labeling base map and the corresponding land utilization vector labeling base map to generate a standard land utilization type training sample.
When the training samples are manufactured, the proportion of each land cover type or land utilization type on each sample image is counted, sample screening is carried out according to the counting result, and a training set with balanced categories is constructed.
In step 2, constructing a deep learning semantic segmentation network model, specifically:
based on deep learning, a parallel high-resolution feature extraction network model is used as a basic network to construct a deep learning semantic segmentation model, and when the deep learning semantic segmentation model is trained through a land cover type training sample, reasonable model training parameters are set, so that the deep learning semantic segmentation model can automatically extract classification features, and pixel-level land cover classification is realized.
In step 2, a type conversion network model is constructed based on a gating mechanism, and the type conversion network model is trained through a land use type training sample, specifically comprising the following steps:
constructing a geographic space incidence relation feature extraction unit based on a gate control mechanism, taking a geographic space incidence relation feature unit as a basic unit, introducing an attention module to establish a type conversion network model, wherein the type conversion network model is divided into a coding part, an attention module and a decoding part, the coding part comprises a plurality of geographic space incidence relation feature extraction units, the plurality of geographic space incidence relation feature extraction units form a plurality of feature extraction layers, the extracted features are subjected to feature cross-layer combination by adopting a cross-layer feature combination module among different feature extraction layers, the attention module is connected with the decoding part and the coding part, weights of different land cover ground objects in various land utilization types are learned through the interior of the attention module, the decoding part comprises a plurality of geographic space incidence relation feature extraction units, and the plurality of geographic space incidence relation feature extraction units form a plurality of feature extraction layers, performing characteristic cross-layer combination on the extracted characteristics by adopting a cross-layer characteristic combination module between different characteristic extraction layers, and outputting a final classification result by an output layer through a softmax function;
dividing the land use type training samples according to lines, arranging the training samples from top to bottom to form a first group of image sequences, forming a second group of image sequences according to the sequence from bottom to top, dividing the land use type training samples according to lines, arranging the training samples from left to right to form a third group of image sequences, forming a fourth group of image sequences according to the sequence from right to left, and inputting the four groups of image sequences into a type conversion network model for training.
In step 4, inputting the remote sensing images of the land to be detected in different periods into a land utilization classification model to obtain a land utilization classification image, which specifically comprises the following steps:
and acquiring remote sensing images of the land to be detected in different periods, and inputting the remote sensing images into the land utilization classification model to obtain a land utilization classification image.
The land use type of the land use classification image is generally grassland, cultivated land, forest land, water area, construction land, unused land, and the like.
The characteristics and the rules of the land utilization space-time change are general land utilization overall change, change amplitude, dynamic degree, change space and the like.
In the step 6, a driving factor is selected to construct a CA-Markov prediction model, which specifically comprises the following steps:
selecting a plurality of driving factors, generating a distribution probability schematic diagram of a land use type according to the driving factors, obtaining a conversion rule of a cellular automaton CA model according to the distribution probability schematic diagram of the land use type, and constructing a CA-Markov prediction model according to the conversion rules of the Markov model and the CA model.
In step 6, obtaining and verifying land remote sensing image data, and verifying the precision and the applicability of the CA-Markov prediction model, specifically:
obtaining a verified land remote sensing image and actual land utilization change data, inputting the verified land remote sensing image into a land utilization classification model to obtain a verified land utilization classification image, drawing a verification classification vector diagram according to the verification land utilization classification image, calculating land variation amplitude, dynamic degree and spatial variation according to the verification classification vector diagram, and analyzes the dynamic change of the land in different periods through the land change amplitude, the dynamic degree and the space change, induces the space-time change characteristics and rules of the land utilization in different periods, and predicting land use change by using a CA-Markov prediction model according to the characteristics and rules of the land use time-space change to obtain verification land use change data, comparing the verification land use change data with the actual land use change data, and judging that the verification is passed if the verification is within a reasonable error range.
The driving factors at least comprise elevation, gradient, slope, distance from the highway and distance from the administrative center.
The invention provides a land use classification and change prediction method based on deep learning, which comprises the steps of obtaining historical remote sensing image data and corresponding land vector data, making a land cover type training sample and a corresponding land use type training sample, constructing a deep learning semantic segmentation network model, training the deep learning semantic segmentation network model through the land cover type training sample, constructing a type conversion network model based on a gating mechanism, training the type conversion network model through the land use type training sample, performing series integration on the trained deep learning semantic segmentation network model and the type conversion network model to generate the land use classification model, inputting land remote sensing images to be detected in different periods into the land use classification model to obtain land use classification images, drawing a classification vector diagram according to the land use classification images, calculating land change amplitude, dynamic degree and space change according to the classification vector diagram, analyzing dynamic change of land in different periods through the land change amplitude, the dynamic degree and the space change, inducing land utilization space-time change characteristics and rules in different periods, selecting a driving factor to construct a CA-Markov prediction model, and predicting future land utilization change by using the CA-Markov prediction model according to the land utilization space-time change characteristics and rules in different periods; the classification of land utilization is realized based on deep learning, a deep learning semantic segmentation technology is adopted, classification model training is carried out based on historical remote sensing image data and corresponding historical land cover vector labeling base maps to obtain pixel-level historical land cover vector labeling base maps which are accurately classified, a geographic space incidence relation feature extraction unit is constructed based on a gating mechanism to extract geographic space incidence relations of various land utilization types, the segmentation results of the land cover types can be merged into the same land utilization type, and the conversion from the land cover ground object type classification base maps to the land utilization classification labeling maps is completed; and a driving factor is introduced, a CA-Markov prediction model is constructed based on a Markov model and a CA model, the land utilization change is accurately analyzed and predicted, and the CA-Markov prediction model is verified by verifying land remote sensing image data, so that the precision and the applicability of the CA-Markov prediction model are ensured.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111115256.3A CN113807278A (en) | 2021-09-23 | 2021-09-23 | Deep learning-based land use classification and change prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111115256.3A CN113807278A (en) | 2021-09-23 | 2021-09-23 | Deep learning-based land use classification and change prediction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113807278A true CN113807278A (en) | 2021-12-17 |
Family
ID=78940277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111115256.3A Pending CN113807278A (en) | 2021-09-23 | 2021-09-23 | Deep learning-based land use classification and change prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113807278A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114565178A (en) * | 2022-03-14 | 2022-05-31 | 江苏恒月地理信息技术有限公司 | Temporal data monitoring and management method based on geographic information survey system |
| CN115545334A (en) * | 2022-10-25 | 2022-12-30 | 石河子大学 | Land use type prediction method, land use type prediction device, electronic device, and storage medium |
| CN117541940A (en) * | 2024-01-10 | 2024-02-09 | 日照市自然资源和规划局 | Land utilization classification method and system based on remote sensing data |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458371A (en) * | 2019-08-21 | 2019-11-15 | 中国地质大学（北京） | Method and device for analyzing and predicting land use change |
| CN112070078A (en) * | 2020-11-16 | 2020-12-11 | 武汉思众空间信息科技有限公司 | Deep learning-based land utilization classification method and system |
-
2021
- 2021-09-23 CN CN202111115256.3A patent/CN113807278A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458371A (en) * | 2019-08-21 | 2019-11-15 | 中国地质大学（北京） | Method and device for analyzing and predicting land use change |
| CN112070078A (en) * | 2020-11-16 | 2020-12-11 | 武汉思众空间信息科技有限公司 | Deep learning-based land utilization classification method and system |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114565178A (en) * | 2022-03-14 | 2022-05-31 | 江苏恒月地理信息技术有限公司 | Temporal data monitoring and management method based on geographic information survey system |
| CN114565178B (en) * | 2022-03-14 | 2024-06-04 | 江苏恒月地理信息技术有限公司 | Temporal data monitoring and management method based on geographic information survey system |
| CN115545334A (en) * | 2022-10-25 | 2022-12-30 | 石河子大学 | Land use type prediction method, land use type prediction device, electronic device, and storage medium |
| CN115545334B (en) * | 2022-10-25 | 2024-04-02 | 石河子大学 | Land utilization type prediction method and device, electronic equipment and storage medium |
| CN117541940A (en) * | 2024-01-10 | 2024-02-09 | 日照市自然资源和规划局 | Land utilization classification method and system based on remote sensing data |
| CN117541940B (en) * | 2024-01-10 | 2024-03-22 | 日照市自然资源和规划局 | Land utilization classification method and system based on remote sensing data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Unsupervised machine learning in urban studies: A systematic review of applications | |
| CN110136170B (en) | Remote sensing image building change detection method based on convolutional neural network | |
| CN110245709B (en) | 3D point cloud data semantic segmentation method based on deep learning and self-attention | |
| CN113807278A (en) | Deep learning-based land use classification and change prediction method | |
| CN112070078B (en) | Land use classification method and system based on deep learning | |
| CN109614985A (en) | A target detection method based on densely connected feature pyramid network | |
| CN112446591A (en) | Evaluation system for student comprehensive capacity evaluation and zero sample evaluation method | |
| CN106529721B (en) | A kind of ad click rate forecasting system and its prediction technique that depth characteristic is extracted | |
| CN110533631A (en) | SAR image change detection based on the twin network of pyramid pondization | |
| Zeng et al. | Revisiting the modifiable areal unit problem in deep traffic prediction with visual analytics | |
| CN109800631A (en) | Fluorescence-encoded micro-beads image detecting method based on masked areas convolutional neural networks | |
| Shen et al. | Visual interpretation of recurrent neural network on multi-dimensional time-series forecast | |
| CN109117883A (en) | SAR image sea ice classification method and system based on long memory network in short-term | |
| CN111709318B (en) | High-resolution remote sensing image classification method based on generation countermeasure network | |
| CN116681176B (en) | A traffic flow prediction method based on clustering and heterogeneous graph neural network | |
| CN111639587A (en) | Hyperspectral image classification method based on multi-scale spectrum space convolution neural network | |
| Li et al. | Learning to holistically detect bridges from large-size vhr remote sensing imagery | |
| CN114021837A (en) | Regional Electricity Consumption Prediction Method Based on Hybrid Machine Learning and Spatial Address Matching | |
| CN114898089A (en) | Functional area extraction and classification method integrating high-resolution images and POI data | |
| CN110796360A (en) | A multi-scale data fusion method for stationary traffic detection sources | |
| Palaniyappan Velumani et al. | AQX: Explaining air quality forecast for verifying domain knowledge using feature importance visualization | |
| KR102492843B1 (en) | Apparatus and method for analyzing spatio temporal data for geo-location | |
| Mohammadi et al. | A framework for classification of volunteered geographic data based on user’s need | |
| CN114639015B (en) | Landslide detection method based on high-resolution images combining spectrum, vegetation index and texture features | |
| CN119441613A (en) | Intelligent recommendation system for urban monitoring points based on spatiotemporal collaborative analysis of meteorological data |
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 |