CN117114627A - land resource management system - Google Patents
land resource management system Download PDFInfo
- Publication number
- CN117114627A CN117114627A CN202311345629.5A CN202311345629A CN117114627A CN 117114627 A CN117114627 A CN 117114627A CN 202311345629 A CN202311345629 A CN 202311345629A CN 117114627 A CN117114627 A CN 117114627A
- Authority
- CN
- China
- Prior art keywords
- feature map
- feature
- soil type
- neural network
- land
- 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.)
- Withdrawn
Links
- 239000002689 soil Substances 0.000 claims abstract description 140
- 238000012876 topography Methods 0.000 claims abstract description 69
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 29
- 230000007704 transition Effects 0.000 claims abstract description 22
- 238000006073 displacement reaction Methods 0.000 claims abstract description 21
- 235000013399 edible fruits Nutrition 0.000 claims abstract description 15
- 230000007246 mechanism Effects 0.000 claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims description 51
- 239000013598 vector Substances 0.000 claims description 51
- 238000013528 artificial neural network Methods 0.000 claims description 35
- 230000004927 fusion Effects 0.000 claims description 35
- 238000012545 processing Methods 0.000 claims description 35
- 238000010586 diagram Methods 0.000 claims description 18
- 230000004913 activation Effects 0.000 claims description 15
- 238000011176 pooling Methods 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 8
- 238000000034 method Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 6
- 230000008707 rearrangement Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 230000000903 blocking effect Effects 0.000 claims description 2
- 238000000638 solvent extraction Methods 0.000 abstract description 6
- 238000003860 storage Methods 0.000 description 11
- 238000004590 computer program Methods 0.000 description 9
- 230000008901 benefit Effects 0.000 description 8
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 235000021049 nutrient content Nutrition 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000035558 fertility Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- 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
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- 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
- G06N3/0455—Auto-encoder networks; Encoder-decoder 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Human Resources & Organizations (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Multimedia (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Databases & Information Systems (AREA)
Abstract
The application relates to the field of land resource management, and particularly discloses a land resource management system, which comprises the steps of firstly acquiring land topography images acquired by cameras and soil type text data acquired from a soil database, then fusing the land topography images acquired by the cameras through a first convolutional neural network and a second convolutional neural network using a spatial attention mechanism to obtain a topography feature map, then partitioning the soil type text data acquired from the soil database, then obtaining a soil type feature map through a context encoder comprising an embedded layer and a third convolutional neural network, and finally performing order-based displacement transition on the topography feature map and the soil type feature map, and judging whether resources of each area of the land are suitable for planting fruits through a classifier, thereby providing scientific basis for fruit planting decisions and further improving the efficiency of land resource management.
Description
Technical Field
The present application relates to the field of land resource management, and more particularly, to a land resource management system.
Background
The land resource management system can integrate and analyze a large amount of land related data, including data of soil texture, land topography and the like. By analyzing the data, the system can evaluate the suitability of the land and help agricultural decision makers make scientific decisions.
In the prior art, agricultural decision makers decide on crops planted in the land based on personal subjective judgment and experience, and such decisions may be subject to subjective bias and limitation, and lack of scientific basis and systematic analysis of decisions may result in selection of unsuitable crops, thereby affecting agricultural production benefits. Meanwhile, when an agricultural decision maker makes a decision, only single factors such as soil fertility conditions or climate conditions are often considered, and the comprehensive influence of other key factors is ignored, so that the overall potential and limitation of the land may not be fully considered by the one-sided decision making, and the scientific and effective decision is not enough.
Thus, a more optimal land resource management system is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a land resource management system, which comprises the steps of firstly acquiring land topography images acquired by a camera and soil type text data acquired from a soil database, then fusing the land topography images acquired by the camera through a first convolution neural network and a second convolution neural network using a space attention mechanism to obtain a topography feature map, then partitioning the soil type text data acquired from the soil database, then obtaining a soil type feature map through a context encoder containing an embedded layer and a third convolution neural network, finally performing order-based displacement transition on the topography feature map and the soil type feature map, and judging whether resources of each area of the land are suitable for planting fruits through a classifier, thereby providing scientific basis for fruit planting decisions and further improving the efficiency of land resource management.
According to an aspect of the present application, there is provided a land resource management system, comprising:
the data acquisition module is used for acquiring the land topography image acquired by the camera and the soil type text data acquired from the soil database;
the first convolution module is used for enabling the land topography image acquired by the camera to pass through a first convolution neural network to obtain a first feature map, wherein the first convolution neural network comprises a deep-shallow feature fusion module;
a second spatial attention module for passing the land topography image acquired by the camera through a second convolutional neural network using a spatial attention mechanism to obtain a second feature map;
the feature fusion module is used for fusing the first feature map and the second feature map to generate a topography feature map;
the context coding module is used for blocking the soil type text data which is called from the soil database and then obtaining a plurality of third feature vectors through a context coder comprising an embedded layer;
the third convolution module is used for obtaining a soil type feature map through a third convolution neural network after the plurality of third feature vectors are arranged in two dimensions;
The fusion module is used for carrying out displacement transition based on order on the topography feature map and the soil type feature map so as to obtain a classification feature map;
and the classification module is used for passing the classification characteristic diagram through a classifier to obtain a classification result, wherein the classification result is used for indicating whether resources of each area of the land are suitable for planting fruits.
In the land resource management system, the first convolution module comprises a shallow feature extraction unit, a first convolution module and a second convolution module, wherein the shallow feature extraction unit is used for obtaining a shallow feature map from an M-th layer of the first convolution neural network, and M is more than or equal to 1 and less than or equal to 6; the deep feature map extraction unit is used for obtaining a deep feature map from the last layer of the first convolutional neural network; and the association unit is used for associating the shallow feature map and the deep feature map through a deep-shallow feature fusion module of the first convolutional neural network so as to obtain the first feature map.
In the above land resource management system, the association unit includes: using the deep-shallow feature fusion module to correlate the shallow feature map and the deep feature map with a correlation formula to obtain the first feature map, wherein the correlation formula is: Wherein (1)>For the first profile, +_>For the shallow feature map, ++>For the deep feature map, ">"means that the elements at the corresponding positions of the shallow feature map and the deep feature map are added,">For controlling the shallow layer in the first characteristic diagramWeighting parameters for the balance between the feature map and the deep feature map.
In the above land resource management system, the second spatial attention module includes: each layer of the second convolutional neural network model performs input data in forward transfer of the layer: a convolution unit, configured to perform a convolution process based on a two-dimensional convolution kernel on the input data to generate a convolution feature map; the pooling unit is used for pooling the convolution feature graphs to generate pooled feature graphs; the activation unit is used for carrying out activation processing on the pooled feature map so as to generate an activated feature map; the global average pooling unit is used for carrying out global average pooling along the channel dimension on the activation feature map so as to obtain a space feature matrix; the weight unit is used for carrying out convolution processing and activation processing on the space feature matrix to generate a weight vector; the weighting unit is used for respectively weighting each feature matrix of the activated feature map by using the weight value of each position in the weight vector so as to obtain a generated feature map; and the generated feature map output by the last layer of the second convolutional neural network model is the second feature map.
In the above land resource management system, the context encoding module includes: the block unit is used for carrying out block processing on the soil type text data which are called from the soil database to obtain word sequences corresponding to the soil types; a conversion unit, configured to convert each word in the word sequence of each soil type into a word embedding vector by using an embedding layer of the context encoder, so as to obtain a sequence of word embedding vectors corresponding to each soil type; a context semantic coding unit, configured to perform context semantic coding on a sequence of word embedding vectors of each soil type by using a Bert model based on a converter of the context encoder to obtain a plurality of word feature vectors corresponding to each soil type; and the word characteristic cascading unit is used for cascading a plurality of word characteristic vectors of each soil type to obtain third characteristic vectors corresponding to each soil type.
In the above land resource management system, the third convolution module includes: two-dimensionally arranging the plurality of third feature vectors to obtain a third feature matrix; and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transmission of layers by using each layer of the third convolution neural network so as to output the soil type feature map by the last layer of the third convolution neural network, wherein the input of the third convolution neural network is the third feature matrix.
In the above land resource management system, the fusion module includes: based on the global average value of each feature matrix of the topographic and topographic feature map along the channel dimension, carrying out feature matrix serialization rearrangement on the topographic and topographic feature map to obtain a serialized topographic and topographic feature map; based on the global average value of each feature matrix of the soil type feature map along the channel dimension, carrying out feature matrix serialization rearrangement on the soil type feature map to obtain a serialized soil type feature map; calculating Euclidean distance between feature matrixes of any two channel dimensions in the serialized topography feature map and the serialized soil type feature map to obtain a displacement transition topology matrix; calculating a position-based mean matrix between feature matrices of each group of corresponding channel dimensions of the serialized topography feature map and the serialized soil type feature map to obtain a sequence of fused feature matrices; and the sequence of the fusion feature matrix and the displacement transition topology matrix pass through a graph neural network model to obtain the classification feature graph.
In the above land resource management system, the classification module includes: processing the classification feature map using the classifier in the following classification formula to obtain the classification result;
Wherein, the classification formula is:wherein->Representing projection of the classification feature map as a vector, < >>Is a weight matrix>Representing the bias vector +_>Representing normalized exponential function, ++>Representing the classification result.
According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory having stored therein computer program instructions that, when executed by the processor, cause the processor to perform the land resource management system as described above.
According to a further aspect of the present application there is provided a computer readable medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform a land resource management system as described above.
Compared with the prior art, the land resource management system provided by the application has the advantages that firstly, the land topography image acquired by the camera and the soil type text data acquired from the soil database are acquired, then, the land topography image acquired by the camera is fused after passing through the first convolution neural network and the second convolution neural network using the spatial attention mechanism respectively to obtain the topography feature map, then, the soil type text data acquired from the soil database are segmented and then pass through the context encoder comprising the embedded layer and the third convolution neural network to obtain the soil type feature map, finally, the topography feature map and the soil type feature map are subjected to order-based displacement transition and pass through the classifier to judge whether the resources of each area of the land are suitable for planting fruits, so that scientific basis is provided for fruit planting decisions, and the efficiency of land resource management is further improved.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
Fig. 1 is a block diagram of a land resource management system according to an embodiment of the present application.
Fig. 2 is a block diagram of a land resource management system according to an embodiment of the present application.
Fig. 3 is a block diagram of a first convolution module in a land resource management system according to an embodiment of the present application.
Fig. 4 is a block diagram of a second spatial attention module in a land resource management system according to an embodiment of the present application.
Fig. 5 is a block diagram of a context encoding module in a land resource management system according to an embodiment of the present application.
Fig. 6 is a block diagram of an electronic device according to an embodiment of the application.
Detailed Description
Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Exemplary System
Fig. 1 is a block diagram of a land resource management system according to an embodiment of the present application. As shown in fig. 1, a land resource management system 100 according to an embodiment of the present application includes: a data acquisition module 110 for acquiring a land topography image acquired by a camera and soil type text data retrieved from a soil database; a first convolution module 120, configured to pass the land topography image acquired by the camera through a first convolution neural network to obtain a first feature map, where the first convolution neural network includes a deep-shallow feature fusion module; a second spatial attention module 130 for passing the land topography image acquired by the camera through a second convolutional neural network using a spatial attention mechanism to obtain a second feature map; a feature fusion module 140, configured to fuse the first feature map and the second feature map to generate a topography feature map; the context coding module 150 is configured to block the soil type text data retrieved from the soil database, and obtain a plurality of third feature vectors through a context encoder including an embedded layer; the third convolution module 160 is configured to two-dimensionally arrange the plurality of third feature vectors, and then obtain a soil type feature map through a third convolution neural network; the fusion module 170 is configured to perform order-based displacement transition on the topography feature map and the soil type feature map to obtain a classification feature map; and the classification module 180 is used for passing the classification characteristic diagram through a classifier to obtain a classification result, wherein the classification result is used for indicating whether resources of each area of the land are suitable for planting fruits.
Fig. 2 is a block diagram of a land resource management system according to an embodiment of the present application. As shown in fig. 2, the land resource management system according to an embodiment of the present application includes: first, a land topography image acquired by a camera and soil type text data called from a soil database are acquired. And then, the land topography image acquired by the camera passes through a first convolution neural network to obtain a first characteristic map, wherein the first convolution neural network comprises a deep-shallow characteristic fusion module. The camera-acquired land topography image is then passed through a second convolutional neural network using a spatial attention mechanism to obtain a second feature map. Further, the first feature map and the second feature map are fused to generate a topography feature map. Next, the soil type text data retrieved from the soil database is segmented and then passed through a context encoder comprising an embedded layer to obtain a plurality of third feature vectors. And further, the plurality of third eigenvectors are subjected to two-dimensional arrangement and then pass through a third convolutional neural network to obtain a soil type characteristic map. Specifically, the topography feature map and the soil type feature map are subjected to order-based displacement transition to obtain a classification feature map. And finally, the classification characteristic diagram is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the resources of each area of the land are suitable for planting fruits.
In the above-described land resource management system 100, the data acquisition module 110 is configured to acquire a land topography image acquired by a camera and soil type text data retrieved from a soil database. It will be appreciated that the land topography images may provide information about land surface morphology, elevation, slope, etc., which have important effects on the growth and development of crops, for example, land with flatter terrain and moderate slope may be easier to mechanically plant and irrigate for management, while land with steep terrain may require measures to prevent water and soil loss, and thus the land topography images acquired by the cameras may help the system to analyze and understand the morphological features of the land. The soil types are classified on physical, chemical and biological characteristics of the soil, different soil types have the characteristics of different water holding capacity, nutrient content, air permeability and the like, the soil types have direct influence on the growth and development of crops, and the system can obtain the classification information of the soil through the text data of the soil types which are called from a soil database, so that the soil quality and characteristics of the soil can be further known. The land feature and soil type information are comprehensively considered, the suitability of the land for planting fruits can be more comprehensively evaluated, the morphological feature of the land is considered, and factors such as the soil texture, nutrient content and drainage condition are considered, so that more land information can be provided by acquiring land topography images and soil type text data, farmers or agricultural managers can be helped to make more accurate planting decisions, and the utilization of land resources is optimized.
In the land resource management system 100, the first convolution module 120 is configured to pass the land topography image acquired by the camera through a first convolution neural network to obtain a first feature map, where the first convolution neural network includes a deep-shallow feature fusion module. Here, it should be appreciated by those skilled in the art that the convolutional neural network model has excellent performance in the field of image feature extraction, and in particular, in the technical solution of the present application, since capturing a change pattern of a land topography image acquired by a camera in a time dimension is expected, a time attention mechanism is introduced into the first convolutional neural network model. It should be appreciated that convolutional neural networks are widely used in the field of image processing to effectively capture local and global features in an image. The first convolutional neural network comprises a deep-shallow feature fusion module, so that feature fusion is carried out among different levels of the network, the deep-shallow feature fusion module is used for combining feature graphs from the different levels, feature representation capacity of the different levels is comprehensively considered, features in the land topography image can be better extracted and expressed through the deep-shallow feature fusion, and the understanding and classifying capacity of the system on land resources is enhanced.
Fig. 3 is a block diagram of a first convolution module in a land resource management system according to an embodiment of the present application. As shown in fig. 3, in a specific embodiment of the present application, the first convolution module 120 includes a shallow feature extraction unit 121, configured to obtain a shallow feature map from an mth layer of the first convolution neural network, where M is greater than or equal to 1 and less than or equal to 6; a deep feature map extracting unit 122, configured to obtain a deep feature map from a last layer of the first convolutional neural network; and the associating unit 123 is configured to associate the shallow feature map and the deep feature map by using a deep-shallow feature fusion module of the first convolutional neural network to obtain the first feature map.
In a specific embodiment of the present application, the association unit 123 includes: using the deep-shallow feature fusion module to correlate the shallow feature map and the deep feature map with a correlation formula to obtain the first feature map, wherein the correlation formula is:wherein (1)>For the first profile, +_>For the shallow feature map, ++>For the deep feature map, ">"means that elements at corresponding positions of the shallow feature map and the deep feature map are added, Is a weighting parameter for controlling the balance between the shallow feature map and the deep feature map in the first feature map.
In the land resource management system 100, the second spatial attention module 130 is configured to obtain the second feature map by using a second convolutional neural network of a spatial attention mechanism for the land topography image acquired by the camera. It will be appreciated that the spatial attention mechanism is a mechanism that helps the network to focus more on important areas and features when processing images, and by introducing the spatial attention mechanism, the second convolutional neural network can automatically learn and adjust feature weights at different locations when processing land topography images, focusing more on areas that are helpful in determining land suitability. Through the application of the spatial attention mechanism, the second convolution neural network can be focused on the characteristics related to the land suitability in the land image, such as the texture, the shape, the structure and the like of the land, so that the perception and understanding capability of the system on the land characteristics can be improved, and the land suitability can be judged better.
Fig. 4 is a block diagram of a second spatial attention module in a land resource management system according to an embodiment of the present application. As shown in fig. 4, in a specific embodiment of the present application, the second spatial attention module 130 includes: each layer of the second convolutional neural network model performs input data in forward transfer of the layer: a convolution unit 131 configured to perform a convolution process based on a two-dimensional convolution kernel on the input data to generate a convolution feature map; a pooling unit 132, configured to perform pooling processing on the convolution feature map to generate a pooled feature map; an activation unit 133, configured to perform activation processing on the pooled feature map to generate an activated feature map; a pooling unit 134, configured to perform global average pooling along a channel dimension on the activation feature map to obtain a spatial feature matrix; a weight unit 135 for performing convolution processing and activation processing on the spatial feature matrix to generate a weight vector; a weighting unit 136, configured to weight each feature matrix of the activated feature map with a weight value of each position in the weight vector to obtain a generated feature map; and the generated feature map output by the last layer of the second convolutional neural network model is the second feature map.
In the land resource management system 100, the feature fusion module 140 is configured to fuse the first feature map and the second feature map to generate a topography feature map. It should be understood that the advantages of the first feature map and the second feature map can be combined by fusing the first feature map and the second feature map, and the fused topographic and geomorphic feature map synthesizes visual information and key features of the image, so that more complete and comprehensive land feature representation is provided, the suitability of the land can be accurately judged by the system, and more reliable planting decision basis can be provided.
In the land resource management system 100, the context encoding module 150 is configured to block the text data of the soil type retrieved from the soil database, and obtain a plurality of third feature vectors by using a context encoder including an embedded layer. It should be appreciated that the purpose of partitioning the soil type text data is to better process the long text sequence and capture the semantic relationships between the different parts, by partitioning the long text sequence into a plurality of short sequences, each of which is passed through a context encoder to obtain a feature vector. Each feature vector after the segmentation represents semantic information and contextual features of the corresponding text block, which can represent different parts of soil type text data more finely, capture richer semantic information, and provide more feature representations for subsequent processing and fusion. Soil type text data is typically in the form of natural language containing information about the description, characteristics, and attributes of the soil type, however, natural language typically requires conversion into a numerical vector representation in computer processing for further processing and analysis, in order to convert the soil type text data into a vector representation, a combination of an embedding layer that maps discrete words into a continuous vector space, each word into a dense vector, and a context encoder that uses these embedded vectors and takes into account the context of the words, encodes the entire text sequence into a continuous vector representation may be used.
Fig. 5 is a block diagram of a context encoding module in a land resource management system according to an embodiment of the present application. As shown in fig. 5, in a specific embodiment of the present application, the context encoding module 150 includes: a partitioning unit 151 for performing a partitioning process on the soil type text data retrieved from the soil database to obtain word sequences corresponding to the respective soil types; a conversion unit 152, configured to convert each word in the word sequence of each soil type into a word embedding vector by using an embedding layer of the context encoder to obtain a sequence of word embedding vectors corresponding to each soil type; a context semantic coding unit 153, configured to perform context semantic coding on a sequence of word embedding vectors of each soil type by using a Bert model based on a converter of the context encoder to obtain a plurality of word feature vectors corresponding to each soil type; a word feature concatenation unit 154 for concatenating the plurality of word feature vectors for each of the soil types to obtain a third feature vector corresponding to each of the soil types.
In the land resource management system 100, the third convolution module 160 is configured to two-dimensionally arrange the plurality of third feature vectors and then obtain a soil type feature map through a third convolution neural network. It should be appreciated that the two-dimensional matrix may dimensionally align the soil type feature vector with the image feature map for subsequent fusion and processing, and then process the soil type feature map through a third convolutional neural network, which may extract and learn a feature representation of the soil type, and the convolutional neural network has a strong feature extraction capability when processing the two-dimensional data, and may capture spatial correlation and structural information of the soil type feature.
In a specific embodiment of the present application, the third convolution module 160 includes: two-dimensionally arranging the plurality of third feature vectors to obtain a third feature matrix; and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transmission of layers by using each layer of the third convolution neural network so as to output the soil type feature map by the last layer of the third convolution neural network, wherein the input of the third convolution neural network is the third feature matrix.
In the land resource management system 100, the fusion module 170 is configured to perform order-based shift transition on the topography map and the soil type map to obtain a classification map. It should be understood that, in the technical solution of the present application, noise and abnormal values exist in the topography feature map and the soil type feature map, and on the other hand, the topography feature map and the soil type feature map are different phase feature characterizations of the same entity, so in the technical solution of the present application, order-based displacement transition is performed on the topography feature map and the soil type feature map, so that order information of data in an original feature space and associated information of data in the original feature space are utilized to perform order-based displacement transition fusion on the topography feature map and the soil type feature map to obtain the classification feature map, and in this way, noise and abnormal values in the data can be effectively eliminated, and meanwhile, a relative relationship between the data is maintained, thereby improving robustness and accuracy of the classifier.
Specifically, in the technical scheme of the application, firstly, the relative sequence relation between semantic information amounts of each feature matrix of the topographic and topographic feature map and the soil type feature map along the channel dimension is utilized to carry out serialization reconstruction on the topographic and topographic feature map and the soil type feature map so as to optimize the alignment degree between the topographic and topographic feature map and the feature structure of the soil type feature map to obtain the serialized topographic and topographic feature map and the serialized soil type feature map. Further, euclidean distance between feature matrixes of any two channel dimensions in the serialized topography feature map and the serialized soil type feature map is calculated to obtain a displacement transition topology matrix, so that the displacement transition feature value of order between any two feature matrixes of the serialized topography feature map and the serialized soil type feature map is measured. And representing a feature level initial association feature representation between the serialized topography feature map and the serialized soil type feature map with a per-position mean feature map between the serialized topography feature map and the serialized soil type feature map. And then, topological fusion is carried out on the sequence of the fusion feature matrix and the displacement transition topology matrix by using a graph neural network model so as to obtain the classification feature graph.
In this way, through carrying out displacement transition based on order on the topography feature map and the soil type feature map, the relative distance and the topological structure between each feature matrix between the topography feature map and the soil type feature map can be reserved, so that the intrinsic mode and the clustering of the data are reflected, and meanwhile, the abnormal points and the outlier points in the data can be highlighted by the classification feature map obtained through fusion.
In a specific embodiment of the present application, the fusion module 170 includes: performing order-based displacement transition on the topography feature map and the soil type feature map to obtain a classification feature map, comprising: based on the global average value of each feature matrix of the topographic and topographic feature map along the channel dimension, carrying out feature matrix serialization rearrangement on the topographic and topographic feature map to obtain a serialized topographic and topographic feature map; based on the global average value of each feature matrix of the soil type feature map along the channel dimension, carrying out feature matrix serialization rearrangement on the soil type feature map to obtain a serialized soil type feature map; calculating Euclidean distance between feature matrixes of any two channel dimensions in the serialized topography feature map and the serialized soil type feature map to obtain a displacement transition topology matrix; calculating a position-based mean matrix between feature matrices of each group of corresponding channel dimensions of the serialized topography feature map and the serialized soil type feature map to obtain a sequence of fused feature matrices; and the sequence of the fusion feature matrix and the displacement transition topology matrix pass through a graph neural network model to obtain the classification feature graph.
In the land resource management system 100, the classification module 180 is configured to pass the classification feature map through a classifier to obtain a classification result, where the classification result is used to indicate whether the resources of each area of land are suitable for planting fruits.
In a specific embodiment of the present application, the classification module 180 includes: processing the classification feature map using the classifier in the following classification formula to obtain the classification result;
wherein, the classification formula is:wherein->Representing projection of the classification feature map as a vector, < >>Is a weight matrix>Representing the bias vector +_>Representing normalized exponential function, ++>Representing the classification result.
In summary, the embodiment of the application firstly acquires the land topography and topography image acquired by the camera and the soil type text data acquired from the soil database, then respectively fuses the land topography and topography image acquired by the camera through a first convolution neural network and a second convolution neural network using a spatial attention mechanism to obtain a topography and topography feature map, then blocks the soil type text data acquired from the soil database, then obtains a soil type feature map through a context encoder containing an embedded layer and a third convolution neural network, finally, carries out order-based displacement transition on the topography feature map and the soil type feature map, and judges whether resources of each area of the land are suitable for planting fruits through a classifier, thereby providing scientific basis for fruit planting decisions and further improving the efficiency of land resource management.
As described above, the land resource management system 100 according to the embodiment of the present application may be implemented in various terminal devices, for example, a server or the like where a land resource management control algorithm is deployed. In one example, land resource management system 100 may be integrated into a terminal device as a software module and/or hardware module. For example, the land resource management system 100 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the land resource management system 100 could equally be one of a number of hardware modules of the terminal device.
Alternatively, in another example, the land resource management system 100 and the terminal device may be separate devices, and the land resource management system 100 may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information in a agreed data format.
Exemplary electronic device
Next, an electronic device according to an embodiment of the present application is described with reference to fig. 6.
Fig. 6 illustrates a block diagram of an electronic device according to an embodiment of the application.
As shown in fig. 6, the electronic device 10 includes one or more processors 11 and a memory 12.
The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that may be executed by processor 11 to perform the land resource management and/or other desired functions of the various embodiments of the application described above. Various contents such as a land topography image acquired by a camera and soil type text data retrieved from a soil database may also be stored in the computer readable storage medium.
In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).
The input means 13 may comprise, for example, a keyboard, a mouse, etc.
The output device 14 can output various information to the outside, including whether the resources of each area of the land are suitable for planting fruits, etc. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device 10 that are relevant to the present application are shown in fig. 6 for simplicity, components such as buses, input/output interfaces, etc. are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer readable storage Medium
In addition to the methods and apparatus described above, embodiments of the application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps of a land resource management system according to the various embodiments of the application described in the "exemplary systems" section of this specification.
The computer program product may write program code for performing operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the application may also be a computer-readable storage medium, on which computer program instructions are stored which, when executed by a processor, cause the processor to perform the steps for a land resource management system according to the various embodiments of the application described in the "exemplary systems" section of this specification.
The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.
The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (8)
1. A land resource management system, comprising:
the data acquisition module is used for acquiring the land topography image acquired by the camera and the soil type text data acquired from the soil database;
The first convolution module is used for enabling the land topography image acquired by the camera to pass through a first convolution neural network to obtain a first feature map, wherein the first convolution neural network comprises a deep-shallow feature fusion module;
a second spatial attention module for passing the land topography image acquired by the camera through a second convolutional neural network using a spatial attention mechanism to obtain a second feature map;
the feature fusion module is used for fusing the first feature map and the second feature map to generate a topography feature map;
the context coding module is used for blocking the soil type text data which is called from the soil database and then obtaining a plurality of third feature vectors through a context coder comprising an embedded layer;
the third convolution module is used for obtaining a soil type feature map through a third convolution neural network after the plurality of third feature vectors are arranged in two dimensions;
the fusion module is used for carrying out displacement transition based on order on the topography feature map and the soil type feature map so as to obtain a classification feature map;
and the classification module is used for passing the classification characteristic diagram through a classifier to obtain a classification result, wherein the classification result is used for indicating whether resources of each area of the land are suitable for planting fruits.
2. The land resource management system of claim 1, wherein said first convolution module comprises:
the shallow feature extraction unit is used for obtaining a shallow feature map from an M-th layer of the first convolutional neural network, wherein M is more than or equal to 1 and less than or equal to 6;
the deep feature map extraction unit is used for obtaining a deep feature map from the last layer of the first convolutional neural network;
and the association unit is used for associating the shallow feature map and the deep feature map through a deep-shallow feature fusion module of the first convolutional neural network so as to obtain the first feature map.
3. The land resource management system of claim 2, wherein said association unit comprises:
using the deep-shallow feature fusion module to correlate the shallow feature map and the deep feature map with a correlation formula to obtain the first feature map, wherein the correlation formula is:wherein (1)>For the first profile, +_>For the shallow feature map, ++>For the deep feature map, ">"means that the elements at the corresponding positions of the shallow feature map and the deep feature map are added,">Is a weighting parameter for controlling the balance between the shallow feature map and the deep feature map in the first feature map.
4. A land resource management system as claimed in claim 3, wherein said second spatial attention module comprises:
each layer of the second convolutional neural network model performs input data in forward transfer of the layer:
a convolution unit, configured to perform a convolution process based on a two-dimensional convolution kernel on the input data to generate a convolution feature map;
the pooling unit is used for pooling the convolution feature graphs to generate pooled feature graphs;
the activation unit is used for carrying out activation processing on the pooled feature map so as to generate an activated feature map;
the global average pooling unit is used for carrying out global average pooling along the channel dimension on the activation feature map so as to obtain a space feature matrix;
the weight unit is used for carrying out convolution processing and activation processing on the space feature matrix to generate a weight vector;
the weighting unit is used for respectively weighting each feature matrix of the activated feature map by using the weight value of each position in the weight vector so as to obtain a generated feature map;
and the generated feature map output by the last layer of the second convolutional neural network model is the second feature map.
5. The land resource management system of claim 4, wherein said context encoding module comprises:
The block unit is used for carrying out block processing on the soil type text data which are called from the soil database to obtain word sequences corresponding to the soil types;
a conversion unit, configured to convert each word in the word sequence of each soil type into a word embedding vector by using an embedding layer of the context encoder, so as to obtain a sequence of word embedding vectors corresponding to each soil type;
a context semantic coding unit, configured to perform context semantic coding on a sequence of word embedding vectors of each soil type by using a Bert model based on a converter of the context encoder to obtain a plurality of word feature vectors corresponding to each soil type;
and the word characteristic cascading unit is used for cascading a plurality of word characteristic vectors of each soil type to obtain third characteristic vectors corresponding to each soil type.
6. The land resource management system of claim 5, wherein said third convolution module comprises:
two-dimensionally arranging the plurality of third feature vectors to obtain a third feature matrix;
and respectively carrying out convolution processing, mean pooling processing based on a feature matrix and nonlinear activation processing on input data in forward transmission of layers by using each layer of the third convolution neural network so as to output the soil type feature map by the last layer of the third convolution neural network, wherein the input of the third convolution neural network is the third feature matrix.
7. The land resource management system of claim 6, wherein said fusion module comprises:
based on the global average value of each feature matrix of the topographic and topographic feature map along the channel dimension, carrying out feature matrix serialization rearrangement on the topographic and topographic feature map to obtain a serialized topographic and topographic feature map;
based on the global average value of each feature matrix of the soil type feature map along the channel dimension, carrying out feature matrix serialization rearrangement on the soil type feature map to obtain a serialized soil type feature map;
calculating Euclidean distance between feature matrixes of any two channel dimensions in the serialized topography feature map and the serialized soil type feature map to obtain a displacement transition topology matrix;
calculating a position-based mean matrix between feature matrices of each group of corresponding channel dimensions of the serialized topography feature map and the serialized soil type feature map to obtain a sequence of fused feature matrices;
and the sequence of the fusion feature matrix and the displacement transition topology matrix pass through a graph neural network model to obtain the classification feature graph.
8. The land resource management system of claim 7, wherein said classification module comprises:
Processing the classification feature map using the classifier in the following classification formula to obtain the classification result;
wherein, the classification formula is:wherein->Representing projection of the classification feature map as a vector, < >>Is a weight matrix>Representing the bias vector +_>Representing normalized exponential function, ++>Representing the classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311345629.5A CN117114627A (en) | 2023-10-18 | 2023-10-18 | land resource management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311345629.5A CN117114627A (en) | 2023-10-18 | 2023-10-18 | land resource management system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117114627A true CN117114627A (en) | 2023-11-24 |
Family
ID=88805756
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311345629.5A Withdrawn CN117114627A (en) | 2023-10-18 | 2023-10-18 | land resource management system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117114627A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117743975A (en) * | 2024-02-21 | 2024-03-22 | 君研生物科技(山西)有限公司 | Hillside cultivated land soil environment improvement method |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170351933A1 (en) * | 2016-06-01 | 2017-12-07 | Intel Corporation | Vision enhanced drones for precision farming |
| CN107766828A (en) * | 2017-10-29 | 2018-03-06 | 北京工业大学 | UAV Landing Geomorphological Classification method based on wavelet convolution neutral net |
| US20190050948A1 (en) * | 2017-08-08 | 2019-02-14 | Indigo Ag, Inc. | Machine learning in agricultural planting, growing, and harvesting contexts |
| CN109522942A (en) * | 2018-10-29 | 2019-03-26 | 中国科学院深圳先进技术研究院 | A kind of image classification method, device, terminal device and storage medium |
| CN111985694A (en) * | 2020-07-29 | 2020-11-24 | 北京派得伟业科技发展有限公司 | Planting planning system |
| CN112906300A (en) * | 2021-02-09 | 2021-06-04 | 北京化工大学 | Polarized SAR (synthetic Aperture Radar) soil humidity inversion method based on two-channel convolutional neural network |
| CN214579998U (en) * | 2021-04-16 | 2021-11-02 | 日照市自然资源和规划局山海天旅游度假区分局 | Municipal works topography survey and drawing auxiliary assembly |
| KR20220095342A (en) * | 2020-12-29 | 2022-07-07 | 아주대학교산학협력단 | The diagnostic method and system of lymph node metastasis in thyroid cancer using ct image |
| CN115169728A (en) * | 2022-07-28 | 2022-10-11 | 长春嘉诚信息技术股份有限公司 | Soil fertility prediction method based on simplified neural network |
| CN115860581A (en) * | 2023-02-28 | 2023-03-28 | 北京市农林科学院信息技术研究中心 | Method, device, equipment and storage medium for evaluating suitability of crop variety |
| CN116030296A (en) * | 2022-11-17 | 2023-04-28 | 江西东煜智能科技有限公司 | Social platform data mining method and system for graphic data collaboration |
| CN116448019A (en) * | 2023-06-14 | 2023-07-18 | 山西首科工程质量检测有限公司 | Intelligent detection device and method for quality flatness of building energy-saving engineering |
| CN116721389A (en) * | 2023-07-28 | 2023-09-08 | 吉林隆源农业服务有限公司 | Crop planting management method |
| CN116740567A (en) * | 2023-06-13 | 2023-09-12 | 平顶山学院 | Soil moisture content detection method and system for Paulownia seedling cultivation |
| CN116782466A (en) * | 2023-07-15 | 2023-09-19 | 上饶市美宇达光学仪器有限公司 | LOGO lamp for vehicle and imaging method thereof |
| CN116840447A (en) * | 2023-07-07 | 2023-10-03 | 华南农业大学 | Soil compactness detection method based on multiple sensors and neural network |
| US20230316555A1 (en) * | 2020-08-14 | 2023-10-05 | Agriculture Victoria Services Pty Ltd | System and Method for Image-Based Remote Sensing of Crop Plants |
-
2023
- 2023-10-18 CN CN202311345629.5A patent/CN117114627A/en not_active Withdrawn
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170351933A1 (en) * | 2016-06-01 | 2017-12-07 | Intel Corporation | Vision enhanced drones for precision farming |
| US20190050948A1 (en) * | 2017-08-08 | 2019-02-14 | Indigo Ag, Inc. | Machine learning in agricultural planting, growing, and harvesting contexts |
| CN107766828A (en) * | 2017-10-29 | 2018-03-06 | 北京工业大学 | UAV Landing Geomorphological Classification method based on wavelet convolution neutral net |
| CN109522942A (en) * | 2018-10-29 | 2019-03-26 | 中国科学院深圳先进技术研究院 | A kind of image classification method, device, terminal device and storage medium |
| CN111985694A (en) * | 2020-07-29 | 2020-11-24 | 北京派得伟业科技发展有限公司 | Planting planning system |
| US20230316555A1 (en) * | 2020-08-14 | 2023-10-05 | Agriculture Victoria Services Pty Ltd | System and Method for Image-Based Remote Sensing of Crop Plants |
| KR20220095342A (en) * | 2020-12-29 | 2022-07-07 | 아주대학교산학협력단 | The diagnostic method and system of lymph node metastasis in thyroid cancer using ct image |
| CN112906300A (en) * | 2021-02-09 | 2021-06-04 | 北京化工大学 | Polarized SAR (synthetic Aperture Radar) soil humidity inversion method based on two-channel convolutional neural network |
| CN214579998U (en) * | 2021-04-16 | 2021-11-02 | 日照市自然资源和规划局山海天旅游度假区分局 | Municipal works topography survey and drawing auxiliary assembly |
| CN115169728A (en) * | 2022-07-28 | 2022-10-11 | 长春嘉诚信息技术股份有限公司 | Soil fertility prediction method based on simplified neural network |
| CN116030296A (en) * | 2022-11-17 | 2023-04-28 | 江西东煜智能科技有限公司 | Social platform data mining method and system for graphic data collaboration |
| CN115860581A (en) * | 2023-02-28 | 2023-03-28 | 北京市农林科学院信息技术研究中心 | Method, device, equipment and storage medium for evaluating suitability of crop variety |
| CN116740567A (en) * | 2023-06-13 | 2023-09-12 | 平顶山学院 | Soil moisture content detection method and system for Paulownia seedling cultivation |
| CN116448019A (en) * | 2023-06-14 | 2023-07-18 | 山西首科工程质量检测有限公司 | Intelligent detection device and method for quality flatness of building energy-saving engineering |
| CN116840447A (en) * | 2023-07-07 | 2023-10-03 | 华南农业大学 | Soil compactness detection method based on multiple sensors and neural network |
| CN116782466A (en) * | 2023-07-15 | 2023-09-19 | 上饶市美宇达光学仪器有限公司 | LOGO lamp for vehicle and imaging method thereof |
| CN116721389A (en) * | 2023-07-28 | 2023-09-08 | 吉林隆源农业服务有限公司 | Crop planting management method |
Non-Patent Citations (3)
| Title |
|---|
| ZHICHENG HUANG ，ET AL.: "Multigraph Convolutional Networks for Rainfall Estimation in Complex Terrain", IEEE, vol. 19, pages 327 - 331 * |
| 董洪义 等: "智能机器人技术：安保、巡逻、处置类警用机器人研究实践", 北京：机械工业出版社, pages: 112 - 114 * |
| 蔡浩杰 等: "基于地形特征融合的卷积神经网络滑坡识别", 地球科学与环境学报, vol. 44, no. 03 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117743975A (en) * | 2024-02-21 | 2024-03-22 | 君研生物科技(山西)有限公司 | Hillside cultivated land soil environment improvement method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shanthamallu et al. | A brief survey of machine learning methods and their sensor and IoT applications | |
| CN107767408B (en) | Image processing method, processing device and processing equipment | |
| CN110796199B (en) | Image processing method and device and electronic medical equipment | |
| CN117114627A (en) | land resource management system | |
| CN111738280A (en) | Image identification method, device, equipment and readable storage medium | |
| Xia et al. | A deep Siamese postclassification fusion network for semantic change detection | |
| CN112183672A (en) | Image classification method, and training method and device of feature extraction network | |
| CN112242002B (en) | Object identification and panoramic roaming method based on deep learning | |
| CN112818849B (en) | Crowd density detection algorithm based on context attention convolutional neural network for countermeasure learning | |
| CN114676777A (en) | Self-supervision learning fine-grained image classification method based on twin network | |
| CN115013298A (en) | Real-time performance on-line monitoring system and monitoring method of sewage pump | |
| CN111062431A (en) | Image clustering method, image clustering device, electronic device, and storage medium | |
| CN112733710A (en) | Method for training a neural network for irrigation water pressure control of an irrigation device | |
| Xu et al. | Combining random forest and support vector machines for object-based rural-land-cover classification using high spatial resolution imagery | |
| Zhaosheng et al. | Rapid detection of wheat ears in orthophotos from unmanned aerial vehicles in fields based on YOLOX | |
| CN112419342A (en) | Image processing method, image processing device, electronic equipment and computer readable medium | |
| CN115424264A (en) | Panorama segmentation method, related device, electronic equipment and storage medium | |
| Xia et al. | Cloud/snow recognition of satellite cloud images based on multiscale fusion attention network | |
| Li et al. | An automatic plant leaf stoma detection method based on YOLOv5 | |
| CN111709766B (en) | User behavior prediction method and device, storage medium and electronic equipment | |
| CN116851856B (en) | Pure waterline cutting processing technology and system thereof | |
| CN110751201B (en) | SAR equipment task failure cause reasoning method based on textural feature transformation | |
| CN117036933A (en) | Intelligent crop pest control system and method | |
| CN115620083B (en) | Model training method, face image quality evaluation method, equipment and medium | |
| CN116610735A (en) | Intelligent management method and system for data storage |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20231124 |
|
| WW01 | Invention patent application withdrawn after publication |