CN111476197A - Oil palm identification and area extraction method and system based on multi-source satellite remote sensing image - Google Patents
Oil palm identification and area extraction method and system based on multi-source satellite remote sensing image Download PDFInfo
- Publication number
- CN111476197A CN111476197A CN202010332738.3A CN202010332738A CN111476197A CN 111476197 A CN111476197 A CN 111476197A CN 202010332738 A CN202010332738 A CN 202010332738A CN 111476197 A CN111476197 A CN 111476197A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- sensing image
- oil palm
- data
- image data
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The method and the system are based on oil palm identification and area extraction of the multisource satellite remote sensing image, the multisource satellite remote sensing image acquisition unit acquires at least two satellite remote sensing image data of a target area through a multisource satellite, and the data are subjected to standardized processing to generate multisource remote sensing image data; the image segmentation unit carries out multilevel gridding segmentation processing on the multi-source remote sensing image data to obtain gridding remote sensing image data; the model selection unit collects sample vector data, acquires corresponding sample data in the grid remote sensing image data, and evaluates and selects the deep learning model by using the sample data to obtain an optimal deep learning model; the oil palm extraction unit inputs the grid remote sensing image data into an optimal deep learning model, identifies and segments oil palms in the grid remote sensing image data and counts the area of the oil palms; and the output unit is used for outputting and displaying the oil palm area statistical result. The oil palm identification, extraction and area statistics can be carried out on multi-source remote sensing image data or regions.
Description
Technical Field
The invention relates to the technical field of oil palm planting, growing and management, in particular to a method and a system for oil palm identification and area extraction based on a multi-source satellite remote sensing image.
Background
The area of the oil palm is always important reference information and decision-making basis concerned by international organizations, national governments, industrial institutions, large oil palm planting institutions and enterprises, oil palm pressing plants, bulk oil traders and bulk oil futures.
The planting organization needs the oil palm area to carry out planting planning and management, the international organization, the national government and the industry organization need the oil palm planting area to carry out unified coordination and planning on important natural resources, the degradation, felling and damage of tropical rain forest are monitored, important reference of sustainable development policy is made, and the oil palm distribution and area are important basis for making oil palm product yield judgment by a large number of oil traders and futures.
At present, the method for obtaining the area of the oil palm is mainly based on the plan or the statistics of the field measurement of various planting areas or the reclamation of the area of the oil palm arranged by ground investigators by planting institutions and enterprises or government institutions, and the method needs a large amount of personnel, is long in time consumption, slow in updating, low in precision and complex in implementation process, and is difficult to implement large-scale coordination and uniform drawing. Therefore, the method brings serious information loss when the oil palm area needs to be known with high reliability, rapidness, high precision and large range, so that further planting and management planning decision making or deeper correlation analysis by using the oil palm area is difficult.
Therefore, the problems of the prior art are to be further improved and developed.
Disclosure of Invention
The object of the invention is: in order to solve the problems in the prior art, the invention aims to provide an artificial intelligent identification and extraction method and system for multisource satellite remote sensing images with rapid oil palm area, high precision, large range and high update rate, so as to solve the problems that a large amount of personnel are needed, the time consumption is long, the update is slow, and large-range coordinated implementation and unified mapping are difficult to perform when ground investigators are arranged to perform field measurement of a planting area based on planting institutions, enterprises or government institutions at present.
The technical scheme is as follows: in order to solve the technical problems, the technical scheme provides a method for identifying and extracting the area of the oil palm based on the multi-source satellite remote sensing image, which comprises the following steps,
step one, a multi-source satellite remote sensing image acquisition unit acquires at least two satellite remote sensing image data of a target area through a multi-source satellite, and carries out standardized processing to generate the multi-source remote sensing image data;
step two, the image segmentation unit carries out multilevel gridding segmentation processing on the multi-source remote sensing image data to obtain gridding remote sensing image data;
thirdly, the model selecting unit collects sample vector data, acquires corresponding sample data in the grid remote sensing image data, and evaluates and selects the deep learning model by using the sample data to obtain an optimal deep learning model;
inputting the grid remote sensing image data into an optimal deep learning model by an oil palm extraction unit, identifying and segmenting oil palms in the grid remote sensing image data and counting the area of the oil palms;
and fifthly, the output unit displays the output of the oil palm area statistical result.
The oil palm identification and area extraction method based on the multisource satellite remote sensing image comprises the following steps:
a3, collecting sample vector data by a data collection module, acquiring corresponding sample data in grid remote sensing image data, and carrying out standardized processing on the sample data;
step b3, the dividing module divides the sample data into training sample data and verification sample data;
step c3, selecting a deep learning semantic segmentation model by a model selection module, and determining model structure parameters of the deep learning semantic segmentation model; carrying out verification training on model structure parameters of the deep learning semantic segmentation model; and evaluating the model structure parameters of the deep learning semantic segmentation model, and selecting the optimal deep learning model.
The oil palm identification and area extraction method based on the multisource satellite remote sensing image comprises the following four steps:
step a4, the oil palm recognition module inputs the grid remote sensing image data into an optimal deep learning model, and oil palm recognition is carried out on the grid remote sensing image data;
and b4, the oil palm statistical module divides the grid remote sensing image data containing the oil palm and counts the area of the oil palm.
Based on multisource satellite remote sensing image oil palm discernment and area extraction system includes:
the multi-source satellite remote sensing image acquisition unit is used for carrying out standardized processing on at least two satellite remote sensing image data of a target area to generate multi-source remote sensing image data;
the image segmentation unit is used for carrying out multi-level gridding segmentation processing on the multi-source remote sensing image data to obtain gridding remote sensing image data;
the model selection unit is used for collecting sample vector data, acquiring corresponding sample data in the grid remote sensing image data, and evaluating and selecting the deep learning model by using the sample data to obtain an optimal deep learning model;
the oil palm extraction unit is used for inputting the grid remote sensing image data into an optimal deep learning model, identifying and segmenting oil palms in the grid remote sensing image data and counting the area of the oil palms;
and the output unit is used for outputting and displaying the oil palm area statistical result.
The oil palm identification and area extraction system based on the multi-source satellite remote sensing image is characterized in that the standardized processing of the multi-source satellite remote sensing image acquisition unit comprises processing the resolution, projection and image enhancement of satellite remote sensing image data according to a unified form.
The oil palm identification and area extraction system based on the multi-source satellite remote sensing image is characterized in that the sample vector data comprises sample vector data marked as oil palms on the ground in the industry, and oil palm sample vector data and non-oil palm sample vector data of different geographic positions and time are collected from the remote sensing image according to the experience of industry experts.
The oil palm identification and area extraction system based on the multisource satellite remote sensing image is characterized in that the model selection unit comprises a data collection module, a division module and a model selection module;
the data collection module is used for collecting sample vector data, acquiring corresponding sample data in the grid remote sensing image data, and carrying out standardized processing on the sample data; the dividing module divides the sample data into training sample data and verification sample data; the model selection module selects a deep learning semantic segmentation model and determines model structure parameters of the deep learning semantic segmentation model; carrying out verification training on model structure parameters of the deep learning semantic segmentation model; and evaluating the model structure parameters of the deep learning semantic segmentation model, and selecting the optimal deep learning model.
The oil palm identification and area extraction system based on the multi-source satellite remote sensing image is characterized in that the sample data standardization processing is to standardize the acquired remote sensing image data block containing the oil palm, and comprises the size of the image block, the proportion of the oil palm and the identification degree of the oil palm on the image.
The oil palm identification and area extraction system based on the multi-source satellite remote sensing image is characterized in that the model structure parameters comprise model parameters and model hyper-parameters, and the model parameters comprise the number of convolution layers, the size of a convolution kernel, the number of pooling layers, a pooling mode and an activation function.
The oil palm identification and area extraction system based on the multi-source satellite remote sensing image is characterized in that the determined model structure parameters of the deep learning semantic segmentation model are used for identifying the oil palm according to the texture structure, the shape, the spectral characteristics and the oil palm cluster characteristics of the oil palm.
The oil palm identification and area extraction system based on the multisource satellite remote sensing image is characterized in that the model selection module evaluates the deep learning semantic segmentation model through identification accuracy, precision, recall rate and an ROC curve, and selects an optimal deep learning model and corresponding parameter configuration.
The oil palm identification and area extraction system based on the multi-source satellite remote sensing image is characterized in that the oil palm extraction unit comprises an oil palm identification module and an oil palm statistical module;
the oil palm recognition module inputs the grid remote sensing image data into an optimal deep learning model and carries out oil palm recognition on the grid remote sensing image data;
the oil palm statistical module is used for segmenting grid remote sensing image data containing oil palms and counting the areas of the oil palms.
The oil palm identification and area extraction system based on the multi-source satellite remote sensing image is characterized in that an oil palm area statistical result comprises oil palm data, and the oil palm data are grid data or vector data with geographic coordinates.
(III) the beneficial effects are as follows: the invention provides a method and a system for identifying and extracting oil palm areas based on multi-source satellite remote sensing images, which can not only identify and extract the oil palm and count the areas of multi-source remote sensing image data or regions, but also save various resources, configuration and coordination of oil palm areas obtained by ground investigation when identifying and extracting the oil palm in a large range.
Drawings
FIG. 1 is a schematic diagram of the steps of the oil palm identification and area extraction method based on the multisource satellite remote sensing image;
FIG. 2 is a schematic diagram of a connection relationship of an oil palm identification and area extraction system based on a multi-source satellite remote sensing image according to an embodiment of the invention;
fig. 3 is a schematic flow chart of oil palm identification and area extraction according to a preferred embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to preferred embodiments, and more details are set forth in the following description in order to provide a thorough understanding of the present invention, but it is apparent that the present invention can be embodied in many other forms different from the description herein and can be similarly generalized and deduced by those skilled in the art based on the practical application without departing from the spirit of the present invention, and therefore, the scope of the present invention should not be limited by the contents of this detailed embodiment.
The drawings are schematic representations of embodiments of the invention, and it is noted that the drawings are intended only as examples and are not drawn to scale and should not be construed as limiting the true scope of the invention.
As shown in fig. 1, the method for oil palm identification and area extraction based on multi-source satellite remote sensing images comprises the following steps,
step one, a multi-source satellite remote sensing image acquisition unit acquires at least two satellite remote sensing image data of a target area through a multi-source satellite, and carries out standardized processing to generate the multi-source remote sensing image data;
step two, the image segmentation unit carries out multilevel gridding segmentation processing on the multi-source remote sensing image data to obtain gridding remote sensing image data;
thirdly, the model selecting unit collects sample vector data, acquires corresponding sample data in the grid remote sensing image data, and evaluates and selects the deep learning model by using the sample data to obtain an optimal deep learning model;
inputting the grid remote sensing image data into an optimal deep learning model by an oil palm extraction unit, identifying and segmenting oil palms in the grid remote sensing image data and counting the area of the oil palms;
and fifthly, the output unit displays the output of the oil palm area statistical result.
The third step further comprises:
a3, collecting sample vector data by a data collection module, acquiring corresponding sample data in grid remote sensing image data, and carrying out standardized processing on the sample data;
step b3, the dividing module divides the sample data into training sample data and verification sample data;
step c3, selecting a deep learning semantic segmentation model by a model selection module, and determining model structure parameters of the deep learning semantic segmentation model; carrying out verification training on model structure parameters of the deep learning semantic segmentation model; and evaluating the model structure parameters of the deep learning semantic segmentation model, and selecting the optimal deep learning model.
The fourth step further comprises:
step a4, the oil palm recognition module inputs the grid remote sensing image data into an optimal deep learning model, and oil palm recognition is carried out on the grid remote sensing image data;
and b4, the oil palm statistical module divides the grid remote sensing image data containing the oil palm and counts the area of the oil palm.
The following is a description of a specific process for oil palm identification and area extraction according to a preferred embodiment of the present application.
As shown in fig. 2 and 3, a system for identifying oil palm and extracting area based on multi-source satellite remote sensing image includes: the device comprises a multi-source satellite remote sensing image acquisition unit, an image segmentation unit, a model selection unit, an oil palm extraction unit and an output unit. The multi-source satellite remote sensing image acquisition unit is sequentially connected with the image segmentation unit, the model selection unit, the oil palm extraction unit and the output unit, wherein the image segmentation unit is also connected with the oil palm extraction unit.
And the multi-source satellite remote sensing image acquisition unit is used for carrying out standardized processing on at least two satellite remote sensing image data of the target area to generate multi-source remote sensing image data.
The satellite remote sensing image data is the satellite remote sensing image data of more than two satellites which aim at a target area and contain the high spatial resolution of the oil palm.
The normalization processing may be processing the satellite remote sensing image data according to a uniform form, wherein the specific form may be a uniform form of resolution, projection, image enhancement, and the like.
The method comprises the steps of carrying out normalization processing on satellite remote sensing image data, preferably carrying out fine-granularity gray level processing on the satellite remote sensing image data and an image picture, preferably simplifying picture resolution to be one third to one fourth of fineness of the shot image picture, carrying out gray level processing on the simplified image picture, obtaining an object contour of the image picture according to the gray level picture of the image picture, and rapidly removing non-oil palm data in the satellite remote sensing image data through comparison of contour pictures of the image pictures with the same image size proportion. The oil palm trees are in vertical tree shapes and are as high as 10 meters or more, the outline of an object is quickly obtained through the outline image of the gray-scale image of the image picture, short plant images are quickly eliminated according to the outline of the object and the height of the object, and the accuracy rate of oil palm identification is improved.
And the image segmentation unit is used for carrying out multi-level gridding segmentation processing on the multi-source remote sensing image data to obtain gridding remote sensing image data so as to ensure that the deep learning model carries out extraction of the oil palm in the target range according to a specified form. The multi-source remote sensing image data comprise multi-source satellite remote sensing images across time phases in a target range.
And the model selection unit is used for collecting sample vector data and acquiring corresponding sample data in the grid remote sensing image data. And evaluating and selecting the deep learning model by using the sample data to obtain the optimal deep learning model.
The model selection unit comprises a data collection module, a division module and a model selection module.
The data collection module is used for collecting sample vector data, acquiring corresponding sample data in the grid remote sensing image data, and carrying out standardized processing on the sample data.
Collecting sample vector data, namely collecting the sample vector data marked as oil palm on the existing ground in the industry, acquiring oil palm sample vector data of different geographic positions and time from a remote sensing image according to the experience of an industry expert, and collecting non-oil palm sample vector data;
obtaining corresponding sample data in the grid remote sensing image data, wherein the corresponding remote sensing image data block containing the oil palm and the non-oil palm in the grid remote sensing image data is obtained through vector data. The non-oil palm sample data can be data of cities, forests, water bodies, farmlands and the like, and is not particularly limited.
The standard sample data processing comprises respectively carrying out standard processing on the acquired remote sensing image data blocks containing the oil palm and the non-oil palm to obtain standard remote sensing image data, namely sample data. The standardization processing specifically comprises the size of an image block, the proportion of oil palm contained, and the recognition degree of the oil palm on the image.
The dividing module divides the sample data into training sample data and verification sample data. The method specifically comprises the steps of randomly dividing standard remote sensing image data into training sample data and verification sample data according to a fixed proportion, and dividing the training sample data and the verification sample data for multiple times according to the same method to obtain multiple groups of matched training sample data and verification sample data. The fixed proportion may be 80% of training sample data and 20% of verification sample data. The training sample data is mainly used for training the deep learning semantic segmentation model to obtain optimized parameters, and the verification sample data is used for evaluating the stability of the model. The fixed ratio may be other specified ratios, and is not limited herein.
The model selection module selects a deep learning semantic segmentation model and determines model structure parameters of the deep learning semantic segmentation model; carrying out verification training on model structure parameters of the deep learning semantic segmentation model; and evaluating the deep learning semantic segmentation model and the corresponding model structure parameters, and selecting the optimal deep learning model.
The deep learning semantic segmentation model can be Mask R-CNN, DenseNet, U-Net, SegNet, PSPNet and the like, and is not limited in particular.
The model structure parameters comprise model parameters and model hyper-parameters, and the model parameters comprise convolution layer number, convolution kernel size, pooling layer number, pooling mode, activation function and the like.
The verification training specifically may be to input training sample data and verification sample data into a model designed by preliminary parameters, and perform multiple times of cross verification training, so that the constructed deep learning semantic segmentation model can identify the oil palm according to the texture structure, shape, spectral characteristics, oil palm cluster characteristics, and the like of the oil palm.
The model selection module can evaluate the deep learning semantic segmentation model through the identification accuracy, precision, recall rate, ROC curve and the like, and selects the optimal deep learning model and the corresponding parameter configuration.
And the oil palm extraction unit inputs the grid remote sensing image data into an optimal deep learning model, identifies and segments the oil palms in the grid remote sensing image data and counts the area of the oil palms.
The oil palm extraction unit comprises an oil palm identification module and an oil palm statistics module.
And the oil palm recognition module inputs the grid remote sensing image data into an optimal deep learning model and recognizes the oil palm from the grid remote sensing image data in a large range.
The oil palm statistical module is used for segmenting grid remote sensing image data containing oil palms to obtain oil palm image data, optimizing the oil palm image data and further counting the area of the oil palms in the oil palm image data.
The oil palm area statistical result comprises oil palm data, and the oil palm data are grid data or vector data with geographic coordinates.
And the output unit is used for outputting and displaying the oil palm area statistical result, and presenting the grid data or vector data with geographic coordinates, oil palm area statistical data and the like comprising the oil palm to a data user.
The method and the system can identify and extract oil palm in a large range, save time and labor compared with the traditional ground investigation method, and save configuration and coordination of various resources when the area of the oil palm is acquired in ground investigation.
The following is a brief description of the title in this application:
ROC curve, receiver operating characteristic curve;
mask R-CNN, which is one of example segmentation, wherein the example segmentation is to detect a target object in an image and segment the detected target object;
DenseNet, dense convolutional network;
U-Net, a convolutional neural network for biomedical image segmentation, is a semantic segmentation network based on a full convolutional network, and is suitable for segmentation of medical images;
SegNet, semantic segmentation, specifically classifying pixels in an image;
PSPNet, English is called Pyramid Scene Parsing Network, and Chinese is Pyramid Scene Parsing Network.
The above description is provided for the purpose of illustrating the preferred embodiments of the present invention and will assist those skilled in the art in more fully understanding the technical solutions of the present invention. However, these examples are merely illustrative, and the embodiments of the present invention are not to be considered as being limited to the description of these examples. For those skilled in the art to which the invention pertains, several simple deductions and changes can be made without departing from the inventive concept, and all should be considered as falling within the protection scope of the invention.
Claims (10)
1. The method for oil palm identification and area extraction based on the multisource satellite remote sensing image is characterized by comprising the following steps,
step one, a multi-source satellite remote sensing image acquisition unit acquires at least two satellite remote sensing image data of a target area through a multi-source satellite, and carries out standardized processing to generate the multi-source remote sensing image data;
step two, the image segmentation unit carries out multilevel gridding segmentation processing on the multi-source remote sensing image data to obtain gridding remote sensing image data;
thirdly, the model selecting unit collects sample vector data, acquires corresponding sample data in the grid remote sensing image data, and evaluates and selects the deep learning model by using the sample data to obtain an optimal deep learning model;
inputting the grid remote sensing image data into an optimal deep learning model by an oil palm extraction unit, identifying and segmenting oil palms in the grid remote sensing image data and counting the area of the oil palms;
and fifthly, the output unit displays the output of the oil palm area statistical result.
2. The method for oil palm identification and area extraction based on multi-source satellite remote sensing images according to claim 1, wherein the third step comprises:
a3, collecting sample vector data by a data collection module, acquiring corresponding sample data in grid remote sensing image data, and carrying out standardized processing on the sample data;
step b3, the dividing module divides the sample data into training sample data and verification sample data;
step c3, selecting a deep learning semantic segmentation model by a model selection module, and determining model structure parameters of the deep learning semantic segmentation model; carrying out verification training on model structure parameters of the deep learning semantic segmentation model; and evaluating the model structure parameters of the deep learning semantic segmentation model, and selecting the optimal deep learning model.
3. The method for oil palm identification and area extraction based on multi-source satellite remote sensing images according to claim 1, wherein the fourth step comprises:
step a4, the oil palm recognition module inputs the grid remote sensing image data into an optimal deep learning model, and oil palm recognition is carried out on the grid remote sensing image data;
and b4, the oil palm statistical module divides the grid remote sensing image data containing the oil palm and counts the area of the oil palm.
4. Based on multisource satellite remote sensing image oil palm discernment and area extraction system, its characterized in that includes:
the multi-source satellite remote sensing image acquisition unit is used for carrying out standardized processing on at least two satellite remote sensing image data of a target area to generate multi-source remote sensing image data;
the image segmentation unit is used for carrying out multi-level gridding segmentation processing on the multi-source remote sensing image data to obtain gridding remote sensing image data;
the model selection unit is used for collecting sample vector data, acquiring corresponding sample data in the grid remote sensing image data, and evaluating and selecting the deep learning model by using the sample data to obtain an optimal deep learning model;
the oil palm extraction unit is used for inputting the grid remote sensing image data into an optimal deep learning model, identifying and segmenting oil palms in the grid remote sensing image data and counting the area of the oil palms;
and the output unit is used for outputting and displaying the oil palm area statistical result.
5. The oil palm identification and area extraction system based on the multi-source satellite remote sensing image according to claim 4, wherein the sample vector data comprises sample vector data of oil palm marked on the ground in the industry, oil palm sample vector data of different geographic positions and time and non-oil palm sample vector data acquired from the remote sensing image according to industry expert experience.
6. The oil palm identification and area extraction system based on the multisource satellite remote sensing image according to claim 4, wherein the model selection unit comprises a data collection module, a division module and a model selection module;
the data collection module is used for collecting sample vector data, acquiring corresponding sample data in the grid remote sensing image data, and carrying out standardized processing on the sample data; the dividing module divides the sample data into training sample data and verification sample data; the model selection module selects a deep learning semantic segmentation model and determines model structure parameters of the deep learning semantic segmentation model; carrying out verification training on model structure parameters of the deep learning semantic segmentation model; and evaluating the model structure parameters of the deep learning semantic segmentation model, and selecting the optimal deep learning model.
7. The oil palm identification and area extraction system based on multi-source satellite remote sensing images of claim 6, wherein the model structure parameters comprise model parameters and model hyper-parameters, and the model parameters comprise convolution layer number, convolution kernel size, pooling layer number, pooling mode and activation function.
8. The multi-source satellite remote sensing image-based oil palm recognition and area extraction system according to claim 6, wherein the model selection module evaluates the deep learning semantic segmentation model through recognition accuracy, precision, recall and ROC curves, and selects an optimal deep learning model and corresponding parameter configuration.
9. The oil palm recognition and area extraction system based on the multi-source satellite remote sensing image according to claim 4, wherein the oil palm extraction unit comprises an oil palm recognition module and an oil palm statistics module;
the oil palm recognition module inputs the grid remote sensing image data into an optimal deep learning model and carries out oil palm recognition on the grid remote sensing image data;
the oil palm statistical module is used for segmenting grid remote sensing image data containing oil palms and counting the areas of the oil palms.
10. The oil palm recognition and area extraction system based on the multi-source satellite remote sensing image according to claim 4, wherein oil palm area statistics results comprise oil palm data, and the oil palm data are grid data or vector data with geographic coordinates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010332738.3A CN111476197A (en) | 2020-04-24 | 2020-04-24 | Oil palm identification and area extraction method and system based on multi-source satellite remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010332738.3A CN111476197A (en) | 2020-04-24 | 2020-04-24 | Oil palm identification and area extraction method and system based on multi-source satellite remote sensing image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111476197A true CN111476197A (en) | 2020-07-31 |
Family
ID=71764064
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010332738.3A Pending CN111476197A (en) | 2020-04-24 | 2020-04-24 | Oil palm identification and area extraction method and system based on multi-source satellite remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111476197A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112634349A (en) * | 2020-12-22 | 2021-04-09 | 福建省星云大数据应用服务有限公司 | Remote-sensing-image-based tea garden area estimation method and system |
| CN112966548A (en) * | 2021-01-18 | 2021-06-15 | 中国农业科学院农业信息研究所 | Soybean plot identification method and system |
| CN113706050A (en) * | 2021-09-07 | 2021-11-26 | 合肥工业大学 | Multi-satellite region observation planning method based on adaptive genetic algorithm |
| CN117522956A (en) * | 2023-11-20 | 2024-02-06 | 青海大学 | Method for calculating degradation indication grass seed coverage by using unmanned aerial vehicle image |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104915636A (en) * | 2015-04-15 | 2015-09-16 | 北京工业大学 | Remote sensing image road identification method based on multistage frame significant characteristics |
| CN105116464A (en) * | 2015-08-12 | 2015-12-02 | 南京大学 | Polar sea ice melting pool extraction method based on neural network model |
| CN105354534A (en) * | 2015-09-29 | 2016-02-24 | 南京林业大学 | Tree species classification method based on multi-source simultaneous high-resolution remote sensing data |
| CN106950177A (en) * | 2016-11-16 | 2017-07-14 | 中国科学院遥感与数字地球研究所 | A kind of Clean water withdraw method of the satellite remote-sensing images of GF 4 |
| CN107132539A (en) * | 2017-05-03 | 2017-09-05 | 中国地质科学院探矿工艺研究所 | Landslide early-stage identification method of time sequence InSAR (interferometric synthetic Aperture Radar) based on small baseline set |
| CN110309809A (en) * | 2019-07-09 | 2019-10-08 | 广西壮族自治区基础地理信息中心 | High Resolution Remote Sensing Satellites image sugarcane extracting method based on deep neural network |
| CN110826509A (en) * | 2019-11-12 | 2020-02-21 | 云南农业大学 | Grassland fence information extraction system and method based on high-resolution remote sensing image |
| CN110849814A (en) * | 2019-11-20 | 2020-02-28 | 平衡机器科技(深圳)有限公司 | Remote sensing image processing method based on multi-source remote sensing satellite |
-
2020
- 2020-04-24 CN CN202010332738.3A patent/CN111476197A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104915636A (en) * | 2015-04-15 | 2015-09-16 | 北京工业大学 | Remote sensing image road identification method based on multistage frame significant characteristics |
| CN105116464A (en) * | 2015-08-12 | 2015-12-02 | 南京大学 | Polar sea ice melting pool extraction method based on neural network model |
| CN105354534A (en) * | 2015-09-29 | 2016-02-24 | 南京林业大学 | Tree species classification method based on multi-source simultaneous high-resolution remote sensing data |
| CN106950177A (en) * | 2016-11-16 | 2017-07-14 | 中国科学院遥感与数字地球研究所 | A kind of Clean water withdraw method of the satellite remote-sensing images of GF 4 |
| CN107132539A (en) * | 2017-05-03 | 2017-09-05 | 中国地质科学院探矿工艺研究所 | Landslide early-stage identification method of time sequence InSAR (interferometric synthetic Aperture Radar) based on small baseline set |
| CN110309809A (en) * | 2019-07-09 | 2019-10-08 | 广西壮族自治区基础地理信息中心 | High Resolution Remote Sensing Satellites image sugarcane extracting method based on deep neural network |
| CN110826509A (en) * | 2019-11-12 | 2020-02-21 | 云南农业大学 | Grassland fence information extraction system and method based on high-resolution remote sensing image |
| CN110849814A (en) * | 2019-11-20 | 2020-02-28 | 平衡机器科技(深圳)有限公司 | Remote sensing image processing method based on multi-source remote sensing satellite |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112634349A (en) * | 2020-12-22 | 2021-04-09 | 福建省星云大数据应用服务有限公司 | Remote-sensing-image-based tea garden area estimation method and system |
| CN112966548A (en) * | 2021-01-18 | 2021-06-15 | 中国农业科学院农业信息研究所 | Soybean plot identification method and system |
| CN112966548B (en) * | 2021-01-18 | 2023-10-27 | 中国农业科学院农业信息研究所 | Soybean plot identification method and system |
| CN113706050A (en) * | 2021-09-07 | 2021-11-26 | 合肥工业大学 | Multi-satellite region observation planning method based on adaptive genetic algorithm |
| CN113706050B (en) * | 2021-09-07 | 2023-06-27 | 合肥工业大学 | Multi-satellite region observation planning method based on adaptive genetic algorithm |
| CN117522956A (en) * | 2023-11-20 | 2024-02-06 | 青海大学 | Method for calculating degradation indication grass seed coverage by using unmanned aerial vehicle image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111709379B (en) | Remote sensing image-based hilly area citrus planting land plot monitoring method and system | |
| CN107016405B (en) | A kind of pest image classification method based on classification prediction convolutional neural networks | |
| CN111476713B (en) | Intelligent weather image identification method and system based on multi-depth convolution neural network fusion | |
| CN112101159B (en) | Multi-temporal forest remote sensing image change monitoring method | |
| CN111476197A (en) | Oil palm identification and area extraction method and system based on multi-source satellite remote sensing image | |
| CN112749627A (en) | Method and device for dynamically monitoring tobacco based on multi-source remote sensing image | |
| CN111191628B (en) | Remote sensing image earthquake damage building identification method based on decision tree and feature optimization | |
| CN107016403A (en) | A kind of method that completed region of the city threshold value is extracted based on nighttime light data | |
| CN107292314A (en) | A kind of lepidopterous insects species automatic identification method based on CNN | |
| CN113887515A (en) | Remote sensing landslide identification method and system based on convolutional neural network | |
| CN114067219A (en) | Farmland crop identification method based on semantic segmentation and superpixel segmentation fusion | |
| CN110287882A (en) | A kind of big chrysanthemum kind image-recognizing method based on deep learning | |
| CN113312993B (en) | Remote sensing data land cover classification method based on PSPNet | |
| CN112950780A (en) | Intelligent network map generation method and system based on remote sensing image | |
| CN116543316B (en) | Method for identifying turf in paddy field by utilizing multi-time-phase high-resolution satellite image | |
| CN112906537A (en) | Crop identification method and system based on convolutional neural network | |
| Baraldi et al. | Operational performance of an automatic preliminary spectral rule-based decision-tree classifier of spaceborne very high resolution optical images | |
| CN110598516A (en) | Random forest based multi-azimuth layered collection combined paddy field weed identification method | |
| CN112347926B (en) | High-resolution image city village detection method based on building morphology distribution | |
| CN115147726B (en) | City form map generation method and device, electronic equipment and readable storage medium | |
| CN111444824A (en) | Vegetation spatial distribution pattern investigation method and vegetation classification method based on unmanned aerial vehicle technology | |
| CN109784168A (en) | A kind of high-definition remote sensing passway for transmitting electricity inspection method and system | |
| Li et al. | Image recognition and empirical application of desert plant species based on convolutional neural network | |
| Hakim et al. | Automatic rain detection system based on digital images of CCTV cameras using convolutional neural network method | |
| Cao et al. | Vegetation phenology detection of deciduous broad-leaf forest using YOLOv3 from PhenoCam |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200731 |
|
| RJ01 | Rejection of invention patent application after publication |