CN115995005B - Crop extraction method and device based on single-period high-resolution remote sensing image - Google Patents
Crop extraction method and device based on single-period high-resolution remote sensing image Download PDFInfo
- Publication number
- CN115995005B CN115995005B CN202310280509.5A CN202310280509A CN115995005B CN 115995005 B CN115995005 B CN 115995005B CN 202310280509 A CN202310280509 A CN 202310280509A CN 115995005 B CN115995005 B CN 115995005B
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- target
- image
- resolution remote
- sensing image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Landscapes
- Image Processing (AREA)
Abstract
The invention provides a crop extraction method and device based on single-period high-resolution remote sensing images, which relate to the technical field of natural resource remote sensing monitoring and comprise the following steps: acquiring a sample single-phase high-resolution remote sensing image of a crop in a growing period; preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image; dividing the target single-period high-resolution remote sensing image by using a scale set dividing algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data; training a preset convolutional neural network model by utilizing target sample data to obtain a target convolutional neural network model; after the single-phase high-resolution remote sensing image to be extracted is obtained, a target convolutional neural network model is utilized to determine the extraction result of crops in the single-phase high-resolution remote sensing image to be extracted, and the technical problem that the existing crop extraction method based on the high-resolution remote sensing image is low in efficiency is solved.
Description
Technical Field
The invention relates to the technical field of remote sensing monitoring of natural resources, in particular to a crop extraction method and device based on single-period high-resolution remote sensing images.
Background
Common remote sensing data sources for crop identification are optical and SAR (synthetic aperture radar) remote sensing data with medium and low resolution, and are limited by image resolution, and the precision of the area and attribute extracted by crops is limited. The method for extracting crops is based on multi-stage images of key waiting periods, even adopts long-time sequence data, combines various vegetation indexes such as ratio vegetation index RVI, normalized vegetation index NDVI and the like, adopts machine learning classification methods such as maximum likelihood, support vector machine, decision tree and the like, or non-supervision classification methods such as ISODATA, K-means and the like, and adopts knowledge rule classification methods if the methods are not supervised, so that the method is still effective for extracting single crops. When the land is crushed and various crops are planted in a mixed mode, due to the fact that the similarity of the characteristics of the spectrum, the texture, the shape and the like is high, no matter the pixel level or the object level analysis method is adopted, the proper characteristics are difficult to find by the traditional shallow machine learning and knowledge rule method, and different crops are distinguished. In conclusion, the existing crop extraction method based on the high-resolution remote sensing image is low in efficiency and accuracy.
An effective solution to the above-mentioned problems has not been proposed yet.
Disclosure of Invention
In view of the above, the invention aims to provide a crop extraction method and device based on single-period high-resolution remote sensing images, so as to solve the technical problem of low efficiency of the existing crop extraction method based on high-resolution remote sensing images.
In a first aspect, an embodiment of the present invention provides a method for extracting crops based on single-period high-resolution remote sensing images, including: acquiring a sample single-phase high-resolution remote sensing image of a crop in a growing period; preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image; dividing the target single-period high-resolution remote sensing image by using a scale set dividing algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data, wherein the labels are used for representing the ground object types of the sample data; training a preset convolutional neural network model by utilizing the target sample data to obtain a target convolutional neural network model; after obtaining a single-stage high-resolution remote sensing image to be extracted, determining an extraction result of crops in the single-stage high-resolution remote sensing image to be extracted by using the target convolutional neural network model, wherein the extraction result comprises the following steps: the type, area and spatial distribution of the crop.
Further, the sample single-period high-resolution remote sensing image comprises: multispectral imaging and panchromatic imaging; preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image, wherein the preprocessing comprises the following steps: performing radiation calibration treatment and atmospheric correction treatment on the multispectral image and the full-color image in sequence respectively to obtain a target multispectral image and a target full-color image; carrying out orthographic correction processing on the target full-color image to obtain a corrected target full-color image; performing geometric registration processing and image fusion processing on the corrected target full-color image and the corrected target multispectral image to obtain a fusion image; and carrying out uniform color mosaic processing on the fusion image to obtain the target single-period high-resolution remote sensing image.
Further, the target single-period high-resolution remote sensing image is segmented by using a scale set segmentation algorithm to obtain sample data, which comprises the following steps: determining homogeneous pixels in the target single-phase high-resolution remote sensing image based on target parameters of the target single-phase high-resolution remote sensing image, wherein the target parameters comprise: color and geometry; and merging homogeneous pixels in the target single-phase high-resolution remote sensing image based on the scale set segmentation algorithm to obtain a first merged image spot, and determining the first merged image spot as the sample data.
Further, training a preset convolutional neural network model by using the target sample data to obtain a target convolutional neural network model, including: performing hierarchical sampling on the target sample data by using a hierarchical sampling method to obtain a training set and an accuracy evaluation set; taking the center of the sample data in the training set as a segmentation center, and segmenting the sample data in the training set according to a preset size to obtain a sample block; training and testing the preset convolutional neural network model by using the sample block to obtain an initial convolutional neural network model; and performing precision evaluation on the initial convolutional neural network model by using the precision evaluation set, and determining the initial convolutional neural network model as the target convolutional neural network model when the precision evaluation result of the initial convolutional neural network model is greater than a preset threshold value.
Further, the preset convolutional neural network model is a 6-layer convolutional neural network model constructed based on PyTorch.
Further, determining an extraction result of the crop in the to-be-extracted single-period high-resolution remote sensing image by using the target convolutional neural network model comprises the following steps: the pretreatment is carried out on the single-phase high-resolution remote sensing image to be extracted, so that a pretreated single-phase high-resolution remote sensing image to be extracted is obtained; determining homogeneous pixels in the preprocessed single-phase high-resolution remote sensing image to be extracted based on target parameters of the preprocessed single-phase high-resolution remote sensing image to be extracted; based on the scale set segmentation algorithm, merging homogeneous pixels in the preprocessed single-phase high-resolution remote sensing image to be extracted to obtain a second merged image spot; taking the center of the segmented image as a segmentation center, and segmenting the second merging image spots according to a preset size to obtain an image block; and inputting the image block into the target convolutional neural network model, and determining the extraction result of the crops in the single-stage high-resolution remote sensing image to be extracted.
In a second aspect, an embodiment of the present invention further provides an apparatus for extracting a crop based on a single-period high-resolution remote sensing image, including: the acquisition unit is used for acquiring a sample single-period high-resolution remote sensing image of the crop in the vigorous growth period; the preprocessing unit is used for preprocessing the sample single-stage high-resolution remote sensing image to obtain a target single-stage high-resolution remote sensing image; the segmentation unit is used for segmenting the target single-period high-resolution remote sensing image by using a scale set segmentation algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data, wherein the labels are used for representing the types of ground features of the sample data; the training unit is used for training a preset convolutional neural network model by utilizing the target sample data to obtain a target convolutional neural network model; the extraction unit is used for determining an extraction result of crops in the single-period high-resolution remote sensing image to be extracted by using the target convolutional neural network model after the single-period high-resolution remote sensing image to be extracted is acquired, wherein the extraction result comprises the following steps: the type, area and spatial distribution of the crop.
Further, the sample single-period high-resolution remote sensing image comprises: multispectral imaging and panchromatic imaging; the preprocessing unit is used for: performing radiation calibration treatment and atmospheric correction treatment on the multispectral image and the full-color image in sequence respectively to obtain a target multispectral image and a target full-color image; carrying out orthographic correction processing on the target full-color image to obtain a corrected target full-color image; performing geometric registration processing and image fusion processing on the corrected target full-color image and the corrected target multispectral image to obtain a fusion image; and carrying out uniform color mosaic processing on the fusion image to obtain the target single-period high-resolution remote sensing image.
In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory and a processor, where the memory is configured to store a program for supporting the processor to execute the method described in the first aspect, and the processor is configured to execute the program stored in the memory.
In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon.
In the embodiment of the invention, a sample single-period high-resolution remote sensing image of the crop in the vigorous growth period is obtained; preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image; dividing the target single-period high-resolution remote sensing image by using a scale set dividing algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data, wherein the labels are used for representing the ground object types of the sample data; training a preset convolutional neural network model by utilizing the target sample data to obtain a target convolutional neural network model; after obtaining a single-stage high-resolution remote sensing image to be extracted, determining an extraction result of crops in the single-stage high-resolution remote sensing image to be extracted by using the target convolutional neural network model, wherein the extraction result comprises the following steps: the crop type, the planting area and the spatial distribution achieve the purpose of extracting the crops by utilizing the single-period high-resolution remote sensing image, and further solve the technical problem that the existing crop extraction method based on the high-resolution remote sensing image needs multi-period remote sensing images for extracting the crops, so that the extraction efficiency is low, and the technical effect of improving the extraction efficiency of the crops is achieved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for extracting crops based on single-period high-resolution remote sensing images according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an extraction device for crops based on single-period high-resolution remote sensing images according to an embodiment of the present invention;
fig. 3 is a schematic diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Embodiment one:
according to an embodiment of the present invention, there is provided an embodiment of a method for extracting crops based on single-period high-resolution remote sensing images, it should be noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order different from that herein.
Fig. 1 is a flowchart of a method for extracting crops based on single-period high-resolution remote sensing images according to an embodiment of the present invention, as shown in fig. 1, the method includes the following steps:
step S102, acquiring a sample single-period high-resolution remote sensing image of a crop in a growing period;
it should be noted that, the crops in the sample single-period high-resolution remote sensing image include any one or more of the following: soybean, corn, rice and ginseng.
Since the growth period of soybean, corn and rice is 4-10 months, the growth period of ginseng is 6-8 months, in order to extract various crops, the single-period high-resolution remote sensing image at the beginning of 8 months is preferably used as a sample single-period high-resolution remote sensing image, and in addition, the sample single-period high-resolution remote sensing image at least comprises blue, green, red and near infrared 4 multispectral wave bands in order to extract the result.
Step S104, preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image;
step S106, dividing the target single-period high-resolution remote sensing image by using a scale set dividing algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data, wherein the labels are used for representing the types of ground features of the sample data;
step S108, training a preset convolutional neural network model by using the target sample data to obtain a target convolutional neural network model;
step S110, after obtaining a single-stage high-resolution remote sensing image to be extracted, determining an extraction result of crops in the single-stage high-resolution remote sensing image to be extracted by using the target convolutional neural network model, wherein the extraction result comprises: the type, area and spatial distribution of the crop.
In the embodiment of the invention, a sample single-period high-resolution remote sensing image of the crop in the vigorous growth period is obtained; preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image; dividing the target single-period high-resolution remote sensing image by using a scale set dividing algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data, wherein the labels are used for representing the ground object types of the sample data; training a preset convolutional neural network model by utilizing the target sample data to obtain a target convolutional neural network model; after obtaining a single-stage high-resolution remote sensing image to be extracted, determining an extraction result of crops in the single-stage high-resolution remote sensing image to be extracted by using the target convolutional neural network model, wherein the extraction result comprises the following steps: the crop type, the planting area and the spatial distribution achieve the purpose of extracting the whole of various crops by utilizing the single-period high-resolution remote sensing image, and further solve the technical problem that the existing crop extraction method based on the high-resolution remote sensing image needs the multi-period remote sensing image for extracting the crops, so that the extraction efficiency is low, thereby realizing the technical effect of improving the extraction efficiency of the crops.
In an embodiment of the present invention, the sample single-period high-resolution remote sensing image includes: multispectral image and panchromatic image, step S104 includes the steps of;
performing radiation calibration treatment and atmospheric correction treatment on the multispectral image and the full-color image in sequence respectively to obtain a target multispectral image and a target full-color image;
carrying out orthographic correction processing on the target full-color image to obtain a corrected target full-color image;
performing geometric registration processing and image fusion processing on the corrected target full-color image and the corrected target multispectral image to obtain a fusion image;
and carrying out uniform color mosaic processing on the fusion image to obtain the target single-period high-resolution remote sensing image.
In the embodiment of the invention, in order to eliminate radiation changes caused by atmospheric conditions, solar altitude, satellite observation angles, terrains and other factors, pretreatment such as radiation calibration, atmospheric correction, orthographic correction, fusion, color balancing, mosaic and the like is required to be carried out on a sample single-period high-resolution remote sensing image. The sample single-stage high-resolution remote sensing image consists of a multispectral image and a panchromatic image, and radiation calibration and atmospheric correction are respectively carried out on the multispectral image and the panchromatic image to obtain a target multispectral image and a target panchromatic image.
Then, selecting a control point, combining DEM data, carrying out orthographic correction on the target panchromatic image to obtain a corrected target panchromatic image, carrying out geometric registration on the target multispectral image and the corrected target panchromatic image by taking the corrected target panchromatic image as a reference, fusing the registered panchromatic image and the multispectral image to obtain a fused image, and further carrying out color homogenization and mosaic to form an orthographic image (namely, a target single-period high-resolution remote sensing image).
The radiometric calibration is to eliminate the radiation error caused by the sensor and convert the dimensionless DN value recorded by the sensor into the atmospheric top layer radiation brightness or reflectivity with practical physical significance.
The atmospheric correction is to convert the radiation brightness value of the top layer of the atmosphere (or the reflectivity of the top layer of the atmosphere) into the solar radiation brightness value reflected by the earth surface (or the reflectivity of the earth surface), and is mainly used for eliminating the influence of atmospheric absorption and scattering on radiation transmission.
The orthographic correction is to correct image point displacement caused by sensor errors, topographic relief and the like by means of satellite sensor model parameters, a geometric model, control points and DEM data, eliminate geometric deformation of images and generate orthographic images.
Geometric registration is the process of precisely matching and superimposing identical feature elements in a dataset to each other.
The image fusion is to fuse the geometrically registered full-color and multispectral images by adopting a certain algorithm, so that the fused images have the spatial resolution of the full-color images and the spectral resolution of the multispectral images at the same time.
And (3) color homogenizing and embedding, and selecting a proper color homogenizing mode according to the data condition. If the historical images which have the same geographical range as the images to be processed and have similar imaging seasons and better colors exist, the geographical templates can be used for evenly coloring. Without the template image, area networks may be used for even color. The controlled area network is generally used for uniformly coloring by using an image with better one or more scenes in the images to be uniformly colored as a control image. And after the color homogenization is finished, automatically generating an embedded line for mosaic and splicing processing, and generating an orthographic image covering the working area.
In the embodiment of the present invention, step S106 includes the following steps:
determining homogeneous pixels in the target single-phase high-resolution remote sensing image based on target parameters of the target single-phase high-resolution remote sensing image, wherein the target parameters comprise: color and geometry;
and merging homogeneous pixels in the target single-phase high-resolution remote sensing image based on the scale set segmentation algorithm to obtain a first merged image spot, and determining the first merged image spot as the sample data.
In the embodiment of the invention, homogeneous pixels are combined to form a first combined image spot according to the color and the geometric shape of the target single-period high-resolution remote sensing image, and then the first combined image spot is segmented by adopting a scale set segmentation algorithm to obtain sample data.
After the sample data are obtained, the ground features in the sample data are randomly marked according to the pattern spots, and the marking principle is that the samples of all the places are complete and the number is balanced. Even though the class is not interesting, it is sometimes necessary to label a certain amount of negative samples in order not to confuse. The related semantic information of a certain class of samples is expressed by different visual features, the visual features have complementarity, the quantity of the labeled samples of the ground classes with small intra-class differences can be slightly smaller, and the labeled samples of the ground classes with large intra-class differences are needed to be labeled more, so that the intra-class diversity is ensured. After the sample labeling is completed, all the sample and non-sample image spots are stored as a shape vector format together with the category attribute.
In the embodiment of the present invention, step S108 includes the steps of:
performing hierarchical sampling on the target sample data by using a hierarchical sampling method to obtain a training set and an accuracy evaluation set;
taking the center of the sample data in the training set as a segmentation center, and segmenting the sample data in the training set according to a preset size to obtain a sample block;
training and testing the preset convolutional neural network model by using the sample block to obtain an initial convolutional neural network model;
and performing precision evaluation on the initial convolutional neural network model by using the precision evaluation set, and determining the initial convolutional neural network model as the target convolutional neural network model when the precision evaluation result of the initial convolutional neural network model is greater than a preset threshold value.
In the embodiment of the invention, because the manually marked sample data has certain randomness, a hierarchical sampling method is adopted to divide the samples into two groups: one set is used for model training and testing, i.e., training set, and the other set is used for predicting result accuracy evaluation, i.e., accuracy evaluation set. The specific method comprises the following steps:
firstly, calculating brightness values of green, red and near infrared bands in each sample data according to the ground type, sequencing according to the sequence from small brightness values to large brightness values, grouping and extracting sample data with a certain proportion, and constructing a training set.
And cutting sample data in the training set into sample blocks with the required size of the model by taking the geometric center of the sample data as the center, and storing the sample blocks into a designated catalog for subsequent model training, testing and verification.
And identifying sample data in the precision evaluation set for future precision evaluation of the prediction result.
And then training the preset convolutional neural network model by using a training set by using a small-batch gradient descent method to obtain an initial convolutional neural network model, setting the initial learning rate to be 0.0001, setting the learning times to be 500, and adjusting by adopting a Cosine attenuation mode. Batch size was 64, accelerated convergence using adam optimization, and training time was about 20 minutes with dual GPU.
The preset convolutional neural network model is a 6-layer convolutional neural network built based on PyTorch, the size of an input image block is 64 x 64, 5 convolutional layers are total, the size of a convolutional kernel is 3*3, and the step length is 1. The first layer is subjected to pooling operation, the rest layers are subjected to pooling with a convolution kernel of 2 x 2 and a step length of 2, and the obtained characteristic maps have side lengths of 62, 60, 28, 12 and 4 respectively. The loss function employs cross entropy loss.
And finally, performing precision evaluation on the initial convolutional neural network model by using a precision evaluation set, and determining the initial convolutional neural network model as a target convolutional neural network model when the precision evaluation result of the initial convolutional neural network model is greater than a preset threshold value.
In the embodiment of the present invention, step S110 includes the following steps:
the pretreatment is carried out on the single-phase high-resolution remote sensing image to be extracted, so that a pretreated single-phase high-resolution remote sensing image to be extracted is obtained;
determining homogeneous pixels in the preprocessed single-phase high-resolution remote sensing image to be extracted based on target parameters of the preprocessed single-phase high-resolution remote sensing image to be extracted;
based on the scale set segmentation algorithm, merging homogeneous pixels in the preprocessed single-phase high-resolution remote sensing image to be extracted to obtain a second merged image spot;
taking the center of the second combined image spot as a segmentation center, and segmenting the second combined image spot according to a preset size to obtain an image block;
and inputting the image block into the target convolutional neural network model, and determining the extraction result of the crops in the single-stage high-resolution remote sensing image to be extracted.
The target convolutional neural network model in the embodiment of the invention is different from a general convolutional neural network model, and the general convolutional neural network model adopts sliding window cutting blocks and pixel-by-pixel prediction. And the target convolutional neural network model cuts the image by taking the geometric center of the merged image spots corresponding to the single-period high-resolution remote sensing image to be extracted as the center, predicts the image spots one by one, and obtains the probability value of each merged image spot belonging to different ground object categories. And finally, according to the set threshold, assigning the category attribute to the corresponding merged patch vector polygon.
After the ground object category corresponding to each merging map spot is determined, the planting area and the spatial distribution are determined according to the area of the merging map spots.
In the embodiment of the invention, the single-phase high-resolution remote sensing image of the crop in the vigorous growth period is adopted, so that the problem of low extraction efficiency caused by large data processing workload in the traditional crop extraction method based on medium-low resolution multi-phase or long-time sequence data is solved.
Meanwhile, the defects of low attribute and area precision caused by serious spectrum confusion due to the crushing of plots and the mixed planting of contemporaneous crops are overcome.
The embodiment of the invention is different from a common single crop extraction method, and can realize the integral extraction of the planting area and the spatial distribution of various crops such as regional soybeans, corns, rice and the like and ginseng by one-time investment, thereby avoiding the complicated and repeated investment of each crop for independent extraction.
Embodiment two:
the embodiment of the invention also provides a crop extraction device based on the single-period high-resolution remote sensing image, which is used for executing the crop extraction method based on the single-period high-resolution remote sensing image provided by the embodiment of the invention, and the following is a specific introduction of the crop extraction device based on the single-period high-resolution remote sensing image provided by the embodiment of the invention.
As shown in fig. 2, fig. 2 is a schematic diagram of the device for extracting a crop based on a single-period high-resolution remote sensing image, where the device for extracting a crop based on a single-period high-resolution remote sensing image includes:
an acquisition unit 10, configured to acquire a sample single-period high-resolution remote sensing image of a crop in a vigorous growth period;
the preprocessing unit 20 is configured to preprocess the sample single-stage high-resolution remote sensing image to obtain a target single-stage high-resolution remote sensing image;
the segmentation unit 30 is configured to segment the target single-period high-resolution remote sensing image by using a scale set segmentation algorithm to obtain sample data, and add a label to the sample data to obtain target sample data, where the label is used to characterize a feature type of the sample data;
the training unit 40 is configured to train a preset convolutional neural network model by using the target sample data, so as to obtain a target convolutional neural network model;
the extracting unit 50 is configured to determine, after obtaining the single-period high-resolution remote sensing image to be extracted, an extraction result of the crop in the single-period high-resolution remote sensing image to be extracted by using the target convolutional neural network model, where the extraction result includes: the type, area and spatial distribution of the crop.
In the embodiment of the invention, a sample single-period high-resolution remote sensing image of the crop in the vigorous growth period is obtained; preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image; dividing the target single-period high-resolution remote sensing image by using a scale set dividing algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data, wherein the labels are used for representing the ground object types of the sample data; training a preset convolutional neural network model by utilizing the target sample data to obtain a target convolutional neural network model; after obtaining a single-stage high-resolution remote sensing image to be extracted, determining an extraction result of crops in the single-stage high-resolution remote sensing image to be extracted by using the target convolutional neural network model, wherein the extraction result comprises the following steps: the crop type, the planting area and the spatial distribution achieve the purpose of extracting the whole of various crops by utilizing the single-period high-resolution remote sensing image, and further solve the technical problem that the existing crop extraction method based on the high-resolution remote sensing image needs the multi-period remote sensing image for extracting the crops, so that the extraction efficiency is low, thereby realizing the technical effect of improving the extraction efficiency of the crops.
Embodiment III:
an embodiment of the present invention further provides an electronic device, including a memory and a processor, where the memory is configured to store a program that supports the processor to execute the method described in the first embodiment, and the processor is configured to execute the program stored in the memory.
Referring to fig. 3, an embodiment of the present invention further provides an electronic device 100, including: a processor 60, a memory 61, a bus 62 and a communication interface 63, the processor 60, the communication interface 63 and the memory 61 being connected by the bus 62; the processor 60 is arranged to execute executable modules, such as computer programs, stored in the memory 61.
The memory 61 may include a high-speed random access memory (RAM, random Access Memory), and may further include a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. The communication connection between the system network element and at least one other network element is achieved via at least one communication interface 63 (which may be wired or wireless), and may use the internet, a wide area network, a local network, a metropolitan area network, etc.
Bus 62 may be an ISA bus, a PCI bus, an EISA bus, or the like. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in FIG. 3, but not only one bus or type of bus.
The memory 61 is configured to store a program, and the processor 60 executes the program after receiving an execution instruction, and the method executed by the apparatus for flow defining disclosed in any of the foregoing embodiments of the present invention may be applied to the processor 60 or implemented by the processor 60.
The processor 60 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuitry in hardware or instructions in software in the processor 60. The processor 60 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a digital signal processor (Digital Signal Processing, DSP for short), application specific integrated circuit (Application Specific Integrated Circuit, ASIC for short), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA for short), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory 61 and the processor 60 reads the information in the memory 61 and in combination with its hardware performs the steps of the method described above.
Embodiment four:
the embodiment of the invention also provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of the method in the first embodiment are executed.
In addition, in the description of embodiments of the present invention, unless explicitly stated and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
Finally, it should be noted that: the above examples are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention, but it should be understood by those skilled in the art that the present invention is not limited thereto, and that the present invention is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (9)
1. The method for extracting the crops based on the single-period high-resolution remote sensing image is characterized by comprising the following steps of:
acquiring a sample single-phase high-resolution remote sensing image of a crop in a growing period;
preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image;
dividing the target single-period high-resolution remote sensing image by using a scale set dividing algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data, wherein the labels are used for representing the ground object types of the sample data;
training a preset convolutional neural network model by utilizing the target sample data to obtain a target convolutional neural network model;
after obtaining a single-stage high-resolution remote sensing image to be extracted, determining an extraction result of crops in the single-stage high-resolution remote sensing image to be extracted by using the target convolutional neural network model, wherein the extraction result comprises the following steps: the type, planting area and spatial distribution of crops;
the method for segmenting the target single-period high-resolution remote sensing image by using a scale set segmentation algorithm to obtain sample data comprises the following steps:
determining homogeneous pixels in the target single-phase high-resolution remote sensing image based on target parameters of the target single-phase high-resolution remote sensing image, wherein the target parameters comprise: color and geometry;
and merging homogeneous pixels in the target single-phase high-resolution remote sensing image based on the scale set segmentation algorithm to obtain a first merged image spot, and determining the first merged image spot as the sample data.
2. The method of claim 1, wherein the sample single-phase high-resolution remote sensing image comprises: multispectral imaging and panchromatic imaging;
preprocessing the sample single-phase high-resolution remote sensing image to obtain a target single-phase high-resolution remote sensing image, wherein the preprocessing comprises the following steps:
performing radiation calibration treatment and atmospheric correction treatment on the multispectral image and the full-color image in sequence respectively to obtain a target multispectral image and a target full-color image;
carrying out orthographic correction processing on the target full-color image to obtain a corrected target full-color image;
performing geometric registration processing and image fusion processing on the corrected target full-color image and the corrected target multispectral image to obtain a fusion image;
and carrying out uniform color mosaic processing on the fusion image to obtain the target single-period high-resolution remote sensing image.
3. The method of claim 1, wherein training a predetermined convolutional neural network model using the target sample data to obtain a target convolutional neural network model comprises:
performing hierarchical sampling on the target sample data by using a hierarchical sampling method to obtain a training set and an accuracy evaluation set;
taking the center of the sample data in the training set as a segmentation center, and segmenting the sample data in the training set according to a preset size to obtain a sample block;
training and testing the preset convolutional neural network model by using the sample block to obtain an initial convolutional neural network model;
and performing precision evaluation on the initial convolutional neural network model by using the precision evaluation set, and determining the initial convolutional neural network model as the target convolutional neural network model when the precision evaluation result of the initial convolutional neural network model is greater than a preset threshold value.
4. The method of claim 1, wherein the step of determining the position of the substrate comprises,
the preset convolutional neural network model is a 6-layer convolutional neural network model constructed based on PyTorch.
5. The method of claim 1, wherein determining, using the target convolutional neural network model, an extraction result of the crop in the single-phase high-resolution remote sensing image to be extracted comprises:
the pretreatment is carried out on the single-phase high-resolution remote sensing image to be extracted, so that a pretreated single-phase high-resolution remote sensing image to be extracted is obtained;
determining homogeneous pixels in the preprocessed single-phase high-resolution remote sensing image to be extracted based on target parameters of the preprocessed single-phase high-resolution remote sensing image to be extracted;
based on the scale set segmentation algorithm, merging homogeneous pixels in the preprocessed single-phase high-resolution remote sensing image to be extracted to obtain a second merged image spot;
taking the center of the second combined image spot as a segmentation center, and segmenting the second combined image spot according to a preset size to obtain an image block;
and inputting the image block into the target convolutional neural network model, and determining the extraction result of the crops in the single-stage high-resolution remote sensing image to be extracted.
6. Crop extraction device based on single-phase high-resolution remote sensing image, which is characterized by comprising:
the acquisition unit is used for acquiring a sample single-period high-resolution remote sensing image of the crop in the vigorous growth period;
the preprocessing unit is used for preprocessing the sample single-stage high-resolution remote sensing image to obtain a target single-stage high-resolution remote sensing image;
the segmentation unit is used for segmenting the target single-period high-resolution remote sensing image by using a scale set segmentation algorithm to obtain sample data, and adding labels to the sample data to obtain target sample data, wherein the labels are used for representing the types of ground features of the sample data;
the training unit is used for training a preset convolutional neural network model by utilizing the target sample data to obtain a target convolutional neural network model;
the extraction unit is used for determining an extraction result of crops in the single-period high-resolution remote sensing image to be extracted by using the target convolutional neural network model after the single-period high-resolution remote sensing image to be extracted is acquired, wherein the extraction result comprises the following steps: the type, planting area and spatial distribution of crops;
wherein the segmentation unit is used for:
determining homogeneous pixels in the target single-phase high-resolution remote sensing image based on target parameters of the target single-phase high-resolution remote sensing image, wherein the target parameters comprise: color and geometry;
and merging homogeneous pixels in the target single-phase high-resolution remote sensing image based on the scale set segmentation algorithm to obtain a first merged image spot, and determining the first merged image spot as the sample data.
7. The apparatus of claim 6, wherein the sample single-phase high-resolution remote sensing image comprises: multispectral imaging and panchromatic imaging;
the preprocessing unit is used for:
performing radiation calibration treatment and atmospheric correction treatment on the multispectral image and the full-color image in sequence respectively to obtain a target multispectral image and a target full-color image;
carrying out orthographic correction processing on the target full-color image to obtain a corrected target full-color image;
performing geometric registration processing and image fusion processing on the corrected target full-color image and the corrected target multispectral image to obtain a fusion image;
and carrying out uniform color mosaic processing on the fusion image to obtain the target single-period high-resolution remote sensing image.
8. An electronic device comprising a memory for storing a program supporting the processor to perform the method of any one of claims 1 to 5, and a processor configured to execute the program stored in the memory.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, performs the steps of the method according to any of the preceding claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310280509.5A CN115995005B (en) | 2023-03-22 | 2023-03-22 | Crop extraction method and device based on single-period high-resolution remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310280509.5A CN115995005B (en) | 2023-03-22 | 2023-03-22 | Crop extraction method and device based on single-period high-resolution remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115995005A CN115995005A (en) | 2023-04-21 |
| CN115995005B true CN115995005B (en) | 2023-08-01 |
Family
ID=85993714
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310280509.5A Active CN115995005B (en) | 2023-03-22 | 2023-03-22 | Crop extraction method and device based on single-period high-resolution remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115995005B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117095299B (en) * | 2023-10-18 | 2024-01-26 | 浙江省测绘科学技术研究院 | Grain crop extraction method, system, equipment and medium for crushing cultivation area |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113077458A (en) * | 2021-04-25 | 2021-07-06 | 北京艾尔思时代科技有限公司 | Cloud and shadow detection method and system in remote sensing image |
| WO2021184891A1 (en) * | 2020-03-20 | 2021-09-23 | 中国科学院深圳先进技术研究院 | Remotely-sensed image-based terrain classification method, and system |
| WO2023000159A1 (en) * | 2021-07-20 | 2023-01-26 | 海南长光卫星信息技术有限公司 | Semi-supervised classification method, apparatus and device for high-resolution remote sensing image, and medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751019B (en) * | 2019-09-03 | 2023-03-07 | 武汉珈和科技有限公司 | High-resolution image crop automatic extraction method and device based on deep learning |
| CN113920441A (en) * | 2021-08-30 | 2022-01-11 | 广东海洋大学 | High-precision farmland vegetation information extraction method |
| CN115631419B (en) * | 2022-12-08 | 2023-05-09 | 航天宏图信息技术股份有限公司 | Rice planting area and spatial distribution extraction method and device based on change detection |
-
2023
- 2023-03-22 CN CN202310280509.5A patent/CN115995005B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021184891A1 (en) * | 2020-03-20 | 2021-09-23 | 中国科学院深圳先进技术研究院 | Remotely-sensed image-based terrain classification method, and system |
| CN113077458A (en) * | 2021-04-25 | 2021-07-06 | 北京艾尔思时代科技有限公司 | Cloud and shadow detection method and system in remote sensing image |
| WO2023000159A1 (en) * | 2021-07-20 | 2023-01-26 | 海南长光卫星信息技术有限公司 | Semi-supervised classification method, apparatus and device for high-resolution remote sensing image, and medium |
Non-Patent Citations (4)
| Title |
|---|
| Study on Ensemble Crop Information Extractin of Remote Sensing Images Based on SVM and BPNN;Dawei Li等;Journal of the Indian Society of Remote Sensing;第45卷;第229-237 * |
| 光谱、形状特征结合的多精度图像分割算法与应用;汪闽;万其明;张大骞;张青峰;;地球信息科学学报;第12卷(第02期);第261-268页 * |
| 基于面向对象分类的南方水稻种植面积提取方法;陈燕丽;莫伟华;莫建飞;王君华;钟仕全;;遥感技术与应用;第26卷(第02期);第163-168页 * |
| 遥感组合指数与不同分类技术结合提取农业用地方法;张明月;杨贵军;宋伟东;徐涛;;测绘科学;第36卷(第05期);第73-76页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115995005A (en) | 2023-04-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11521380B2 (en) | Shadow and cloud masking for remote sensing images in agriculture applications using a multilayer perceptron | |
| CN112183209B (en) | Regional crop classification method and system based on multidimensional feature fusion | |
| US8682079B1 (en) | Automatic extraction of built-up footprints from high resolution overhead imagery through manipulation of alpha-tree data structures | |
| Yue et al. | Method for accurate multi-growth-stage estimation of fractional vegetation cover using unmanned aerial vehicle remote sensing | |
| CN109726705B (en) | Mangrove forest information extraction method and device and electronic equipment | |
| Shu et al. | Application of UAV multisensor data and ensemble approach for high-throughput estimation of maize phenotyping traits | |
| Deng et al. | Cloud detection in satellite images based on natural scene statistics and gabor features | |
| CN115481368B (en) | Vegetation coverage estimation method based on full remote sensing machine learning | |
| Wang et al. | Unsupervised discrimination between lodged and non-lodged winter wheat: A case study using a low-cost unmanned aerial vehicle | |
| Ghanea et al. | Automatic building extraction in dense urban areas through GeoEye multispectral imagery | |
| CN114219847B (en) | Method and system for determining crop planting area based on phenological characteristics and storage medium | |
| CN114463637B (en) | Winter wheat remote sensing identification analysis method and system based on deep learning | |
| CN115995005B (en) | Crop extraction method and device based on single-period high-resolution remote sensing image | |
| CN111144250A (en) | Land coverage classification method fusing radar and optical remote sensing data | |
| Ozdarici Ok et al. | A segment-based approach to classify agricultural lands by using multi-temporal optical and microwave data | |
| Tian et al. | Machine learning-based crop recognition from aerial remote sensing imagery | |
| Kuswidiyanto et al. | Airborne hyperspectral imaging for early diagnosis of kimchi cabbage downy mildew using 3D-ResNet and leaf segmentation | |
| CN109784320B (en) | Ginseng facility agricultural domain determination method | |
| CN117557897A (en) | Lodging monitoring method and device for target crops, electronic equipment and storage medium | |
| CN115631419B (en) | Rice planting area and spatial distribution extraction method and device based on change detection | |
| CN109726679B (en) | Remote sensing classification error spatial distribution mapping method | |
| Zhao et al. | Improving object-oriented land use/cover classification from high resolution imagery by spectral similarity-based post-classification | |
| JP6060377B2 (en) | Image cluster analyzer | |
| CN115049900A (en) | Rice planting area identification method and system based on optical and radar data | |
| Li et al. | Semi-supervised cooperative regression model for small sample estimation of citrus leaf nitrogen content with UAV images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |