CN114494319A - Sand dune shape extraction method and device based on remote sensing image - Google Patents

Sand dune shape extraction method and device based on remote sensing image Download PDF

Info

Publication number
CN114494319A
CN114494319A CN202210264843.7A CN202210264843A CN114494319A CN 114494319 A CN114494319 A CN 114494319A CN 202210264843 A CN202210264843 A CN 202210264843A CN 114494319 A CN114494319 A CN 114494319A
Authority
CN
China
Prior art keywords
image
sand dune
layer
remote sensing
image set
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
Application number
CN202210264843.7A
Other languages
Chinese (zh)
Inventor
罗伦
刘用
赵凤鸣
毕云香
李丽
徐孟杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guojiao Space Information Technology Beijing Co ltd
Original Assignee
Guojiao Space Information Technology Beijing Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guojiao Space Information Technology Beijing Co ltd filed Critical Guojiao Space Information Technology Beijing Co ltd
Priority to CN202210264843.7A priority Critical patent/CN114494319A/en
Publication of CN114494319A publication Critical patent/CN114494319A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • 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
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Image Analysis (AREA)

Abstract

The invention provides a sand dune shape extraction method and device based on remote sensing images, wherein the method comprises the following steps: acquiring a remote sensing image of a target sand dune; cutting the remote sensing image into a standard image set with a preset size; preprocessing the standard image set to generate an image set to be identified; inputting the image set to be recognized into a pre-trained sand dune shape extraction model according to a preset sequence, and outputting a corresponding recognition image set; and overlapping and cutting the identification image set, and splicing according to the preset sequence to generate an identification result image, wherein the identification result image comprises the edge shape of the target sand dune. In this way, the sand dune position information can be accurately grasped, and the sand dune shape can be accurately predicted at the same time.

Description

Sand dune shape extraction method and device based on remote sensing image
Technical Field
Embodiments of the present disclosure relate generally to the field of remote sensing image processing data preprocessing technology, and more particularly, to a remote sensing image-based sand dune shape extraction method and apparatus.
Background
Desert areas increasingly face imminent threats and are susceptible to influences of climate change, urbanization, land degradation, water shortage and the like. Deserts are extremely important habitats for specialized plant and zoo populations. In addition, deserts possess a wide range of minerals such as lithium, boron, etc. Knowledge of the changes that occur at the boundaries of a sand dune and their orientation will also aid in the ability to predict the movement of the sand dune. From a human perspective, it is a very dangerous natural behavior if a dune is changing directions and moving towards a densely populated direction. Therefore, it is necessary to accurately monitor the desert to make and implement various effective measures to monitor the sand dune, however, most of the prior methods for extracting sand dunes focus on manually outlining or extracting sand dune outlines by using semi-automatic programs and manually modifying the contours, which requires a lot of labor cost and cannot extract semantic information of the sand dune in a wide range and in an all-round manner.
When a traditional edge extraction operator such as a Canny operator extracts a sand dune contour, the extracted sand dune contour information has a lot of noise points due to weak pertinence, and although noise points of a contour extraction algorithm based on deep learning such as HED and RCF are reduced when the sand dune information is extracted, the noise points cannot accurately extract sand dune semantic information, so that the defect exists. Under the background of lacking accurate semantic information extraction, although a method proposes to extract a sand dune by using a Unet network, the network is too simple, the number of times of common convolution of an up-sampling part and a down-sampling part is large, the loss of characteristic diagram information is large, the position information of a sand dune characteristic diagram cannot be accurately grasped, the accurate semantic information of the sand dune in a desert cannot be fully extracted, and the key is that how to train the network by using effective samples and accurately predict the large-amplitude remote sensing image is considered to be small in the actual scene.
Disclosure of Invention
According to the embodiment of the disclosure, a sand dune shape extraction scheme based on a remote sensing image is provided, sand dune position information can be accurately grasped, and meanwhile, the sand dune shape can be accurately predicted.
In a first aspect of the present disclosure, a remote sensing image-based sand dune shape extraction method is provided, including:
acquiring a remote sensing image of a target sand dune;
cutting the remote sensing image into a standard image set with a preset size;
preprocessing the standard image set to generate an image set to be identified;
inputting the image set to be recognized into a pre-trained sand dune shape extraction model according to a preset sequence, and outputting a corresponding recognition image set;
and overlapping and cutting the identification image set, and splicing according to the preset sequence to generate an identification result image, wherein the identification result image comprises the edge shape of the target sand dune.
In some embodiments, the preprocessing the standard image set to generate an image set to be identified includes:
and carrying out filtering and denoising on the standard image set, and replacing the gray value of one point in the standard image with the neighborhood median of the point.
In some embodiments, the filtering denoising the standard image set comprises:
for remote sensing images, { x }ij,(i,j)∈I2Expressing the gray value of each point of the digital image, and the formula of the image median filtering with a filtering window of A is as follows:
y＝Med{xij}＝Med{x(i+r),(j+s)(r,s)∈A(i,j)∈I2}
where Med { } is the median of the elements in the set, { xij,(i,j)∈I2Expressing the gray value of each pixel point of the remote sensing image, I expressing the size of a neighborhood window of the image, I being an odd number, A (I, j) representing an image area to be filtered in the image, and (r, s) representing the neighborhood windowThere is a collection of pixels.
In some embodiments, the sand dune shape extraction model comprises:
the depth residual error expansion convolutional coding part comprises five layers, and the dimension of the output characteristic matrix of each layer is added with the output characteristic matrix of the previous layer after being convolved by the transpose of the layer-by-layer connection decoding part layer by layer to form a layer-by-layer connection network decoding structure.
In some embodiments, the depth residual expansion convolutional coding part comprises a five-layer structure, wherein the first layer structure comprises a three-layer convolutional layer with a convolutional kernel size of 3 x 3, and a normalization layer and an activation layer which are arranged after the three-layer convolutional layer; the second layer structure is different from the first layer structure only in that the sizes of convolution kernels are 1 x 1, 3 x 3 and 1 x 1 respectively, and the first layer structure and the second layer structure are connected through a maximum pooling layer; the third layer structure is the same as the second layer structure in structure, and a characteristic diagram obtained by adding residuals of the input characteristic diagram and the output characteristic diagram of the second layer structure is used as the input characteristic diagram of the third layer structure; the fourth layer structure comprises one convolution layer with convolution kernel size of 1 x 1 and expansion convolution layers with expansion factors of 1, 2 and 5 respectively, and the fifth layer structure and the fourth layer structure have the same structure.
In some embodiments, the sand dune shape extraction model is trained by:
taking a preset number of image sets to be recognized, which are marked with sand dune edge shapes in advance, as training samples, training a sand dune shape extraction model which is constructed in advance, and outputting a feature map with the sand dune shape as a recognition result;
comparing the sand dune shape in the recognition result with a pre-marked sand dune edge shape, and adjusting parameters of each layer in a pre-built sand dune shape extraction model in response to the fact that the error between the sand dune shape in the recognition result and the pre-marked sand dune edge shape is larger than a preset threshold value;
and repeating the process until the error between the sand dune shape in the recognition result and the pre-marked sand dune edge shape in the recognition result is smaller than a preset threshold value, and finishing the training of the sand dune shape extraction model.
In some embodiments, the cropping the remote sensing image, and the cropping the remote sensing image into a standard image set with a preset size includes:
cutting the remote sensing image with overlap to make the overlap ratio between adjacent standard images generated after cutting 1-1/r2Wherein r is the expansion ratio in the expanded convolutional layer.
In a second aspect of the present disclosure, there is provided a remote sensing image-based sand dune shape extraction device, including:
the image acquisition module is used for acquiring a remote sensing image of a target sand dune;
the image cutting module is used for cutting the remote sensing image into a standard image set with a preset size;
the image preprocessing module is used for preprocessing the standard image set to generate an image set to be identified;
the image recognition module is used for inputting the image set to be recognized into a pre-trained sand dune shape extraction model according to a preset sequence and outputting a corresponding recognition image set;
and the image splicing module is used for performing overlapping cutting on the identification image set and splicing according to the preset sequence to generate an identification result image, wherein the identification result image comprises the edge shape of the target sand dune.
In a third aspect of the present disclosure, an electronic device is provided, comprising a memory having stored thereon a computer program and a processor implementing the method as described above when executing the program.
In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out the method as set forth above.
Through the sand dune shape extraction method based on the remote sensing image, sand dune position information can be accurately grasped, and meanwhile the sand dune shape can be accurately predicted.
The statements made in this summary are not intended to limit key or critical features of the embodiments of the disclosure, nor are they intended to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. In the drawings, like or similar reference characters designate like or similar elements, and wherein:
fig. 1 shows a flow chart of a remote sensing image-based sand dune shape extraction method according to an embodiment of the present disclosure;
fig. 2 shows a schematic structural diagram of a remote sensing image-based sand dune shape extraction device according to an embodiment of the disclosure;
fig. 3 shows a schematic structural diagram of a remote sensing image-based sand dune shape extraction device according to an embodiment of the present disclosure;
FIG. 4 illustrates an example flow chart of a remote sensing image-based sand dune shape extraction method of an embodiment of the present disclosure;
FIG. 5 shows a schematic structural diagram of a sand dune shape extraction model of an embodiment of the present disclosure;
FIG. 6 illustrates an image overlay cropping schematic of an embodiment of the present disclosure.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are some, but not all embodiments of the present disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
In addition, the term "and/or" herein is only one kind of association relationship describing an associated object, and means that there may be three kinds of relationships, for example, a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
The sand dune shape extraction method based on the remote sensing image can accurately grasp sand dune position information and accurately predict the sand dune shape.
Specifically, as shown in fig. 1, the method is a flowchart of a remote sensing image-based sand dune shape extraction method according to an embodiment of the present disclosure. Fig. 4 shows an example flowchart of a remote sensing image-based sand dune shape extraction method according to an embodiment of the present disclosure. Referring to fig. 1 and 4, the remote sensing image-based sand dune shape extraction method may include the following steps:
s101: and acquiring a remote sensing image of the target sand dune.
The sand dune shape extraction method based on the remote sensing image can be applied to sand dune shape extraction in a desert area, for example, crescent sand dune shape extraction, and through extracting the sand dune shape, the movement of the sand dune can be predicted, so that the monitoring of the sand dune is realized. Of course, the method of the embodiment of the present disclosure may also be applied to shape extraction of other terrains, such as rivers, lakes, etc.
In this embodiment, when extracting a sand dune shape using the method of the present disclosure, it is first necessary to acquire a remote sensing image of a target sand dune. And preferably, the remote sensing image is a high-resolution remote sensing image. The remote sensing image comprises an image of a target sand dune.
S102: and cutting the remote sensing image into a standard image set with a preset size.
In the model prediction process, if a large remote sensing image to be classified is directly input into a network model, memory overflow can be caused, so that generally, an image to be classified is cut into a series of small images which are respectively input into a network for prediction, and then prediction results are spliced into a final result image according to the cutting sequence. Corresponding to the layer-by-layer connection depth residual error expansion convolution network model, the method of the embodiment of the disclosure adopts the edge neglecting prediction, namely, the image is cut in an overlapped mode and the edge neglecting strategy is adopted during splicing.
In this embodiment, after the remote sensing image of the target sand dune is acquired, the acquired remote sensing image may be further cut, and the remote sensing image may be cut into a standard image set with a preset size. In the method of the embodiment of the disclosure, the neural network model is used for extracting the sand dune shape, so that the remote sensing image needs to be cut into a standard image suitable for neural network input, in order to fully utilize the characteristic diagram information of each layer of the neural network model and provide a larger receptive field for the sand dune remote sensing image, and the sand dune position information is accurately grasped and classified when the sand dune information is extracted, so that an accurate prediction result is obtained, the neural network model of the embodiment of the disclosure adopts a layer-by-layer connection depth residual error cavity convolution network. In order to adapt to the layer-by-layer connection of the depth residual error cavity convolution network, the remote sensing image needs to be cut, and the remote sensing image is cut into a standard image set with a preset size.
Fig. 6 is a schematic diagram of image overlay cropping according to an embodiment of the present disclosure. In the present embodiment, when the remote sensing image is cut, it is necessary to perform overlay cutting, and the dotted line portion in fig. 6 indicates the size of the image after overlay cutting, and the solid line portion in fig. 6 indicates the size of the useful information image.
As shown in fig. 5, a schematic structural diagram of the sand dune shape extraction model of the embodiment of the present disclosure is shown in fig. 5. The sand dune-shaped extraction model comprises a depth residual error expansion convolution coding part and a layer-by-layer connection decoding part, wherein the depth residual error expansion convolution coding part comprises five layers, and the dimension of an output characteristic matrix of each layer is added with an output characteristic matrix of the previous layer after being subjected to transposition convolution of the layer-by-layer connection decoding part layer by layer to form a layer-by-layer connection network decoding structure.
The depth residual error expansion convolution coding part comprises five layers of structures, wherein the first layer of structure comprises three convolution layers with convolution kernel size of 3 x 3, a normalization layer and an activation layer which are arranged behind the three convolution layers; the second layer structure is different from the first layer structure only in that the sizes of convolution kernels are 1 x 1, 3 x 3 and 1 x 1 respectively, and the first layer structure and the second layer structure are connected through a maximum pooling layer; the third layer structure is the same as the second layer structure in structure, and a characteristic diagram obtained by adding residuals of the input characteristic diagram and the output characteristic diagram of the second layer structure is used as the input characteristic diagram of the third layer structure; the fourth layer structure comprises a convolution layer with convolution kernel size of 1 x 1 and expansion convolution layers with expansion factors of 1, 2 and 5 respectively, and the fifth layer structure and the fourth layer structure have the same structure.
Where the dilation convolution is defined as the following equation, where f (i) is the input information, g (i) is the output signal, h (i) represents a filter of length L, r corresponds to the dilation rate, and in standard convolution r is 1:
Figure BDA0003551259530000081
and, the sand dune shape extraction model is trained by the following method:
taking a preset number of image sets to be recognized, which are marked with sand dune edge shapes in advance, as training samples, training a sand dune shape extraction model which is constructed in advance, and outputting a feature map with the sand dune shape as a recognition result;
comparing the sand dune shape in the recognition result with a pre-marked sand dune edge shape, and adjusting parameters of each layer in a pre-built sand dune shape extraction model in response to the fact that the error between the sand dune shape in the recognition result and the pre-marked sand dune edge shape is larger than a preset threshold value;
and repeating the process until the error between the sand dune shape in the recognition result and the pre-marked sand dune edge shape in the recognition result is smaller than a preset threshold value, and finishing the training of the sand dune shape extraction model.
S103: and preprocessing the standard image set to generate an image set to be identified.
After the remote sensing image is cut to generate a standard image set, preprocessing the standard image in the standard image set to generate an image set to be identified. Wherein preprocessing the standard image set comprises: and carrying out filtering and denoising on the standard image set, and replacing the gray value of one point in the standard image with the neighborhood median of the point.
In particular, for remote sensing images, { x }ij,(i,j)∈I2Expressing the gray value of each point of the digital image, and the formula of the image median filtering with a filtering window of A is as follows:
y＝Med{xij}＝Med{x(i+r),(j+s)(r,s)∈A(i,j)∈I2}
where Med { } is the median of the elements in the set, { xij,(i,j)∈I2The gray value of each pixel point of the remote sensing image is represented, I represents the size of a neighborhood window of the image, I is an odd number, A (I, j) represents an image area to be filtered in the image, and (r, s) represents a set of all pixel points of the neighborhood window.
And (3) median filtering is used for the cut image, the sand dune remote sensing image is interfered by the formation of irregular gravels, sand ridge lines are in a discontinuous state, dot-shaped or plaque-shaped noise is formed on the slope surface of the sand dune, and denoising is carried out through a median filtering algorithm. The median filtering algorithm can eliminate noise and protect the edge of the image, and has an image restoration effect. Median filtering can remove isolated noise points (such as gravel and the like) and maintain edge feature shapes.
S104: and inputting the image set to be recognized into a pre-trained sand dune shape extraction model according to a preset sequence, and outputting a corresponding recognition image set.
In this embodiment, after the image set to be recognized is generated, the image set to be recognized may be input to a pre-trained sand dune shape extraction model according to a preset order, and a corresponding recognition image set may be output. The images to be recognized in the image set to be recognized can be numbered according to the sequence of cutting, so that the sequence of inputting the images to be recognized into the sand dune shape extraction model is determined. Alternatively, the order of input to the pre-trained sand dune shape extraction model may be determined according to other means. In this embodiment, the order of inputting the pre-trained sand dune shape extraction model is determined to determine the order of the output recognition images, and further determine the positions of the order of the recognition images in the final recognition result image, so as to facilitate splicing.
S105: and overlapping and cutting the identification image set, and splicing according to the preset sequence to generate an identification result image, wherein the identification result image comprises the edge shape of the target sand dune.
In this embodiment, after the image to be recognized in the image set to be recognized is recognized and the recognition image set is generated, since the recognition image corresponds to the image to be recognized and the image to be recognized is cut in an overlapping manner, the recognition image also includes additional information, as shown in fig. 6, the recognition image is a square with a side length of a, and the useful information is a square with a side length of a, then a2-a2Is additional information. In the present embodiment, when the remote sensing image is cut to recognize or the recognition image is cut to produce a recognition result image, the remote sensing image is cut so as to overlap each other, and the overlap ratio between the adjacent standard images generated by cutting is 1 to 1/r2Wherein r is the expansion ratio in the expanded convolutional layer.
The sand dune shape extraction method based on the remote sensing image can accurately grasp sand dune position information and accurately predict the sand dune shape.
It is noted that while for simplicity of explanation, the foregoing method embodiments have been described as a series of acts or combination of acts, it will be appreciated by those skilled in the art that the present disclosure is not limited by the order of acts, as some steps may, in accordance with the present disclosure, occur in other orders and concurrently. Further, those skilled in the art should also appreciate that the embodiments described in this specification are all alternative embodiments and that the acts and modules involved are not necessarily essential to the disclosure.
The above is a description of embodiments of the method, and the embodiments of the apparatus are further described below.
Fig. 2 is a schematic structural diagram of a remote sensing image-based sand dune shape extraction device according to an embodiment of the present disclosure. The sand dune shape extraction device based on remote sensing image of this embodiment includes:
the image acquisition module 201 is used for acquiring a remote sensing image of a target sand dune;
the image cutting module 202 is used for cutting the remote sensing image into a standard image set with a preset size;
the image preprocessing module 203 is configured to preprocess the standard image set to generate an image set to be identified;
the image recognition module 204 is used for inputting the image set to be recognized into a pre-trained sand dune shape extraction model according to a preset sequence and outputting a corresponding recognition image set;
and the image splicing module 205 is configured to perform overlapping clipping on the recognition image set, and splice the recognition image set according to the preset sequence to generate a recognition result image, where the recognition result image includes an edge shape of the target sand dune.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the described module may refer to the corresponding process in the foregoing method embodiment, and is not described herein again.
FIG. 3 shows a schematic block diagram of an electronic device 300 that may be used to implement embodiments of the present disclosure. As shown, device 300 includes a Central Processing Unit (CPU)301 that may perform various appropriate actions and processes in accordance with computer program instructions stored in a Read Only Memory (ROM)302 or loaded from a storage unit 308 into a Random Access Memory (RAM) 303. In the RAM 303, various programs and data necessary for the operation of the device 300 can also be stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An input/output (I/O) interface 305 is also connected to bus 304.
Various components in device 300 are connected to I/O interface 305, including: an input unit 306 such as a keyboard, a mouse, or the like; an output unit 307 such as various types of displays, speakers, and the like; a storage unit 308 such as a magnetic disk, optical disk, or the like; and a communication unit 309 such as a network card, modem, wireless communication transceiver, etc. The communication unit 309 allows the device 300 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
The processing unit 301, which tangibly embodies a machine-readable medium, such as the storage unit 308, performs the various methods and processes described above. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 300 via ROM 302 and/or communication unit 309. When the computer program is loaded into the RAM 703 and executed by the CPU 301, one or more steps of the method described above may be performed. Alternatively, in other embodiments, the CPU 301 may be configured to perform the above-described method in any other suitable manner (e.g., by way of firmware).
The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a load programmable logic device (CPLD), and the like.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
Further, while operations are depicted in a particular order, this should be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single implementation. Conversely, various features that are described in the context of a single implementation can also be implemented in multiple implementations separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims (10)

1. A sand dune shape extraction method based on remote sensing images is characterized by comprising the following steps:
acquiring a remote sensing image of a target sand dune;
cutting the remote sensing image into a standard image set with a preset size;
preprocessing the standard image set to generate an image set to be identified;
inputting the image set to be recognized into a pre-trained sand dune shape extraction model according to a preset sequence, and outputting a corresponding recognition image set;
and overlapping and cutting the identification image set, and splicing according to the preset sequence to generate an identification result image, wherein the identification result image comprises the edge shape of the target sand dune.
2. The sand dune shape extraction method according to claim 1, wherein said preprocessing said standard image set to generate an image set to be recognized comprises:
and carrying out filtering and denoising on the standard image set, and replacing the gray value of one point in the standard image with the neighborhood median of the point.
3. The sand dune shape extraction method according to claim 2, wherein said filtering and denoising said standard image set comprises:
for remote sensing images, { x }ij,(i,j)∈I2Expressing the gray value of each point of the digital image, and the formula of the image median filtering with a filtering window of A is as follows:
y＝Med{xij}＝Med{x(i+r),(j+s)(r,s)∈A(i,j)∈I2}
where Med { } is the median of the elements in the set, { xij,(i,j)∈I2The gray value of each pixel point of the remote sensing image is represented, I represents the size of a neighborhood window of the image, I is an odd number, A (I, j) represents an image area to be filtered in the image, and (r, s) represents a set of all pixel points of the neighborhood window.
4. The sand dune shape extraction method according to claim 1, wherein said sand dune shape extraction model includes:
the depth residual error expansion convolutional coding part comprises five layers, and the dimension of the output characteristic matrix of each layer is added with the output characteristic matrix of the previous layer after being convoluted by transpose of the layer-by-layer connection decoding part layer by layer to form a layer-by-layer connection network decoding structure.
5. The sand dune shape extraction method according to claim 4, wherein said depth residual expansion convolution coding part comprises a five-layer structure, wherein the first layer structure comprises three convolution layers with convolution kernel size of 3 x 3, and a normalization layer and an activation layer disposed after the three convolution layers; the second layer structure is different from the first layer structure only in that the sizes of convolution kernels are 1 x 1, 3 x 3 and 1 x 1 respectively, and the first layer structure and the second layer structure are connected through a maximum pooling layer; the third layer structure is the same as the second layer structure in structure, and a characteristic diagram obtained by adding residuals of the input characteristic diagram and the output characteristic diagram of the second layer structure is used as the input characteristic diagram of the third layer structure; the fourth layer structure comprises one convolution layer with convolution kernel size of 1 x 1 and expansion convolution layers with expansion factors of 1, 2 and 5 respectively, and the fifth layer structure and the fourth layer structure have the same structure.
6. The sand dune shape extraction method according to claim 4, wherein said sand dune shape extraction model is trained by:
taking a preset number of image sets to be recognized, which are marked with sand dune edge shapes in advance, as training samples, training a sand dune shape extraction model which is constructed in advance, and outputting a feature map with the sand dune shape as a recognition result;
comparing the sand dune shape in the recognition result with a pre-marked sand dune edge shape, and adjusting parameters of each layer in a pre-built sand dune shape extraction model in response to the fact that the error between the sand dune shape in the recognition result and the pre-marked sand dune edge shape is larger than a preset threshold value;
and repeating the process until the error between the sand dune shape in the recognition result and the pre-marked sand dune edge shape in the recognition result is smaller than a preset threshold value, and finishing the training of the sand dune shape extraction model.
7. The sand dune shape extraction method according to claim 6, wherein said cutting the remote sensing image to cut the remote sensing image into a standard image set with a preset size comprises:
cutting the remote sensing image with overlap to make the overlap ratio between adjacent standard images generated after cutting 1-1/r2Wherein r is the expansion ratio in the expanded convolutional layer.
8. Sand dune shape extraction element based on remote sensing image, its characterized in that includes:
the image acquisition module is used for acquiring a remote sensing image of a target sand dune;
the image cutting module is used for cutting the remote sensing image into a standard image set with a preset size;
the image preprocessing module is used for preprocessing the standard image set to generate an image set to be identified;
the image recognition module is used for inputting the image set to be recognized into a pre-trained sand dune shape extraction model according to a preset sequence and outputting a corresponding recognition image set;
and the image splicing module is used for performing overlapping cutting on the identification image set and splicing according to the preset sequence to generate an identification result image, wherein the identification result image comprises the edge shape of the target sand dune.
9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program, wherein the processor, when executing the program, implements the method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the method according to any one of claims 1 to 7.
CN202210264843.7A 2022-03-17 2022-03-17 Sand dune shape extraction method and device based on remote sensing image Pending CN114494319A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210264843.7A CN114494319A (en) 2022-03-17 2022-03-17 Sand dune shape extraction method and device based on remote sensing image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210264843.7A CN114494319A (en) 2022-03-17 2022-03-17 Sand dune shape extraction method and device based on remote sensing image

Publications (1)

Publication Number Publication Date
CN114494319A true CN114494319A (en) 2022-05-13

Family

ID=81487116

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210264843.7A Pending CN114494319A (en) 2022-03-17 2022-03-17 Sand dune shape extraction method and device based on remote sensing image

Country Status (1)

Country Link
CN (1) CN114494319A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116503614A (en) * 2023-04-27 2023-07-28 杭州食方科技有限公司 Dinner plate shape feature extraction network training method and dinner plate shape information generation method

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116503614A (en) * 2023-04-27 2023-07-28 杭州食方科技有限公司 Dinner plate shape feature extraction network training method and dinner plate shape information generation method

Similar Documents

Publication Publication Date Title
CN108510451B (en) Method for reconstructing license plate based on double-layer convolutional neural network
CN110084238B (en) Finger vein image segmentation method and device based on LadderNet network and storage medium
CN111476343B (en) Method and apparatus for utilizing masking parameters
CN115546768B (en) Pavement marking identification method and system based on multi-scale mechanism and attention mechanism
CN112749578A (en) Remote sensing image automatic road extraction method based on deep convolutional neural network
CN111179196B (en) Multi-resolution depth network image highlight removing method based on divide-and-conquer
CN114359130A (en) Road crack detection method based on unmanned aerial vehicle image
CN109064402B (en) Single image super-resolution reconstruction method based on enhanced non-local total variation model prior
CN116912257B (en) Concrete pavement crack identification method based on deep learning and storage medium
CN113887472A (en) Remote sensing image cloud detection method based on cascade color and texture feature attention
CN114764856A (en) Image semantic segmentation method and image semantic segmentation device
CN114822033B (en) Road network traffic flow data restoration method and system based on feature pyramid network
CN113177956B (en) Semantic segmentation method for unmanned aerial vehicle remote sensing image
CN112734675B (en) Image rain removing method based on pyramid model and non-local enhanced dense block
CN117726954A (en) Sea-land segmentation method and system for remote sensing image
CN113920147A (en) Remote sensing image building extraction method and device based on deep learning
CN111914596A (en) Lane line detection method, device, system and storage medium
CN111260564A (en) Image processing method and device and computer storage medium
CN111445446B (en) Concrete surface crack detection method based on improved U-net
CN117788811A (en) Extraction method and device for mine earth surface damage cracks, electronic equipment and medium
CN112560719A (en) High-resolution image water body extraction method based on multi-scale convolution-multi-core pooling
CN116912700A (en) River and lake regulation violation problem change detection method based on feature enhancement and refinement
CN114494319A (en) Sand dune shape extraction method and device based on remote sensing image
CN112183359B (en) Method, device and equipment for detecting violent content in video
CN115311522A (en) Target detection method and device for automatic driving, electronic equipment and medium

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