CN115100541B - Satellite remote sensing data processing method, system and cloud platform - Google Patents

Abstract

The invention provides a satellite remote sensing data processing method, a satellite remote sensing data processing system and a cloud platform, and relates to the technical field of data processing. According to the method, the satellite remote sensing image to be processed is segmented according to the preset image segmentation size, so that multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image are output. And performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images so as to output the image similarity between every two frames of satellite remote sensing sub-images. And according to the image similarity between every two frames of satellite remote sensing sub-images, classifying and storing the multi-frame satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images. Based on the method, the problem of poor storage effect of satellite remote sensing data in the prior art can be solved.

Description

Satellite remote sensing data processing method, system and cloud platform

Technical Field

The invention relates to the technical field of data processing, in particular to a satellite remote sensing data processing method, a satellite remote sensing data processing system and a cloud platform.

Background

On the basis of successful transmission of various remote sensing earth observation satellites, multi-resolution, massive and real-time earth observation data related to the earth and various resource environments thereof are continuously acquired. Moreover, with the development of remote sensing technology, the data size is also increasing at a remarkable speed, so that a large amount of satellite remote sensing data, namely satellite remote sensing images, are stored. When the satellite remote sensing image is stored, the satellite remote sensing image is generally segmented and then stored according to the corresponding positions, so that the problem of poor storage effect may exist.

Disclosure of Invention

Accordingly, the present invention is directed to a method, a system and a cloud platform for processing satellite remote sensing data, so as to solve the problem of poor storage effect of satellite remote sensing data in the prior art.

In order to achieve the above purpose, the embodiment of the present invention adopts the following technical scheme:

the processing method of the satellite remote sensing data is applied to a cloud platform and comprises the following steps:

according to a preset image segmentation size, carrying out segmentation operation on a satellite remote sensing image to be processed so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image, wherein the multi-frame satellite remote sensing sub-images are spliced to form the satellite remote sensing image;

performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images to output image similarity between every two frames of satellite remote sensing sub-images;

and carrying out classified storage operation on the multi-frame satellite remote sensing sub-images according to the image similarity between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

In some preferred embodiments, in the method for processing satellite remote sensing data, the step of performing a segmentation operation on a satellite remote sensing image to be processed according to a preset image segmentation size to output a multi-frame satellite remote sensing sub-image corresponding to the satellite remote sensing image includes:

Under the condition of receiving an original satellite remote sensing image, identifying whether a target storage request instruction for classifying and storing the original satellite remote sensing image is received or not;

under the condition that a target storage request instruction for classifying and storing the original satellite remote sensing image is identified to be received, marking the original satellite remote sensing image as a satellite remote sensing image to be processed;

and carrying out segmentation operation on the satellite remote sensing image to be processed according to the preset image segmentation size so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image, wherein the image size of each frame of satellite remote sensing sub-image is equal to the image segmentation size.

In some preferred embodiments, in the method for processing satellite remote sensing data, the step of marking the original satellite remote sensing image as a satellite remote sensing image to be processed if it is identified that a target storage request instruction for storing the original satellite remote sensing image in a classified manner is received includes:

under the condition that a target storage request instruction for classifying and storing the original satellite remote sensing image is identified to be received, determining the image size of the original satellite remote sensing image so as to output the target image size corresponding to the original satellite remote sensing image;

Performing size comparison operation on the target image size and an image size reference value;

and under the condition that the size of the target image is smaller than or equal to the image size reference value, determining that the original satellite remote sensing image is not marked as the satellite remote sensing image to be processed, and under the condition that the size of the target image is larger than the image size reference value, marking the original satellite remote sensing image as the satellite remote sensing image to be processed.

In some preferred embodiments, in the above satellite remote sensing data processing method, the step of performing a similarity calculation operation on each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images to output an image similarity between each two frames of satellite remote sensing sub-images includes:

performing image marking operation on one of the two frames of satellite remote sensing sub-images to form a target satellite remote sensing sub-image, and performing image marking operation on the other of the two frames of satellite remote sensing sub-images to form a reference satellite remote sensing sub-image;

performing object recognition operation on the target satellite remote sensing sub-image to output a target object type set, and performing object recognition operation on the reference satellite remote sensing sub-image to output a reference object type set, wherein each target object type included in the target object type set belongs to an object type corresponding to one object recognized in the target satellite remote sensing sub-image, and each reference object type included in the reference object type set belongs to an object type corresponding to one object recognized in the reference satellite remote sensing sub-image;

Performing object contour extraction operation on the target satellite remote sensing sub-image to output a target object contour set, and performing object contour extraction operation on the reference satellite remote sensing sub-image to output a reference object contour set, wherein the target object contour set comprises the contour of each object extracted from the target satellite remote sensing sub-image, and the reference object contour set comprises the contour of each object extracted from the reference satellite remote sensing sub-image;

for each target object contour included in the target object contour set, performing contour similarity calculation on the target object contour and each reference object contour included in the reference object contour set, and performing mean value calculation operation on each contour similarity outputted by calculation to output a target contour similarity corresponding to the target object contour;

for each standard object type in a plurality of standard object types, performing a statistical operation on the occurrence frequency of the standard object type in the target object type set to output a first occurrence frequency corresponding to the standard object type, and performing a statistical operation on the occurrence frequency of the standard object type in the reference object type set to output a second occurrence frequency corresponding to the standard object type;

For each standard object type in a plurality of standard object types, according to the number of object types included in the target object type set and the number of object types included in the reference object type set, performing fusion operation on the first occurrence frequency and the second occurrence frequency corresponding to the standard object type to output corresponding target occurrence frequency, and then performing positive correlation value determination operation according to the target occurrence frequency to output type importance corresponding to the standard object type;

for each target object contour, updating the target contour similarity corresponding to the target object contour according to the type importance corresponding to the object type corresponding to the target object contour to output the updated contour similarity corresponding to the target object contour, and determining the type importance corresponding to the object type corresponding to the target object contour according to the minimum value of the plurality of type importance corresponding to the plurality of standard object types when the object type corresponding to the target object contour does not belong to any one of the plurality of standard object types;

and calculating and outputting the image similarity between the two frames of satellite remote sensing sub-images according to the updated contour similarity corresponding to each target object contour included in the target object contour set.

In some preferred embodiments, in the processing method of satellite remote sensing data, for each of the plurality of standard object types, according to the number of object types included in the set of target object types and the number of object types included in the set of reference object types, performing a fusion operation on a first occurrence frequency and a second occurrence frequency corresponding to the standard object type to output a corresponding target occurrence frequency, and then performing a determination operation of a positive correlation value according to the target occurrence frequency to output a type importance corresponding to the standard object type, the method includes:

according to the number of the object types included in the target object type set and the number of the object types included in the reference object type set, calculating and outputting a first weighting coefficient corresponding to the first occurrence frequency and a second weighting coefficient corresponding to the second occurrence frequency;

and for each standard object type in the plurality of standard object types, performing weighted average calculation operation on the first occurrence frequency and the second occurrence frequency corresponding to the standard object type according to the first weighting coefficient and the second weighting coefficient to output a target occurrence frequency corresponding to the standard object type, and performing positive correlation value determination operation according to the target occurrence frequency to output the type importance corresponding to the standard object type.

In some preferred embodiments, in the method for processing satellite remote sensing data, for each of the target object outlines, the step of updating the target outline similarity corresponding to the target object outline according to the type importance corresponding to the object type corresponding to the target object outline to output the updated outline similarity corresponding to the target object outline includes:

for each target object contour, determining whether an object type corresponding to the target object contour belongs to any one of the standard object types;

for each target object contour, marking a minimum value of a plurality of type importance levels corresponding to the plurality of standard object types as the type importance level corresponding to the target object contour when the object type corresponding to the target object contour is the type importance level not belonging to any standard object type of the plurality of standard object types, and marking the type importance level corresponding to the standard object type as the type importance level corresponding to the target object contour when the object type corresponding to the target object contour is the type importance level belonging to one standard object type of the plurality of standard object types;

And for each target object contour, performing product calculation operation on the type importance degree corresponding to the target object contour and the target contour similarity corresponding to the target object contour so as to output updated contour similarity corresponding to the target object contour.

In some preferred embodiments, in the above satellite remote sensing data processing method, the step of performing a classification storage operation on the multi-frame satellite remote sensing sub-image according to the image similarity between every two frames of satellite remote sensing sub-images to store the multi-frame satellite remote sensing sub-image includes:

for each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images, performing region distance calculation operation on geographic regions corresponding to the two frames of satellite remote sensing sub-images to output region distance values between the two frames of satellite remote sensing sub-images, and determining the region correlation values between the two frames of satellite remote sensing sub-images according to the region distance values between the two frames of satellite remote sensing sub-images to output region correlation values between the two frames of satellite remote sensing sub-images;

for each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images, carrying out fusion operation on the region correlation value between the two frames of satellite remote sensing sub-images and the image similarity between the two frames of satellite remote sensing sub-images so as to output the image correlation value between the two frames of satellite remote sensing sub-images;

And carrying out classified storage operation on the multi-frame satellite remote sensing sub-images according to the image correlation degree between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

In some preferred embodiments, in the method for processing satellite remote sensing data, the step of performing a classification storage operation on the multi-frame satellite remote sensing sub-image according to an image correlation between every two frames of satellite remote sensing sub-images to store the multi-frame satellite remote sensing sub-image includes:

step 1, classifying the multi-frame satellite remote sensing sub-images to form a plurality of remote sensing sub-image initial sets corresponding to the multi-frame satellite remote sensing sub-images, wherein each remote sensing sub-image initial set comprises at least one frame of satellite remote sensing sub-images;

step 2, performing set number statistics operation on the plurality of remote sensing sub-image initial sets to output initial set numbers corresponding to the plurality of remote sensing sub-image initial sets, and determining a first classification coefficient with a negative correlation according to the initial set numbers;

step 3, for each of the plurality of remote sensing sub-image initial sets, marking the remote sensing sub-image initial set as a first remote sensing sub-image initial set when the remote sensing sub-image initial set comprises one frame of satellite remote sensing sub-image, assigning a target image correlation corresponding to the first remote sensing sub-image initial set as a maximum image correlation, performing a size comparison operation on an image correlation and a correlation reference value between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set when the remote sensing sub-image initial set comprises at least two frames of satellite remote sensing sub-images, and marking the sub-image initial set as a first remote sensing sub-image initial set when the image correlation between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set is greater than or equal to the correlation reference value, and marking an average value of the image correlation between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set as the target sub-image correlation corresponding to the first remote sensing sub-image initial set;

Step 4, jumping to step 1 when at least one remote sensing sub-image initial set in the plurality of remote sensing sub-image initial sets is not marked as a first remote sensing sub-image initial set, and performing fusion operation on the target image correlation degree corresponding to each first remote sensing sub-image initial set when each remote sensing sub-image initial set in the plurality of remote sensing sub-image initial sets is marked as a first remote sensing sub-image initial set so as to output a second classification coefficient;

step 5, carrying out fusion operation on the first classification coefficient and the second classification coefficient to output a target classification coefficient, and comparing the target classification coefficient with a classification coefficient reference value;

and step 6, jumping to step 1 when the target classification coefficient is smaller than or equal to the classification coefficient reference value, and marking the plurality of remote sensing sub-image initial sets formed in the step 1 as a plurality of corresponding remote sensing sub-image sets when the target classification coefficient is larger than the classification coefficient reference value, and storing each remote sensing sub-image set respectively.

The embodiment of the invention also provides a processing system of the satellite remote sensing data, which is applied to the cloud platform and comprises the following steps:

The image segmentation module is used for carrying out segmentation operation on a satellite remote sensing image to be processed according to a preset image segmentation size so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image, and the multi-frame satellite remote sensing sub-images are spliced to form the satellite remote sensing image;

the similarity calculation module is used for performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images so as to output image similarity between every two frames of satellite remote sensing sub-images;

and the image classification storage module is used for performing classification storage operation on the multi-frame satellite remote sensing sub-images according to the image similarity between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

The embodiment of the invention also provides a cloud platform for executing the satellite remote sensing data processing method.

According to the satellite remote sensing data processing method, system and cloud platform provided by the embodiment of the invention, the satellite remote sensing image to be processed can be segmented according to the preset image segmentation size, so that the multi-frame satellite remote sensing sub-image corresponding to the satellite remote sensing image can be output. And performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images so as to output the image similarity between every two frames of satellite remote sensing sub-images. And according to the image similarity between every two frames of satellite remote sensing sub-images, classifying and storing the multi-frame satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images. Based on the foregoing, the image similarity between the satellite remote sensing sub-images formed by segmentation can be combined to perform classification storage operation on the satellite remote sensing sub-images, so that after storage, the satellite remote sensing sub-images are convenient to search (by classifying, for example, the satellite remote sensing sub-images with high image similarity are stored together as a class, so that satellite remote sensing sub-images with high image similarity are conveniently searched together, or the satellite remote sensing sub-images with low image similarity are obtained by searching together as a class, so that the diversity of the searched images can be improved, and the management and control effect on the storage can be improved to a certain extent, thereby solving the problem that the storage effect of satellite remote sensing data in the prior art is poor.

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

Fig. 1 is a structural block diagram of a cloud platform according to an embodiment of the present invention.

Fig. 2 is a flowchart illustrating steps involved in a satellite remote sensing data processing method according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of each module included in the satellite remote sensing data processing system according to the 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 embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. 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.

Referring to fig. 1, an embodiment of the present invention provides a cloud platform.

In detail, in some possible implementations, the cloud platform may include a memory and a processor. Wherein the memory and the processor may be electrically connected directly or indirectly to enable transmission or interaction of data. For example, electrical connection may be made to each other via one or more communication buses or signal lines. The memory may store at least one software functional module (computer program) that may exist in the form of software or firmware. The processor may be configured to execute the executable computer program stored in the memory, so as to implement the method for processing satellite remote sensing data provided by the embodiment of the present invention.

In detail, in some possible implementations, the Memory may be, but is not limited to, random access Memory (Random Access Memory, RAM), read Only Memory (ROM), programmable Read Only Memory (Programmable Read-Only Memory, PROM), erasable Read Only Memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable Read Only Memory (Electric Erasable Programmable Read-Only Memory, EEPROM), and the like.

In detail, in some possible implementations, the processor may be a general-purpose processor including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), a System on Chip (SoC), etc.; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

In detail, in some possible implementations, the cloud platform may be a server with data processing capabilities.

Referring to fig. 2, the embodiment of the invention further provides a satellite remote sensing data processing method, which can be applied to the cloud platform. The method steps defined by the flow related to the satellite remote sensing data processing method can be realized by the cloud platform.

The specific flow shown in fig. 2 will be described in detail.

Step S110, according to the preset image segmentation size, the satellite remote sensing image to be processed is segmented, so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image.

In the embodiment of the invention, the cloud platform can segment the satellite remote sensing image to be processed according to the preset image segmentation size so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image. The multi-frame satellite remote sensing sub-images (according to the corresponding relation formed by segmentation) can be spliced to form the satellite remote sensing image.

Step S120, performing similarity calculation on each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images, so as to output image similarity between each two frames of satellite remote sensing sub-images.

In the embodiment of the invention, the cloud platform can perform similarity calculation operation on each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images so as to output the image similarity between each two frames of satellite remote sensing sub-images.

Step S130, classifying and storing the multi-frame satellite remote sensing sub-images according to the image similarity between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

In the embodiment of the invention, the cloud platform can perform classified storage operation on the multi-frame satellite remote sensing sub-images according to the image similarity between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

Based on the method, namely the satellite remote sensing data processing method, the satellite remote sensing image to be processed can be segmented according to the preset image segmentation size, so that multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image can be output. And performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images so as to output the image similarity between every two frames of satellite remote sensing sub-images. And according to the image similarity between every two frames of satellite remote sensing sub-images, classifying and storing the multi-frame satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images. Based on the foregoing, the image similarity between the satellite remote sensing sub-images formed by segmentation can be combined to perform classification storage operation on the satellite remote sensing sub-images, so that after storage, the satellite remote sensing sub-images are convenient to search (by classifying, for example, the satellite remote sensing sub-images with high image similarity are stored together as a class, so that satellite remote sensing sub-images with high image similarity are conveniently searched together, or the satellite remote sensing sub-images with low image similarity are obtained by searching together as a class, so that the diversity of the searched images can be improved, and the management and control effect on the storage can be improved to a certain extent, thereby solving the problem that the storage effect of satellite remote sensing data in the prior art is poor.

In detail, in some possible implementation manners, in step S110 included in the method, the following details may be further included:

under the condition of receiving an original satellite remote sensing image, identifying whether a target storage request instruction for classifying and storing the original satellite remote sensing image is received or not;

under the condition that a target storage request instruction for classifying and storing the original satellite remote sensing image is identified to be received, marking the original satellite remote sensing image as a satellite remote sensing image to be processed;

and carrying out segmentation operation on the satellite remote sensing image to be processed according to the preset image segmentation size so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image, wherein the image size of each frame of satellite remote sensing sub-image is equal to the image segmentation size.

In detail, in some possible implementation manners, the step of marking the original satellite remote sensing image as the satellite remote sensing image to be processed when the step of identifying that the target storage request instruction for classifying and storing the original satellite remote sensing image is received may further include the following specific contents:

Under the condition that a target storage request instruction for classifying and storing the original satellite remote sensing image is identified to be received, determining the image size of the original satellite remote sensing image so as to output the target image size corresponding to the original satellite remote sensing image;

performing size comparison operation on the target image size and an image size reference value;

and under the condition that the size of the target image is smaller than or equal to the image size reference value, determining that the original satellite remote sensing image is not marked as the satellite remote sensing image to be processed, and under the condition that the size of the target image is larger than the image size reference value, marking the original satellite remote sensing image as the satellite remote sensing image to be processed.

In detail, in some possible implementation manners, in step S120 included in the above method, the following details may be further included:

performing image marking operation on one of the two frames of satellite remote sensing sub-images to form a target satellite remote sensing sub-image, and performing image marking operation on the other of the two frames of satellite remote sensing sub-images to form a reference satellite remote sensing sub-image;

Performing object recognition operation on the target satellite remote sensing sub-image to output a target object type set, and performing object recognition operation on the reference satellite remote sensing sub-image to output a reference object type set, wherein each target object type included in the target object type set belongs to an object type corresponding to one object recognized in the target satellite remote sensing sub-image, and each reference object type included in the reference object type set belongs to an object type corresponding to one object recognized in the reference satellite remote sensing sub-image;

performing object contour extraction operation on the target satellite remote sensing sub-image to output a target object contour set, and performing object contour extraction operation on the reference satellite remote sensing sub-image to output a reference object contour set, wherein the target object contour set comprises the contour of each object extracted from the target satellite remote sensing sub-image, and the reference object contour set comprises the contour of each object extracted from the reference satellite remote sensing sub-image;

for each target object contour included in the target object contour set, performing contour similarity calculation on the target object contour and each reference object contour included in the reference object contour set, and performing mean value calculation operation on each contour similarity outputted by calculation to output a target contour similarity corresponding to the target object contour;

For each standard object type in a plurality of standard object types, performing a statistical operation on the occurrence frequency of the standard object type in the target object type set to output a first occurrence frequency corresponding to the standard object type (namely, the occurrence frequency of the standard object type), and performing a statistical operation on the occurrence frequency of the standard object type in the reference object type set to output a second occurrence frequency corresponding to the standard object type;

for each standard object type in a plurality of standard object types, according to the number of object types included in the target object type set and the number of object types included in the reference object type set, performing fusion operation on the first occurrence frequency and the second occurrence frequency corresponding to the standard object type to output corresponding target occurrence frequency, and then performing positive correlation value determination operation according to the target occurrence frequency to output type importance corresponding to the standard object type;

for each target object contour, updating the target contour similarity corresponding to the target object contour according to the type importance corresponding to the object type corresponding to the target object contour to output the updated contour similarity corresponding to the target object contour, and determining the type importance corresponding to the object type corresponding to the target object contour according to the minimum value of the plurality of type importance corresponding to the plurality of standard object types when the object type corresponding to the target object contour does not belong to any one of the plurality of standard object types;

And calculating and outputting the image similarity between the two frames of satellite remote sensing sub-images according to the updated contour similarity (for example, the average value of the updated contour similarity corresponding to each target object contour) corresponding to each target object contour included in the target object contour set.

In detail, in some possible implementation manners, the step of performing, for each standard object type of the plurality of standard object types, a fusion operation on the first occurrence frequency and the second occurrence frequency corresponding to the standard object type according to the number of object types included in the target object type set and the number of object types included in the reference object type set to output a corresponding target occurrence frequency, and performing a positive correlation value determining operation according to the target occurrence frequency to output a type importance corresponding to the standard object type according to the target occurrence frequency may further include the following specific contents:

calculating and outputting a first weighting coefficient corresponding to the first occurrence frequency and a second weighting coefficient corresponding to the second occurrence frequency according to the number of object types included in the target object type set and the number of object types included in the reference object type set (the first weighting coefficient is positively correlated with the number of object types included in the target object type set, and the second weighting coefficient is positively correlated with the number of object types included in the target reference object type set);

And for each standard object type in the plurality of standard object types, performing weighted average calculation operation on the first occurrence frequency and the second occurrence frequency corresponding to the standard object type according to the first weighting coefficient and the second weighting coefficient to output a target occurrence frequency corresponding to the standard object type, and performing positive correlation value determination operation according to the target occurrence frequency to output the type importance corresponding to the standard object type.

In detail, in some possible implementation manners, the step of updating, for each of the target object outlines, the target contour similarity corresponding to the target object outline according to the type importance corresponding to the object type corresponding to the target object outline to output the updated contour similarity corresponding to the target object outline, further includes the following specific contents:

for each target object contour, determining whether an object type corresponding to the target object contour belongs to any one of the standard object types;

for each target object contour, marking a minimum value of a plurality of type importance levels corresponding to the plurality of standard object types as the type importance level corresponding to the target object contour when the object type corresponding to the target object contour is the type importance level not belonging to any standard object type of the plurality of standard object types, and marking the type importance level corresponding to the standard object type as the type importance level corresponding to the target object contour when the object type corresponding to the target object contour is the type importance level belonging to one standard object type of the plurality of standard object types;

And for each target object contour, performing product calculation operation on the type importance degree corresponding to the target object contour and the target contour similarity corresponding to the target object contour so as to output updated contour similarity corresponding to the target object contour.

In detail, in some possible implementation manners, in step S120 included in the above method, the following details may be further included:

performing image marking operation on one of the two frames of satellite remote sensing sub-images to form a target satellite remote sensing sub-image, and performing image marking operation on the other of the two frames of satellite remote sensing sub-images to form a reference satellite remote sensing sub-image;

performing image difference operation on the target satellite remote sensing sub-image and a pre-configured standard satellite remote sensing sub-image to output a target difference image corresponding to the target satellite remote sensing sub-image, and performing image difference operation on the reference satellite remote sensing sub-image and the standard satellite remote sensing sub-image to output a reference difference image corresponding to the reference satellite remote sensing sub-image;

performing feature point identification operation on the target satellite remote sensing sub-image (the feature point identification operation may include an ORB feature point identification technique) to output a first feature point set corresponding to the target satellite remote sensing sub-image, and performing feature point identification operation on the target differential image to output a second feature point set corresponding to the target differential image;

Performing characteristic point identification operation on the reference satellite remote sensing sub-image to output a third characteristic point set corresponding to the reference satellite remote sensing sub-image, and performing characteristic point identification operation on the reference differential image to output a fourth characteristic point set corresponding to the reference differential image;

clustering the first feature points included in the first feature point set according to a preset number and pixel positions corresponding to each feature point to output a plurality of first feature point subsets with the preset number, clustering the second feature points included in the second feature point set according to the preset number and pixel positions corresponding to each feature point to output a plurality of second feature point subsets with the preset number, clustering the third feature points included in the third feature point set according to the preset number and pixel positions corresponding to each feature point to output a plurality of third feature point subsets with the preset number, and clustering the fourth feature points included in the fourth feature point set according to the preset number and pixel positions corresponding to each feature point to output a plurality of fourth feature point subsets with the preset number;

For each of the plurality of first feature point subsets, performing a center position determination operation on the first feature point subset according to a pixel position corresponding to each first feature point included in the first feature point subset, so as to output a first center position corresponding to the first feature point subset, and for each of the plurality of second feature point subsets, performing a center position determination operation on the second feature point subset according to a pixel position corresponding to each second feature point included in the second feature point subset, so as to output a second center position corresponding to the second feature point subset;

for each third feature point subset of the plurality of third feature point subsets, performing a center position determination operation on the third feature point subset according to a pixel position corresponding to each third feature point included in the third feature point subset to output a third center position corresponding to the third feature point subset, and for each fourth feature point subset of the plurality of fourth feature point subsets, performing a center position determination operation on the fourth feature point subset according to a pixel position corresponding to each fourth feature point included in the fourth feature point subset to output a fourth center position corresponding to the fourth feature point subset;

Performing one-to-one association operation on the plurality of first feature point subsets and the plurality of third feature point subsets (i.e., the average value of the distances between the associated first center position and third center position is minimum) according to the first center position corresponding to each first feature point subset and the third center position corresponding to each third feature point subset, and performing one-to-one association operation on the plurality of second feature point subsets and the plurality of fourth feature point subsets (i.e., the average value of the distances between the associated second center position and fourth center position is minimum) according to the second center position corresponding to each second feature point subset and the fourth center position corresponding to each fourth feature point subset;

for each first feature point subset, performing a correlation calculation operation on the first feature point subset and the third feature point subset according to a pixel position of each first feature point included in the first feature point subset and a pixel position of each third feature point included in a third feature point subset associated with the first feature point subset (for example, an average value of distances between the pixel positions may be first, and then a negative correlation coefficient corresponding to the average value may be used as a feature correlation) to output a feature correlation corresponding to the first feature point subset;

For each second feature point subset, performing correlation calculation operation (as described above) on the second feature point subset and the fourth feature point subset according to the pixel position of each second feature point included in the second feature point subset and the pixel position of each fourth feature point included in the fourth feature point subset associated with the second feature point subset, so as to output feature correlation corresponding to the second feature point subset;

and according to the feature correlation degree corresponding to each first feature point subset and the feature correlation degree corresponding to each second feature point subset (for example, first calculating a first average value of the feature correlation degree corresponding to each first feature point subset, then calculating a second average value of the feature correlation degree corresponding to each second feature point subset, and then carrying out weighted summation calculation on the first average value and the second average value), and fusing to obtain the image similarity between the two frames of satellite remote sensing sub-images.

In detail, in some possible implementation manners, in step S130 included in the above method, the following details may be further included:

for each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-image, performing region distance calculation operation on geographic regions corresponding to the two frames of satellite remote sensing sub-images to output region distance values between the two frames of satellite remote sensing sub-images, and determining region correlation values between the two frames of satellite remote sensing sub-images according to the region distance values between the two frames of satellite remote sensing sub-images to output region correlation values between the two frames of satellite remote sensing sub-images (the region correlation values can be inversely correlated with the region distance values);

For each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-image, performing fusion operation on the region correlation value between the two frames of satellite remote sensing sub-images and the image similarity between the two frames of satellite remote sensing sub-images (for example, performing weighted average calculation on the region correlation value and the image similarity) so as to output the image correlation value between the two frames of satellite remote sensing sub-images;

and carrying out classified storage operation on the multi-frame satellite remote sensing sub-images according to the image correlation degree between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

In detail, in some possible implementation manners, the step of performing a classification storage operation on the multi-frame satellite remote sensing sub-image according to the image correlation degree between every two frames of satellite remote sensing sub-images to store the multi-frame satellite remote sensing sub-image may further include the following specific contents:

step 1, classifying the multi-frame satellite remote sensing sub-images to form a plurality of remote sensing sub-image initial sets corresponding to the multi-frame satellite remote sensing sub-images, wherein each remote sensing sub-image initial set comprises at least one frame of satellite remote sensing sub-images;

Step 2, performing set number statistics operation on the plurality of remote sensing sub-image initial sets to output initial set numbers corresponding to the plurality of remote sensing sub-image initial sets, and determining a first classification coefficient with a negative correlation according to the initial set numbers;

step 3, for each of the plurality of remote sensing sub-image initial sets, marking the remote sensing sub-image initial set as a first remote sensing sub-image initial set when the remote sensing sub-image initial set comprises one frame of satellite remote sensing sub-image, assigning a target image correlation corresponding to the first remote sensing sub-image initial set as a maximum image correlation, performing a size comparison operation on an image correlation and a correlation reference value between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set when the remote sensing sub-image initial set comprises at least two frames of satellite remote sensing sub-images, and marking the sub-image initial set as a first remote sensing sub-image initial set when the image correlation between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set is greater than or equal to the correlation reference value, and marking an average value of the image correlation between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set as the target sub-image correlation corresponding to the first remote sensing sub-image initial set;

Step 4, jumping to step 1 when at least one remote sensing sub-image initial set in the plurality of remote sensing sub-image initial sets is not marked as a first remote sensing sub-image initial set, and performing fusion operation on the target image correlation degree corresponding to each first remote sensing sub-image initial set when each remote sensing sub-image initial set in the plurality of remote sensing sub-image initial sets is marked as a first remote sensing sub-image initial set so as to output a second classification coefficient;

step 5, carrying out fusion operation on the first classification coefficient and the second classification coefficient to output a target classification coefficient, and comparing the target classification coefficient with a classification coefficient reference value;

and step 6, jumping to step 1 when the target classification coefficient is smaller than or equal to the classification coefficient reference value, and when the target classification coefficient is greater than the classification coefficient reference value, marking the plurality of remote sensing sub-image initial sets formed in the last step 1 as corresponding plurality of remote sensing sub-image sets, and storing each remote sensing sub-image set (for example, for each remote sensing sub-image set, all satellite remote sensing sub-images included in the remote sensing sub-image set may be stored together, which may refer to one storage space stored in a plurality of storage spaces of one device or one storage device, or stored in different storage spaces, but respectively performing association processing so as to facilitate the association output together during the search).

Referring to fig. 3, the embodiment of the invention further provides a processing system of satellite remote sensing data, which can be applied to the cloud platform. The processing system of the satellite remote sensing data can comprise an image segmentation module, a similarity calculation module and an image classification storage module.

In detail, in some possible implementation manners, the image segmentation module is configured to perform a segmentation operation on a satellite remote sensing image to be processed according to a preset image segmentation size, so as to output a multi-frame satellite remote sensing sub-image corresponding to the satellite remote sensing image, where the multi-frame satellite remote sensing sub-image is spliced to form the satellite remote sensing image. The similarity calculation module is used for performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images so as to output image similarity between every two frames of satellite remote sensing sub-images, and the image classification storage module is used for performing classification storage operation on the multi-frame satellite remote sensing sub-images according to the image similarity between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

In summary, according to the method, the system and the cloud platform for processing satellite remote sensing data provided by the invention, the satellite remote sensing image to be processed can be subjected to the segmentation operation according to the preset image segmentation size, so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image. And performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images so as to output the image similarity between every two frames of satellite remote sensing sub-images. And according to the image similarity between every two frames of satellite remote sensing sub-images, classifying and storing the multi-frame satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images. Based on the foregoing, the image similarity between the satellite remote sensing sub-images formed by segmentation can be combined to perform a classification storage operation on the satellite remote sensing sub-images, so that after storage, the satellite remote sensing sub-images are convenient to search (by classifying, for example, the satellite remote sensing sub-images with high image similarity are stored together as a class, so that satellite remote sensing sub-images with high image similarity are conveniently searched together, or the satellite remote sensing sub-images with low image similarity are obtained by searching together as a class, so that the diversity of the searched images is improved, and the management and control effect on the storage is improved to a certain extent, thereby improving the problem of poor storage effect of satellite remote sensing data in the prior art.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (9)

1. The processing method of the satellite remote sensing data is characterized by being applied to a cloud platform, and comprises the following steps:

according to a preset image segmentation size, carrying out segmentation operation on a satellite remote sensing image to be processed so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image, wherein the multi-frame satellite remote sensing sub-images are spliced to form the satellite remote sensing image;

performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images to output image similarity between every two frames of satellite remote sensing sub-images;

according to the image similarity between every two frames of satellite remote sensing sub-images, classifying and storing the multi-frame satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images;

the step of classifying and storing the multi-frame satellite remote sensing sub-images according to the image similarity between every two frames of satellite remote sensing sub-images to store the multi-frame satellite remote sensing sub-images comprises the following steps:

For each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images, performing region distance calculation operation on geographic regions corresponding to the two frames of satellite remote sensing sub-images to output region distance values between the two frames of satellite remote sensing sub-images, and determining the region correlation values between the two frames of satellite remote sensing sub-images according to the region distance values between the two frames of satellite remote sensing sub-images to output region correlation values between the two frames of satellite remote sensing sub-images;

for each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images, carrying out fusion operation on the region correlation value between the two frames of satellite remote sensing sub-images and the image similarity between the two frames of satellite remote sensing sub-images so as to output the image correlation value between the two frames of satellite remote sensing sub-images;

and carrying out classified storage operation on the multi-frame satellite remote sensing sub-images according to the image correlation degree between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

2. The method for processing satellite remote sensing data according to claim 1, wherein the step of performing a segmentation operation on the satellite remote sensing image to be processed according to a preset image segmentation size to output a multi-frame satellite remote sensing sub-image corresponding to the satellite remote sensing image comprises the steps of:

Under the condition of receiving an original satellite remote sensing image, identifying whether a target storage request instruction for classifying and storing the original satellite remote sensing image is received or not;

under the condition that a target storage request instruction for classifying and storing the original satellite remote sensing image is identified to be received, marking the original satellite remote sensing image as a satellite remote sensing image to be processed;

and carrying out segmentation operation on the satellite remote sensing image to be processed according to the preset image segmentation size so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image, wherein the image size of each frame of satellite remote sensing sub-image is equal to the image segmentation size.

3. The method for processing satellite remote sensing data according to claim 2, wherein the step of marking the original satellite remote sensing image as a satellite remote sensing image to be processed in a case that it is recognized that a target storage request instruction for classifying and storing the original satellite remote sensing image is received, comprises:

under the condition that a target storage request instruction for classifying and storing the original satellite remote sensing image is identified to be received, determining the image size of the original satellite remote sensing image so as to output the target image size corresponding to the original satellite remote sensing image;

Performing size comparison operation on the target image size and an image size reference value;

and under the condition that the size of the target image is smaller than or equal to the image size reference value, determining that the original satellite remote sensing image is not marked as the satellite remote sensing image to be processed, and under the condition that the size of the target image is larger than the image size reference value, marking the original satellite remote sensing image as the satellite remote sensing image to be processed.

4. The method for processing satellite remote sensing data according to claim 1, wherein the step of performing a similarity calculation operation on each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images to output an image similarity between each two frames of satellite remote sensing sub-images comprises:

performing image marking operation on one of the two frames of satellite remote sensing sub-images to form a target satellite remote sensing sub-image, and performing image marking operation on the other of the two frames of satellite remote sensing sub-images to form a reference satellite remote sensing sub-image;

performing object recognition operation on the target satellite remote sensing sub-image to output a target object type set, and performing object recognition operation on the reference satellite remote sensing sub-image to output a reference object type set, wherein each target object type included in the target object type set belongs to an object type corresponding to one object recognized in the target satellite remote sensing sub-image, and each reference object type included in the reference object type set belongs to an object type corresponding to one object recognized in the reference satellite remote sensing sub-image;

Performing object contour extraction operation on the target satellite remote sensing sub-image to output a target object contour set, and performing object contour extraction operation on the reference satellite remote sensing sub-image to output a reference object contour set, wherein the target object contour set comprises the contour of each object extracted from the target satellite remote sensing sub-image, and the reference object contour set comprises the contour of each object extracted from the reference satellite remote sensing sub-image;

for each target object contour included in the target object contour set, performing contour similarity calculation on the target object contour and each reference object contour included in the reference object contour set, and performing mean value calculation operation on each contour similarity outputted by calculation to output a target contour similarity corresponding to the target object contour;

for each standard object type in a plurality of standard object types, performing a statistical operation on the occurrence frequency of the standard object type in the target object type set to output a first occurrence frequency corresponding to the standard object type, and performing a statistical operation on the occurrence frequency of the standard object type in the reference object type set to output a second occurrence frequency corresponding to the standard object type;

For each standard object type in a plurality of standard object types, according to the number of object types included in the target object type set and the number of object types included in the reference object type set, performing fusion operation on the first occurrence frequency and the second occurrence frequency corresponding to the standard object type to output corresponding target occurrence frequency, and then performing positive correlation value determination operation according to the target occurrence frequency to output type importance corresponding to the standard object type;

for each target object contour, updating the target contour similarity corresponding to the target object contour according to the type importance corresponding to the object type corresponding to the target object contour to output the updated contour similarity corresponding to the target object contour, and determining the type importance corresponding to the object type corresponding to the target object contour according to the minimum value of the plurality of type importance corresponding to the plurality of standard object types when the object type corresponding to the target object contour does not belong to any one of the plurality of standard object types;

and calculating and outputting the image similarity between the two frames of satellite remote sensing sub-images according to the updated contour similarity corresponding to each target object contour included in the target object contour set.

5. The method for processing satellite remote sensing data according to claim 4, wherein for each of the plurality of standard object types, the step of performing a fusion operation on the first occurrence frequency and the second occurrence frequency corresponding to the standard object type according to the number of object types included in the set of target object types and the number of object types included in the set of reference object types to output a corresponding target occurrence frequency, and performing a positive correlation value determination operation according to the target occurrence frequency to output a type importance corresponding to the standard object type includes:

according to the number of the object types included in the target object type set and the number of the object types included in the reference object type set, calculating and outputting a first weighting coefficient corresponding to the first occurrence frequency and a second weighting coefficient corresponding to the second occurrence frequency;

and for each standard object type in the plurality of standard object types, performing weighted average calculation operation on the first occurrence frequency and the second occurrence frequency corresponding to the standard object type according to the first weighting coefficient and the second weighting coefficient to output a target occurrence frequency corresponding to the standard object type, and performing positive correlation value determination operation according to the target occurrence frequency to output the type importance corresponding to the standard object type.

6. The method of claim 4, wherein for each of the target object profiles, the step of updating the target profile similarity corresponding to the target object profile according to the type importance corresponding to the object type corresponding to the target object profile to output the updated profile similarity corresponding to the target object profile comprises:

for each target object contour, determining whether an object type corresponding to the target object contour belongs to any one of the standard object types;

for each target object contour, marking a minimum value of a plurality of type importance levels corresponding to the plurality of standard object types as the type importance level corresponding to the target object contour when the object type corresponding to the target object contour is the type importance level not belonging to any standard object type of the plurality of standard object types, and marking the type importance level corresponding to the standard object type as the type importance level corresponding to the target object contour when the object type corresponding to the target object contour is the type importance level belonging to one standard object type of the plurality of standard object types;

And for each target object contour, performing product calculation operation on the type importance degree corresponding to the target object contour and the target contour similarity corresponding to the target object contour so as to output updated contour similarity corresponding to the target object contour.

7. The method for processing satellite remote sensing data according to claim 1, wherein the step of performing a classification storage operation on the multi-frame satellite remote sensing sub-images according to an image correlation between each two frames of satellite remote sensing sub-images to store the multi-frame satellite remote sensing sub-images comprises:

step 1, classifying the multi-frame satellite remote sensing sub-images to form a plurality of remote sensing sub-image initial sets corresponding to the multi-frame satellite remote sensing sub-images, wherein each remote sensing sub-image initial set comprises at least one frame of satellite remote sensing sub-images;

step 2, performing set number statistics operation on the plurality of remote sensing sub-image initial sets to output initial set numbers corresponding to the plurality of remote sensing sub-image initial sets, and determining a first classification coefficient with a negative correlation according to the initial set numbers;

step 3, for each of the plurality of remote sensing sub-image initial sets, marking the remote sensing sub-image initial set as a first remote sensing sub-image initial set when the remote sensing sub-image initial set comprises one frame of satellite remote sensing sub-image, assigning a target image correlation corresponding to the first remote sensing sub-image initial set as a maximum image correlation, performing a size comparison operation on an image correlation and a correlation reference value between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set when the remote sensing sub-image initial set comprises at least two frames of satellite remote sensing sub-images, and marking the sub-image initial set as a first remote sensing sub-image initial set when the image correlation between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set is greater than or equal to the correlation reference value, and marking an average value of the image correlation between every two frames of satellite remote sensing sub-images included in the remote sensing sub-image initial set as the target sub-image correlation corresponding to the first remote sensing sub-image initial set;

Step 4, jumping to step 1 when at least one remote sensing sub-image initial set in the plurality of remote sensing sub-image initial sets is not marked as a first remote sensing sub-image initial set, and performing fusion operation on the target image correlation degree corresponding to each first remote sensing sub-image initial set when each remote sensing sub-image initial set in the plurality of remote sensing sub-image initial sets is marked as a first remote sensing sub-image initial set so as to output a second classification coefficient;

step 5, carrying out fusion operation on the first classification coefficient and the second classification coefficient to output a target classification coefficient, and comparing the target classification coefficient with a classification coefficient reference value;

and step 6, jumping to step 1 when the target classification coefficient is smaller than or equal to the classification coefficient reference value, and marking the plurality of remote sensing sub-image initial sets formed in the step 1 as a plurality of corresponding remote sensing sub-image sets when the target classification coefficient is larger than the classification coefficient reference value, and storing each remote sensing sub-image set respectively.

8. A processing system of satellite remote sensing data, which is characterized in that the processing system is applied to a cloud platform, and the processing system of satellite remote sensing data comprises:

The image segmentation module is used for carrying out segmentation operation on a satellite remote sensing image to be processed according to a preset image segmentation size so as to output multi-frame satellite remote sensing sub-images corresponding to the satellite remote sensing image, and the multi-frame satellite remote sensing sub-images are spliced to form the satellite remote sensing image;

the similarity calculation module is used for performing similarity calculation operation on every two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images so as to output image similarity between every two frames of satellite remote sensing sub-images;

the image classification storage module is used for performing classification storage operation on the multi-frame satellite remote sensing sub-images according to the image similarity between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images;

the classifying and storing operation is performed on the multi-frame satellite remote sensing sub-images according to the image similarity between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images, and the classifying and storing operation comprises the following steps:

for each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images, performing region distance calculation operation on geographic regions corresponding to the two frames of satellite remote sensing sub-images to output region distance values between the two frames of satellite remote sensing sub-images, and determining the region correlation values between the two frames of satellite remote sensing sub-images according to the region distance values between the two frames of satellite remote sensing sub-images to output region correlation values between the two frames of satellite remote sensing sub-images;

For each two frames of satellite remote sensing sub-images in the multi-frame satellite remote sensing sub-images, carrying out fusion operation on the region correlation value between the two frames of satellite remote sensing sub-images and the image similarity between the two frames of satellite remote sensing sub-images so as to output the image correlation value between the two frames of satellite remote sensing sub-images;

and carrying out classified storage operation on the multi-frame satellite remote sensing sub-images according to the image correlation degree between every two frames of satellite remote sensing sub-images so as to store the multi-frame satellite remote sensing sub-images.

9. A cloud platform for executing the method for processing satellite remote sensing data according to any one of claims 1 to 7.

Images