CN115965622B - Method and device for detecting change of remote sensing tile data - Google Patents
Method and device for detecting change of remote sensing tile data Download PDFInfo
- Publication number
- CN115965622B CN115965622B CN202310114334.0A CN202310114334A CN115965622B CN 115965622 B CN115965622 B CN 115965622B CN 202310114334 A CN202310114334 A CN 202310114334A CN 115965622 B CN115965622 B CN 115965622B
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- tile data
- target
- area
- sensing tile
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method and a device for detecting the change of remote sensing tile data, which relate to the technical field of change detection and comprise the following steps: acquiring a sample remote sensing tile data pair of a preset change class, and adding a labeling vector for a target area; determining a target parameter corresponding to the target area based on the sample remote sensing tile data pair and the labeling vector; constructing a training set based on target parameters, and training a preset random forest model by using the training set to obtain a target random forest model; after the remote sensing tile data pair to be detected is obtained, the target random forest model and the remote sensing tile data pair to be detected are utilized to determine the change category corresponding to the change area in the area corresponding to the remote sensing tile data pair to be detected, so that the technical problem that the remote sensing tile data cannot be subjected to change detection in the prior art is solved.
Description
Technical Field
The invention relates to the technical field of change detection, in particular to a method and a device for detecting change of remote sensing tile data.
Background
The change detection is an important technical method in the remote sensing field, and provides an important technical means for monitoring the change condition of land features such as cultivated land, buildings, roads, gardens and the like. The traditional change detection method relies on manual visual interpretation, namely, the change of pixels of the phase images before and after comparison is observed simultaneously to obtain the feature change information, but the efficiency of the method is low and is limited by the experience of operators. By means of computer technology, automatic change detection through algorithm programming is becoming the mainstream method. Some students adopt the difference of front and back time phase images to detect the change through difference information, but the noise of the detection result is serious, and the pseudo-change condition is also serious under the influence of the quaternary phase and imaging conditions. With the development of machine learning algorithms, some popular algorithms, such as Support Vector Machines (SVMs), random Forests (RFs), artificial Neural Networks (ANNs), deep Learning (DLs), and the like, are largely put into the change detection, and the algorithms construct a model of the change detection by autonomously learning information of change samples, so that the degree of automation is high, but due to the number of samples, image resolution, region and time phase limitation, the generalization capability of the model is often difficult to meet the application requirements.
On the other hand, most of the existing algorithms are based on multispectral or hyperspectral images for detecting changes, but with the gradual penetration of internet data services, the detection of changes based on remote sensing tile data is particularly important, but the prior art cannot detect changes of remote sensing tile data.
An effective solution to the above-mentioned problems has not been proposed yet.
Disclosure of Invention
Accordingly, the present invention is directed to a method for detecting the change of remote sensing tile data, so as to solve the technical problem that the prior art cannot detect the change of remote sensing tile data.
In a first aspect, an embodiment of the present invention provides a method for detecting a change in remote sensing tile data, including: acquiring a sample remote sensing tile data pair of a preset change type, and adding a labeling vector to a target area, wherein the target area is a change area of a ground object type in an area corresponding to the sample remote sensing tile data, and the labeling vector is used for representing the range of the target area and the preset change type corresponding to the target area; determining a target parameter corresponding to the target area based on the sample remote sensing tile data pair and the labeling vector, wherein the target parameter comprises: geometric features and relative color features; constructing a training set based on the target parameters, and training a preset random forest model by utilizing the training set to obtain a target random forest model; after the remote sensing tile data pair to be detected is obtained, determining a change category corresponding to a change area in the area corresponding to the remote sensing tile data pair to be detected by utilizing the target random forest model and the remote sensing tile data pair to be detected.
Further, the remote sensing tile data pair includes front-time phase remote sensing tile data and rear-time phase remote sensing tile data, and adds a labeling vector for a target area, including: preprocessing the pre-time phase remote sensing tile data and the post-time phase remote sensing tile data respectively to obtain target pre-time phase remote sensing tile data and target post-time phase remote sensing tile data, wherein the preprocessing comprises the following steps: spatial reference system unification, registration and resampling; determining a difference image between the target pre-time phase remote sensing tile data and the target post-time phase remote sensing tile data based on a principal component analysis algorithm; sequentially performing threshold segmentation processing, morphological processing and vectorization processing on the difference image to obtain the annotation vector; and adding the annotation vector to the target area.
Further, determining the target parameter corresponding to the target area based on the sample remote sensing tile data pair and the labeling vector includes: determining a first overlapping region between the annotation vector and the front-time-phase remote sensing tile data, and determining a second overlapping region between the annotation vector and the rear-time-phase remote sensing tile data; dividing the second overlapping region based on a QuickShift dividing algorithm to obtain a dividing mask; and calculating a target parameter corresponding to the target region based on the segmentation mask, the first overlapping region and the second overlapping region.
Further, calculating, based on the segmentation mask, the first overlapping region, and the second overlapping region, a target parameter corresponding to the target region includes: determining a middle region in the segmentation mask, wherein the middle region is a region larger than 1 in the segmentation mask; determining a third overlapping region between the intermediate region and the first overlapping region, and determining a fourth overlapping region between the intermediate region and the second overlapping region; calculating target parameters of the middle area, wherein the target parameters comprise: the ratio of the area of the middle area to the area of the circumscribed rectangle of the middle area, the circumscribed rectangle length-width ratio of the middle area; calculating the average value and standard deviation of a preset wave band between the third overlapping area and the fourth overlapping area; and determining the target parameters, the mean value and the standard deviation as target parameters corresponding to the target area.
Further, determining a change category corresponding to a change region in a region corresponding to the remote sensing tile data pair to be detected by using the target random forest model and the remote sensing tile data pair to be detected, including: adding a labeling vector to the change area of the remote sensing tile data pair to be detected; determining target parameters corresponding to the change areas based on the remote sensing tile data to be detected and the labeling vectors of the change areas of the remote sensing tile data pair to be detected; inputting target parameters corresponding to the change areas into the target random forest model to obtain probability vectors corresponding to the change areas, wherein the probability vectors are used for representing the probability that the change types of the change areas are all preset change types; and if the maximum value in the probability vector is larger than a preset threshold value, determining the preset change type corresponding to the maximum value as the change type of the change area.
In a second aspect, an embodiment of the present invention further provides a device for detecting a change of remote sensing tile data, including: the acquisition unit is used for acquiring a sample remote sensing tile data pair with a preset change category and adding a labeling vector to a target area, wherein the target area is a region with a change in the ground object category in a region corresponding to the sample remote sensing tile data, and the labeling vector is used for representing the range of the target area and the preset change category corresponding to the target area; the determining unit is configured to determine, based on the sample remote sensing tile data pair and the labeling vector, a target parameter corresponding to the target area, where the target parameter includes: geometric features and relative color features; the training unit is used for constructing a training set based on the target parameters, and training a preset random forest model by utilizing the training set to obtain a target random forest model; the detection unit is used for determining the change category corresponding to the change area in the area corresponding to the remote sensing tile data pair to be detected by utilizing the target random forest model and the remote sensing tile data pair to be detected after the remote sensing tile data pair to be detected is acquired.
Further, the remote sensing tile data pair includes a pre-temporal remote sensing tile data and a post-temporal remote sensing tile data, and the acquiring unit is configured to: preprocessing the pre-time phase remote sensing tile data and the post-time phase remote sensing tile data respectively to obtain target pre-time phase remote sensing tile data and target post-time phase remote sensing tile data, wherein the preprocessing comprises the following steps: spatial reference system unification, registration and resampling; determining a difference image between the target pre-time phase remote sensing tile data and the target post-time phase remote sensing tile data based on a principal component analysis algorithm; sequentially performing threshold segmentation processing, morphological processing and vectorization processing on the difference image to obtain the annotation vector; and adding the annotation vector to the target area.
Further, the determining unit is configured to: determining a first overlapping region between the annotation vector and the front-time-phase remote sensing tile data, and determining a second overlapping region between the annotation vector and the rear-time-phase remote sensing tile data; dividing the second overlapping region based on a QuickShift dividing algorithm to obtain a dividing mask; and calculating a target parameter corresponding to the target region based on the segmentation mask, the first overlapping region and the second overlapping region.
In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory and a processor, where the memory is configured to store a program for supporting the processor to execute the method described in the first aspect, and the processor is configured to execute the program stored in the memory.
In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon.
In the embodiment of the invention, a sample remote sensing tile data pair of a preset change type is obtained, and a labeling vector is added to a target area, wherein the target area is a change area of a ground object type in an area corresponding to the sample remote sensing tile data, and the labeling vector is used for representing the range of the target area and the preset change type corresponding to the target area; determining a target parameter corresponding to the target area based on the sample remote sensing tile data pair and the labeling vector, wherein the target parameter comprises: geometric features and relative color features; constructing a training set based on the target parameters, and training a preset random forest model by utilizing the training set to obtain a target random forest model; after the remote sensing tile data pair to be detected is obtained, the target random forest model and the remote sensing tile data pair to be detected are utilized to determine the change category corresponding to the change area in the area corresponding to the remote sensing tile data pair to be detected, so that the purpose of carrying out change detection on the remote sensing tile data is achieved, and the technical problem that the remote sensing tile data cannot be subjected to change detection in the prior art is solved, and the technical effect of providing a new detection mode for change detection is achieved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for detecting changes in remote sensing tile data according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a device for detecting changes in remote sensing tile data according to an embodiment of the present invention;
fig. 3 is a schematic diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Embodiment one:
according to an embodiment of the present invention, there is provided an embodiment of a method of detecting changes in remote sensing tile data, it being noted that the steps illustrated in the flowchart of the figures may be performed in a computer system, such as a set of computer executable instructions, and that although a logical sequence is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in a different order than what is illustrated herein.
Fig. 1 is a flowchart of a method for detecting a change in remote sensing tile data according to an embodiment of the present invention, as shown in fig. 1, the method includes the following steps:
step S102, acquiring a sample remote sensing tile data pair of a preset change type, and adding a labeling vector to a target area, wherein the target area is a region in which the ground object type in the region corresponding to the sample remote sensing tile data changes, and the labeling vector is used for representing the range of the target area and the preset change type corresponding to the target area;
in the embodiment of the invention, the change detection of the remote sensing tile data is mainly performed on three types of features, namely a building, linear features (roads and ditches) and bare soil, and the possibility of the phase change of the three types of features is considered, and the determined detection types comprise the change of bare soil into the building (marked as 1 type change), the change of the building into the bare soil (marked as 2 type change), the change of the bare soil into the linear features (marked as 3 type change) and the change of the linear features into the bare soil (marked as 4 type change), wherein the four types of changes are preset change types.
Step S104, determining a target parameter corresponding to the target area based on the sample remote sensing tile data pair and the labeling vector, wherein the target parameter comprises: geometric features and relative color features;
step S106, a training set is constructed based on the target parameters, and a preset random forest model is trained by using the training set to obtain a target random forest model;
step S108, after the remote sensing tile data pair to be detected is obtained, determining a change category corresponding to a change area in the area corresponding to the remote sensing tile data pair to be detected by using the target random forest model and the remote sensing tile data pair to be detected.
In the embodiment of the invention, a sample remote sensing tile data pair of a preset change type is obtained, and a labeling vector is added to a target area, wherein the target area is a change area of a ground object type in an area corresponding to the sample remote sensing tile data, and the labeling vector is used for representing the range of the target area and the preset change type corresponding to the target area; determining a target parameter corresponding to the target area based on the sample remote sensing tile data pair and the labeling vector, wherein the target parameter comprises: geometric features and relative color features; constructing a training set based on the target parameters, and training a preset random forest model by utilizing the training set to obtain a target random forest model; after the remote sensing tile data pair to be detected is obtained, the target random forest model and the remote sensing tile data pair to be detected are utilized to determine the change category corresponding to the change area in the area corresponding to the remote sensing tile data pair to be detected, so that the purpose of carrying out change detection on the remote sensing tile data is achieved, and the technical problem that the remote sensing tile data cannot be subjected to change detection in the prior art is solved, and the technical effect of providing a new detection mode for change detection is achieved.
In the embodiment of the present invention, step S102 includes the following steps:
preprocessing the pre-time phase remote sensing tile data and the post-time phase remote sensing tile data respectively to obtain target pre-time phase remote sensing tile data and target post-time phase remote sensing tile data, wherein the preprocessing comprises the following steps: spatial reference system unification, registration and resampling;
determining a difference image between the target pre-time phase remote sensing tile data and the target post-time phase remote sensing tile data based on a principal component analysis algorithm;
sequentially performing threshold segmentation processing, morphological processing and vectorization processing on the difference image to obtain the annotation vector;
and adding the annotation vector to the target area.
And (3) performing Principal Component Analysis (PCA) on the front and rear time-phase remote sensing tile data A, B respectively, extracting A, B a first principal component, and performing dimension reduction to obtain first principal component data C1 corresponding to the front time-phase remote sensing tile data and first principal component data C2 corresponding to the rear time-phase remote sensing tile data.
The images C1 and C2 are differenced to obtain a difference image C3.
Then, the average value m and the standard deviation s of the image C3 are calculated, a threshold value y=m+2×s is set, the image C3 is classified into two categories, the position of the image C3 larger than the threshold value y is assigned to be 1, and otherwise, the position of the image C3 larger than the threshold value y is assigned to be 0, so that a binary image C4 is obtained. The value of each pixel in image C4 is either 0 or 1, with a position of 0 indicating no possibility of change and a position of 1 indicating the possibility of change.
And further removing noise and eliminating hollowness by adopting a morphological corrosion and expansion method to obtain an image C5, wherein the morphological corrosion and expansion method is integrated into a morphology module of the Skimage library. And vectorizing the image C5 to obtain a labeling vector D, wherein each image spot of the labeling vector D is the position of an adjacent image block with the value of 1 in the image C5, and the position and range information with the possibility of change are recorded.
In the embodiment of the present invention, step S104 includes the following steps:
determining a first overlapping region between the annotation vector and the front-time-phase remote sensing tile data, and determining a second overlapping region between the annotation vector and the rear-time-phase remote sensing tile data;
dividing the second overlapping region based on a QuickShift dividing algorithm to obtain a dividing mask;
and calculating a target parameter corresponding to the target region based on the segmentation mask, the first overlapping region and the second overlapping region.
In the embodiment of the invention, a front-time-phase remote sensing tile A and a rear-time-phase remote sensing tile B are respectively and spatially matched with a labeling vector D, each image spot in the labeling vector D is traversed, the current image spot is D1, and image blocks E and F (namely, a first overlapping area and a second overlapping area) covered by the D1 in the remote sensing tile A and the remote sensing tile B respectively are obtained.
The acquired image blocks E and F are equal in size, the spatial positions are consistent with the ranges, the image blocks E and F are undetermined areas with possible feature type changes, and the image blocks E and F are possible to have mutual changes among various features.
Therefore, the post-phase image block F is further segmented by using the QuickShift segmentation algorithm to obtain the segmentation mask H. The division mask H is a continuous integer from 0, 0 is a background area, each number from 1 corresponds to a block connected area on the mask, namely one division block, and the type of ground object covered by each division block on the image block F is single, which is a basic unit for giving a change type next.
The QuickShift Segmentation algorithm is a two-dimensional image Segmentation algorithm based on the approximation of a kernel mean shift algorithm, and can calculate layering Segmentation on multiple scales simultaneously, the kernel algorithm is integrated into a segment module of a Skimage library, the Segmentation algorithm is easy to call, the Segmentation effect can keep the continuity of the ground object, and the geometric features of the ground object are not damaged.
In an embodiment of the present invention, based on the segmentation mask, the first overlapping region, and the second overlapping region, a target parameter corresponding to the target region is calculated, including the following steps:
determining a middle region in the segmentation mask, wherein the middle region is a region larger than 1 in the segmentation mask;
determining a third overlapping region between the intermediate region and the first overlapping region, and determining a fourth overlapping region between the intermediate region and the second overlapping region;
calculating target parameters of the middle area, wherein the target parameters comprise: the ratio of the area of the middle area to the area of the circumscribed rectangle of the middle area, the circumscribed rectangle length-width ratio of the middle area;
calculating the average value and standard deviation of a preset wave band between the third overlapping area and the fourth overlapping area;
and determining the target parameters, the mean value and the standard deviation as target parameters corresponding to the target area.
In the embodiment of the invention, the geometric features and the relative color features of the segmented blocks are calculated by taking each segmented block as a basic unit.
And traversing all areas (namely, middle areas) larger than 1 in the segmentation mask H, wherein the current middle area is H1, acquiring segmentation blocks E1 and F1 (namely, a third overlapping area and a fourth overlapping area) covered by H1 in the image blocks E and F, wherein the segmentation blocks E1 and F1 are identical in size, and the positions and the ranges are consistent. The geometric features of the segmentation blocks E1 and F1 are calculated by H1, and the specific steps are as follows:
firstly, calculating an external rectangle of H1;
then, obtaining the length L and the width W of the external rectangle, and calculating the length-width ratio L/W to obtain the length-width ratio characteristic of the external rectangle;
finally, the area1 of H1 and the area2 of the circumscribed rectangle are obtained, and the area ratio area1/area2 is calculated as a second geometric feature.
The relative color feature calculation steps of the segmentation blocks E1 and F1 are as follows:
first, the difference fe=f1-E1 between F1 and E1 is calculated
Then, the mean and standard deviation of FE are calculated, mean FE ={mean FE1 ,mean FE2 ,mean FE3 ' and std FE ={std FE 1,std FE2 ,std FE3 }, mean of FEi And std FEi The mean value and standard deviation of the ith band are respectively.
The obtained aspect ratio and area ratio features have good characterization capability for linear features and block features, and are less influenced by factors such as resolution, time phase, region and the like of images. The mean value and standard deviation characteristics of each wave band obtained based on the difference between the segmentation blocks E1 and F1 weaken the influence of absolute DN values, are beneficial to improving generalization capability, and the statistical characteristics can also improve noise tolerance and have good characterization force on the relative change between ground objects. Finally, for the remote sensing tile data, each pair of segments E1 and F1 is characterized by { mean } FE1 , mean FE2 , mean FE3, std FE1 ,std FE2 , std FE3 , area1/area2, L/W}。
Step S106 will be described below.
And after calculating the target parameters corresponding to the target area based on the sample remote sensing tile data pairs and the labeling vector, taking the target parameters as input, taking the corresponding change types as output, and training by adopting a random forest model to obtain a target random forest model.
And carrying out class probability prediction on the feature vectors of each segmented block E1 and F1 by using a prediction model to obtain a 1*4 probability vector P, wherein each element represents the probability that the segmented block respectively belongs to 4 kinds of variation classes.
In the embodiment of the present invention, step S108 includes the steps of:
adding a labeling vector to the change area of the remote sensing tile data pair to be detected;
determining target parameters corresponding to the change areas based on the remote sensing tile data to be detected and the labeling vectors of the change areas of the remote sensing tile data pair to be detected;
inputting target parameters corresponding to the change areas into the target random forest model to obtain probability vectors corresponding to the change areas, wherein the probability vectors are used for representing the probability that the change types of the change areas are all preset change types;
and if the maximum value in the probability vector is larger than a preset threshold value, determining the preset change type corresponding to the maximum value as the change type of the change area.
In the embodiment of the invention, after the target parameters corresponding to the change area are calculated, the target parameters corresponding to the change area are input into a target random forest model, all the segmentation blocks are traversed, the probability vector of the current segmentation block is P, and the maximum value P of P is calculated max Index P to maximum ind ，P max And P ind Indicating that the current segmentation block belongs to a preset change category P ind The probability of (2) is P max 。
According to actual application requirements, a threshold t is determined, the value of t is equivalent to a confidence level, the larger the value of t is, the fewer the detected change pattern spots are, the higher the confidence level of the detected change pattern spots is, and the lower the value of t is, the opposite is. Typically, a confidence level of 80% is chosen, i.e. a t value of 0.8.
When P max When the current partition block is larger than t, the current partition block is considered to be a preset change type P ind When P is changed to max And when the current segmentation block is smaller than t, the current segmentation block is considered unchanged. By the method for screening the threshold value, the false detection rate can be reduced to a great extent. Finally, when all the dividing blocks of the whole image are subjected to one-time screening and category judgment, a category image is obtained,values 0 to 4,0 as background, 1 to 4 represent 4 variation categories, respectively. And converting the category images into vectors to obtain category vectors, namely the final change detection result.
The embodiment of the invention carries out the change detection of the ground feature based on the front-back time phase remote sensing tile data, fills the blank of the remote sensing tile data change detection method, and provides algorithm support for the engineering application of the remote sensing tile data change detection.
Meanwhile, the adopted geometric features and relative color features weaken the influence of absolute DN values and image resolution, weaken the limitation of regions and time phases to a certain extent, improve the generalization capability of an algorithm and provide a set of feasible schemes for detecting the change of a large-scale and multi-time phase.
Embodiment two:
the embodiment of the invention also provides a change detection device of the remote sensing tile data, which is used for executing the change detection method of the remote sensing tile data provided by the embodiment of the invention, and the following is a specific description of the change detection device of the remote sensing tile data provided by the embodiment of the invention.
As shown in fig. 2, fig. 2 is a schematic diagram of the above apparatus, and the change detection apparatus for remote sensing tile data includes:
the acquiring unit 10 is configured to acquire a sample remote sensing tile data pair of a preset change type, and add a labeling vector to a target area, where the target area is a region where a change occurs in a ground object type in a region corresponding to the sample remote sensing tile data, and the labeling vector is used to characterize a range of the target area and the preset change type corresponding to the target area;
the determining unit 20 is configured to determine, based on the sample remote sensing tile data pair and the labeling vector, a target parameter corresponding to the target area, where the target parameter includes: geometric features and relative color features;
the training unit 30 is configured to construct a training set based on the target parameter, and train a preset random forest model by using the training set to obtain a target random forest model;
the detecting unit 40 is configured to determine, after acquiring the to-be-detected remote sensing tile data pair, a change category corresponding to a change region in the region corresponding to the to-be-detected remote sensing tile data pair by using the target random forest model and the to-be-detected remote sensing tile data pair.
In the embodiment of the invention, a sample remote sensing tile data pair of a preset change type is obtained, and a labeling vector is added to a target area, wherein the target area is a change area of a ground object type in an area corresponding to the sample remote sensing tile data, and the labeling vector is used for representing the range of the target area and the preset change type corresponding to the target area; determining a target parameter corresponding to the target area based on the sample remote sensing tile data pair and the labeling vector, wherein the target parameter comprises: geometric features and relative color features; constructing a training set based on the target parameters, and training a preset random forest model by utilizing the training set to obtain a target random forest model; after the remote sensing tile data pair to be detected is obtained, the target random forest model and the remote sensing tile data pair to be detected are utilized to determine the change category corresponding to the change area in the area corresponding to the remote sensing tile data pair to be detected, so that the purpose of carrying out change detection on the remote sensing tile data is achieved, and the technical problem that the remote sensing tile data cannot be subjected to change detection in the prior art is solved, and the technical effect of providing a new detection mode for change detection is achieved.
Embodiment III:
an embodiment of the present invention further provides an electronic device, including a memory and a processor, where the memory is configured to store a program that supports the processor to execute the method described in the first embodiment, and the processor is configured to execute the program stored in the memory.
Referring to fig. 3, an embodiment of the present invention further provides an electronic device 100, including: a processor 50, a memory 51, a bus 52 and a communication interface 53, the processor 50, the communication interface 53 and the memory 51 being connected by the bus 52; the processor 50 is arranged to execute executable modules, such as computer programs, stored in the memory 51.
The memory 51 may include a high-speed random access memory (RAM, random Access Memory), and may further include a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. The communication connection between the system network element and at least one other network element is achieved via at least one communication interface 53 (which may be wired or wireless), and the internet, wide area network, local network, metropolitan area network, etc. may be used.
Bus 52 may be an ISA bus, a PCI bus, an EISA bus, or the like. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in FIG. 3, but not only one bus or type of bus.
The memory 51 is configured to store a program, and the processor 50 executes the program after receiving an execution instruction, and the method executed by the apparatus for flow defining disclosed in any of the foregoing embodiments of the present invention may be applied to the processor 50 or implemented by the processor 50.
The processor 50 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuitry in hardware in the processor 50 or by instructions in the form of software. The processor 50 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but may also be a digital signal processor (Digital Signal Processing, DSP for short), application specific integrated circuit (Application Specific Integrated Circuit, ASIC for short), off-the-shelf programmable gate array (Field-ProgrammableGate Array, FPGA for short), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory 51 and the processor 50 reads the information in the memory 51 and in combination with its hardware performs the steps of the above method.
Embodiment four:
the embodiment of the invention also provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of the method in the first embodiment are executed.
In addition, in the description of embodiments of the present invention, unless explicitly stated and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
Finally, it should be noted that: the above examples are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention, but it should be understood by those skilled in the art that the present invention is not limited thereto, and that the present invention is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (5)
1. A method for detecting changes in remote sensing tile data, comprising:
acquiring a sample remote sensing tile data pair of a preset change type, and adding a labeling vector to a target area, wherein the target area is a change area of a ground object type in an area corresponding to the sample remote sensing tile data, and the labeling vector is used for representing the range of the target area and the preset change type corresponding to the target area;
determining a target parameter corresponding to the target area based on the sample remote sensing tile data pair and the labeling vector, wherein the target parameter comprises: geometric features and relative color features;
constructing a training set based on the target parameters, and training a preset random forest model by utilizing the training set to obtain a target random forest model;
after the remote sensing tile data pair to be detected is obtained, determining a change category corresponding to a change area in an area corresponding to the remote sensing tile data pair to be detected by utilizing the target random forest model and the remote sensing tile data pair to be detected;
the remote sensing tile data pair comprises front-time-phase remote sensing tile data and rear-time-phase remote sensing tile data, and a labeling vector is added for a target area, and the remote sensing tile data pair comprises:
preprocessing the pre-time phase remote sensing tile data and the post-time phase remote sensing tile data respectively to obtain target pre-time phase remote sensing tile data and target post-time phase remote sensing tile data, wherein the preprocessing comprises the following steps: spatial reference system unification, registration and resampling;
determining a difference image between the target pre-time phase remote sensing tile data and the target post-time phase remote sensing tile data based on a principal component analysis algorithm;
sequentially performing threshold segmentation processing, morphological processing and vectorization processing on the difference image to obtain the annotation vector;
the determining, based on the sample remote sensing tile data pair and the labeling vector, a target parameter corresponding to the target area includes:
determining a first overlapping region between the annotation vector and the front-time-phase remote sensing tile data, and determining a second overlapping region between the annotation vector and the rear-time-phase remote sensing tile data;
dividing the second overlapping region based on a QuickShift dividing algorithm to obtain a dividing mask;
calculating a target parameter corresponding to the target region based on the segmentation mask, the first overlapping region and the second overlapping region;
wherein calculating, based on the segmentation mask, the first overlapping region, and the second overlapping region, a target parameter corresponding to the target region includes:
determining a middle region in the segmentation mask, wherein the middle region is a region larger than 1 in the segmentation mask;
determining a third overlapping region between the intermediate region and the first overlapping region, and determining a fourth overlapping region between the intermediate region and the second overlapping region;
calculating target parameters of the middle area, wherein the target parameters comprise: the ratio of the area of the middle area to the area of the circumscribed rectangle of the middle area, the circumscribed rectangle length-width ratio of the middle area;
calculating the average value and standard deviation of a preset wave band between the third overlapping area and the fourth overlapping area;
and determining the target parameters, the mean value and the standard deviation as target parameters corresponding to the target area.
2. The method of claim 1, wherein determining a change category corresponding to a change region in a region corresponding to the pair of remote sensing tile data to be detected using the target random forest model and the pair of remote sensing tile data to be detected, comprises:
adding a labeling vector to the change area of the remote sensing tile data pair to be detected;
determining target parameters corresponding to the change areas based on the remote sensing tile data to be detected and the labeling vectors of the change areas of the remote sensing tile data pair to be detected;
inputting target parameters corresponding to the change areas into the target random forest model to obtain probability vectors corresponding to the change areas, wherein the probability vectors are used for representing the probability that the change types of the change areas are all preset change types;
and if the maximum value in the probability vector is larger than a preset threshold value, determining the preset change type corresponding to the maximum value as the change type of the change area.
3. A change detection device for remote sensing tile data, comprising:
the acquisition unit is used for acquiring a sample remote sensing tile data pair with a preset change category and adding a labeling vector to a target area, wherein the target area is a region with a change in the ground object category in a region corresponding to the sample remote sensing tile data, and the labeling vector is used for representing the range of the target area and the preset change category corresponding to the target area;
the determining unit is configured to determine, based on the sample remote sensing tile data pair and the labeling vector, a target parameter corresponding to the target area, where the target parameter includes: geometric features and relative color features;
the training unit is used for constructing a training set based on the target parameters, and training a preset random forest model by utilizing the training set to obtain a target random forest model;
the detection unit is used for determining a change category corresponding to a change area in an area corresponding to the remote sensing tile data pair to be detected by utilizing the target random forest model and the remote sensing tile data pair to be detected after the remote sensing tile data pair to be detected is acquired;
the remote sensing tile data pair comprises front-time-phase remote sensing tile data and rear-time-phase remote sensing tile data, and the acquisition unit is used for:
preprocessing the pre-time phase remote sensing tile data and the post-time phase remote sensing tile data respectively to obtain target pre-time phase remote sensing tile data and target post-time phase remote sensing tile data, wherein the preprocessing comprises the following steps: spatial reference system unification, registration and resampling;
determining a difference image between the target pre-time phase remote sensing tile data and the target post-time phase remote sensing tile data based on a principal component analysis algorithm;
sequentially performing threshold segmentation processing, morphological processing and vectorization processing on the difference image to obtain the annotation vector;
adding the annotation vector to the target area;
the determining unit is used for:
determining a first overlapping region between the annotation vector and the front-time-phase remote sensing tile data, and determining a second overlapping region between the annotation vector and the rear-time-phase remote sensing tile data;
dividing the second overlapping region based on a QuickShift dividing algorithm to obtain a dividing mask;
calculating a target parameter corresponding to the target region based on the segmentation mask, the first overlapping region and the second overlapping region;
wherein calculating, based on the segmentation mask, the first overlapping region, and the second overlapping region, a target parameter corresponding to the target region includes:
determining a middle region in the segmentation mask, wherein the middle region is a region larger than 1 in the segmentation mask;
determining a third overlapping region between the intermediate region and the first overlapping region, and determining a fourth overlapping region between the intermediate region and the second overlapping region;
calculating target parameters of the middle area, wherein the target parameters comprise: the ratio of the area of the middle area to the area of the circumscribed rectangle of the middle area, the circumscribed rectangle length-width ratio of the middle area;
calculating the average value and standard deviation of a preset wave band between the third overlapping area and the fourth overlapping area;
and determining the target parameters, the mean value and the standard deviation as target parameters corresponding to the target area.
4. An electronic device comprising a memory for storing a program supporting the processor to perform the method of any one of claims 1 to 2, and a processor configured to execute the program stored in the memory.
5. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, performs the steps of the method according to any of the preceding claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310114334.0A CN115965622B (en) | 2023-02-15 | 2023-02-15 | Method and device for detecting change of remote sensing tile data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310114334.0A CN115965622B (en) | 2023-02-15 | 2023-02-15 | Method and device for detecting change of remote sensing tile data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115965622A CN115965622A (en) | 2023-04-14 |
| CN115965622B true CN115965622B (en) | 2023-06-09 |
Family
ID=85896055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310114334.0A Active CN115965622B (en) | 2023-02-15 | 2023-02-15 | Method and device for detecting change of remote sensing tile data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115965622B (en) |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10325370B1 (en) * | 2016-05-31 | 2019-06-18 | University Of New Brunswick | Method and system of coregistration of remote sensing images |
| CN108761451B (en) * | 2018-05-21 | 2020-09-04 | 中南大学 | Land use change intelligent understanding method based on multi-temporal remote sensing image |
| CN111597861A (en) * | 2019-02-21 | 2020-08-28 | 中科星图股份有限公司 | System and method for automatically interpreting ground object of remote sensing image |
| CN111242050A (en) * | 2020-01-15 | 2020-06-05 | 同济大学 | Automatic change detection method for remote sensing image in large-scale complex scene |
| CN112232171A (en) * | 2020-10-10 | 2021-01-15 | 中煤航测遥感集团有限公司 | Remote sensing image information extraction method and device based on random forest and storage medium |
| CN113569760A (en) * | 2021-07-29 | 2021-10-29 | 重庆交通大学 | Three-dimensional change detection method based on multi-mode deep learning |
| CN114972191A (en) * | 2022-04-25 | 2022-08-30 | 航天宏图信息技术股份有限公司 | Method and device for detecting farmland change |
-
2023
- 2023-02-15 CN CN202310114334.0A patent/CN115965622B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115965622A (en) | 2023-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110245659B (en) | Image salient object segmentation method and device based on foreground and background interrelation | |
| CN110309824B (en) | Character detection method and device and terminal | |
| CN114972191A (en) | Method and device for detecting farmland change | |
| Valentini et al. | New algorithms for shoreline monitoring from coastal video systems | |
| CN108764325A (en) | Image-recognizing method, device, computer equipment and storage medium | |
| CN103544703B (en) | Digital picture splicing detection method | |
| Wang et al. | The poor generalization of deep convolutional networks to aerial imagery from new geographic locations: an empirical study with solar array detection | |
| CN114926511A (en) | High-resolution remote sensing image change detection method based on self-supervision learning | |
| CN111008561A (en) | Livestock quantity determination method, terminal and computer storage medium | |
| CN116453104B (en) | Liquid level identification method, liquid level identification device, electronic equipment and computer readable storage medium | |
| CN115187530A (en) | Method, device, terminal and medium for identifying ultrasonic automatic breast full-volume image | |
| CN112330597A (en) | Image difference detection method and device and computer equipment | |
| CN115953612A (en) | ConvNeXt-based remote sensing image vegetation classification method and device | |
| Yu et al. | Mean shift based clustering of neutrosophic domain for unsupervised constructions detection | |
| CN115965622B (en) | Method and device for detecting change of remote sensing tile data | |
| CN113505643A (en) | Violation target detection method and related device | |
| CN114494999B (en) | Double-branch combined target intensive prediction method and system | |
| CN114742849B (en) | Leveling instrument distance measuring method based on image enhancement | |
| CN113298755B (en) | Method and device for rapidly detecting ecological environment change patch based on time sequence image | |
| CN115861823A (en) | Remote sensing change detection method and device based on self-supervision deep learning | |
| CN116363583A (en) | Human body identification method, device, equipment and medium for top view angle | |
| EP3989162A1 (en) | Defect image generation method for deep learning and system therefore | |
| CN114565858B (en) | Multispectral image change detection method based on geospatial perception low-rank reconstruction network | |
| CN112749670B (en) | Pixel-by-pixel classification method, medium and equipment for multi-source remote sensing image | |
| CN117274817B (en) | Automatic crack identification method and device, terminal equipment and storage 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |