CN113109816B - Echo block tracking method, device and storage medium of radar echo image - Google Patents
Echo block tracking method, device and storage medium of radar echo image Download PDFInfo
- Publication number
- CN113109816B CN113109816B CN202110332840.8A CN202110332840A CN113109816B CN 113109816 B CN113109816 B CN 113109816B CN 202110332840 A CN202110332840 A CN 202110332840A CN 113109816 B CN113109816 B CN 113109816B
- Authority
- CN
- China
- Prior art keywords
- echo
- block
- radar
- blocks
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/95—Radar or analogous systems specially adapted for specific applications for meteorological use
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/66—Radar-tracking systems; Analogous systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/411—Identification of targets based on measurements of radar reflectivity
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The application discloses an echo block tracking method, an echo block tracking device and a storage medium of a radar echo image, which relate to the technical field of weather forecast, wherein the echo block tracking method of the radar echo image comprises the following steps: acquiring a plurality of radar echo images to form a radar echo image sequence, wherein the radar echo images in the radar echo image sequence are from the same radar and are arranged in time sequence; performing hierarchical division on each radar echo image in the radar echo image sequence to obtain a plurality of binarized hierarchical matrixes; determining all echo blocks in the hierarchical matrix according to the communication rule; calculating the similarity between echo blocks in the level matrix of the same level of two adjacent radar echo images; and judging two echo blocks with the similarity larger than the threshold value as the same cloud cluster. By hierarchically comparing the echo block similarities, echo blocks in radar echo images can be accurately tracked.
Description
Technical Field
The present application relates to the field of weather forecast technologies, and in particular, to a method and an apparatus for tracking an echo block of a radar echo image, and a storage medium.
Background
Radar echo images are powerful tools for each weather station to make short-time predictions of strong convection weather, heavy rain and general precipitation. The radar echo image has the characteristics of visual image and quick response to the live ageing of precipitation, creates favorable conditions for weather stations to develop short-time weather forecast service, and achieves good effects in practical work. When the radar electromagnetic wave encounters particles such as cloud drops, raindrops and ice crystals in the propagation process, part of the electromagnetic wave is reflected back to be received by the radar, the energy received by the radar is called echo power, and particularly for a certain radar, two water-falling areas with the same distance as the radar have large echo power and strong echo; and vice versa. Radar echo images are widely applied in the field of weather forecast, and are often combined with a computer to extrapolate the echo intensity of radar in a future period of time, so as to predict weather changes in the future period of time. However, when the computer performs extrapolation according to the radar echo intensity, the state time change of the echo block, such as echo intensity change, echo block moving speed, moving direction of the radar echo block in different areas, etc., may have different effects on the extrapolation result of the computer, and even affect the final output result. Therefore, the computer needs to acquire and analyze echo blocks in the radar echo image before extrapolating the radar echo image sequence.
Disclosure of Invention
In order to solve at least one of the above technical problems, the present application provides a method, an apparatus and a storage medium for tracking an echo block of a radar echo image, which can accurately track an echo block in the radar echo image.
In a first aspect, an embodiment of the present application provides an echo block tracking method for a radar echo image, including the following steps:
acquiring a plurality of radar echo images to form a radar echo image sequence, wherein the radar echo images in the radar echo image sequence are from the same radar and are arranged in time sequence;
performing hierarchical division on each radar echo image in the radar echo image sequence to obtain a plurality of binarized hierarchical matrixes;
determining all echo blocks in the hierarchical matrix according to the communication rule;
calculating the similarity between echo blocks in the level matrix of the same level of two adjacent radar echo images;
and judging two echo blocks with the similarity larger than the threshold value as the same cloud cluster.
In some embodiments, the step of hierarchically dividing each radar echo image in the radar echo image sequence to obtain a plurality of binarized hierarchical matrices includes the steps of:
converting the color values of pixels in the radar echo image into radar echo intensity values to obtain an intensity matrix;
and carrying out hierarchical division on the intensity matrix by combining the radar echo intensity threshold value group to obtain a plurality of binarization hierarchical matrices.
In some embodiments, the layering the intensity matrix in combination with the radar echo intensity threshold set to obtain a plurality of binarized layering matrices includes the steps of:
acquiring a radar intensity echo threshold value of a j-th level according to the radar echo intensity threshold value group, wherein the radar echo intensity threshold value group is an ordered sequence, and the expression of the radar echo intensity threshold value group is as follows:
C＝{C j ＝Z+j(L-Z)(z-1) -1 |j∈[0，z)∩N}，C j l table representing radar intensity echo threshold value of j th levelShowing a set radar echo intensity maximum value, wherein Z represents a set radar echo intensity minimum value, and Z represents a divided total hierarchy;
the ith intensity matrix N according to the radar echo intensity threshold value of the jth level i Performing numerical comparison and assignment to obtain a j-th hierarchical matrix, wherein the element position of the intensity matrix, which is larger than the radar intensity echo threshold value of the j-th hierarchical matrix, is 1, and the element position of the intensity matrix, which is not larger than the radar intensity echo threshold value of the j-th hierarchical matrix, is 0, so as to obtain a j-th binary hierarchical matrix S i，j 。
In some embodiments, the calculating the similarity between echo blocks in the same hierarchical matrix of two adjacent radar echo images includes the following steps:
obtaining a first echo block set according to echo blocks in a j-th hierarchical matrix of the ith radar echo image;
obtaining a second echo block set according to echo blocks in a j-th hierarchical matrix of the i-1 th radar echo image;
calculating the area similarity, the center similarity and the shape similarity of an e-th echo block in the first echo block set and an m-th echo block in the second echo block set;
and determining the similarity of the two echo blocks according to the area similarity, the center similarity and the shape similarity.
In some embodiments, the determining all echo blocks in the hierarchical matrix according to the connectivity rule includes the steps of:
determining a connected set in the hierarchical matrix according to constraint rules of four or eight connections;
and determining all echo blocks in the hierarchical matrix according to the characteristics of the connected set.
In some embodiments, the calculating the similarity between echo blocks in the same hierarchical matrix of two adjacent radar echo images includes the following steps:
determining the relation of echo blocks in adjacent level matrixes according to the characteristics of the echo blocks in the level matrixes of the adjacent level of the same radar echo image, wherein when a first echo block coordinate set in a low level matrix comprises a plurality of second echo block coordinate sets in a high level matrix, the first echo block is a father echo block of the second echo block, and the second echo block is a child echo block of the first echo block;
and calculating the similarity between a third echo block in the j-th level hierarchical matrix of the ith radar echo image and a fourth echo block in the j-th level hierarchical matrix of the i-1 th radar echo image, wherein the similarity between a father echo block of the third echo block and a father echo block of the fourth echo block is larger than a first threshold value.
In some embodiments, the determining that two echo blocks having a similarity greater than a threshold are the same cloud includes the steps of:
judging whether the third echo block has a sub echo block according to the echo block relation;
judging whether the fourth echo block has a sub echo block according to the echo block relation;
and when the third echo block does not have the sub echo block or the fourth echo block does not have the sub echo block, and the similarity of the third echo block and the fourth echo block is larger than the threshold value, the third echo block and the fourth echo block are in the same cloud cluster.
In some embodiments, the method for echo block tracking of radar echo images further comprises the steps of:
when two echo blocks are judged to be the same cloud cluster, determining a previous echo block and a subsequent echo block according to the time sequence of two adjacent radar echo images in the radar echo image sequence;
the motion trajectories of the cloud are displayed by plotting the previous echo block and the subsequent echo block.
In a second aspect, an embodiment of the present application further provides an echo block tracking device for radar echo images, including: the system comprises a memory for storing at least one program and a processor for loading the at least one program to perform the echo block tracking method of the radar echo image of the embodiment of the first aspect.
In a third aspect, embodiments of the present application also provide a computer storage medium in which a processor-executable program is stored, which when executed by the processor is configured to implement the echo block tracking method of the radar echo image of the first aspect described above.
The technical scheme of the application has at least one of the following advantages or beneficial effects: acquiring a plurality of radar echo images which are from the same radar and are arranged in time sequence to form a radar echo image sequence, carrying out hierarchical division on each radar echo image in the radar echo image sequence to obtain a plurality of binarized hierarchical matrixes, determining all echo blocks in the hierarchical matrixes according to a communication rule, and calculating the similarity between echo blocks in the hierarchical matrixes of the same hierarchy of two radar echo images adjacent in time sequence; and judging two echo blocks with the similarity larger than the threshold value as the same cloud cluster. The obtained radar echo images are arranged in time sequence, the echo block similarity between the adjacent radar echo images is calculated to track cloud clusters at different times, meanwhile, the radar echo images are divided into a plurality of layer matrixes, and then the echo block similarities of the adjacent radar echo images are compared layer by layer, so that the echo block tracking result is more accurate, and a basis is provided for accurate extrapolation by a computer.
Drawings
FIG. 1 is a flow chart of an echo block tracking method for radar echo images provided in accordance with an embodiment of the present application;
FIG. 2 is a schematic view of a radar echo image provided in accordance with an embodiment of the present application;
FIG. 3 is a schematic diagram of a set of hierarchical matrices after a radar echo image is hierarchically partitioned according to an embodiment of the present application;
FIG. 4 is a schematic view of a set of radar echo images and their hierarchical matrix of the same hierarchy provided in accordance with an embodiment of the present application;
FIG. 5 is a diagram of a j-th level cloud cluster tracking result record representing intent provided in accordance with an embodiment of the present application;
fig. 6 is a diagram for tracking cloud motion trajectories according to an embodiment of the present application.
Detailed Description
The embodiments described herein should not be construed as limiting the application, and all other embodiments, which may be made by those of ordinary skill in the art without the benefit of the present disclosure, are intended to be within the scope of the present application.
In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with one another without conflict.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the application only and is not intended to be limiting of the application.
An embodiment of the present application provides an echo block tracking method of radar echo images, referring to fig. 1, the method of the embodiment of the present application includes, but is not limited to, step S110, step S120, step S130, step S140 and step S150.
Step S110, a plurality of radar echo images are acquired to form a radar echo image sequence, wherein the radar echo images in the radar echo image sequence are from the same radar and are arranged in time sequence.
Step S120, each radar echo image in the radar echo image sequence is subjected to hierarchical division to obtain a plurality of binarized hierarchical matrixes.
Step S130, determining all echo blocks in the hierarchical matrix according to the connection rule.
Step S140, calculating the similarity between echo blocks in the level matrix of the same level of two adjacent radar echo images.
In step S150, two echo blocks with similarity greater than the threshold are determined to be the same cloud.
In some embodiments, the input is a group of radar echo image sequences composed of n radar echo images which belong to a certain radar, are ordered in time and have consistent time intervals, and the ith radar echo image is subjected to hierarchical division to obtain the ith radar echo imageA plurality of binarized hierarchical matrices, wherein the hierarchical matrix of the j-th hierarchy is denoted as S i，j . Obtaining a hierarchical matrix S according to the connection rule and constraint condition of the coordinates i，j All connected sets omega in (1) i，j ，Ω i，j The kth connected set in (a) is omega i，j，k Each connected set corresponds to an echo block in each hierarchical matrix. According to omega i，j，k Corresponding coordinate set calculation Ω i，j，k And is referred to as echo block P i，j，k Is used for calculating a hierarchical matrix S according to the characteristics of the echo blocks i-1，j And a hierarchical matrix S i，j And when the calculated similarity is greater than a threshold value, judging that the two echo blocks are reflected by the same cloud cluster in adjacent time. Illustratively, one radar echo image is acquired every five minutes, and six are acquired in total. Dividing each radar echo image into 8 level matrixes, extracting a connected set in each level matrix to determine echo block characteristics in the level matrixes, calculating the similarity between echo blocks in the level matrixes of two adjacent radar echo images layer by layer according to the echo block characteristics, judging two echo blocks with the similarity larger than a threshold value as the same cloud cluster, plotting and displaying the same cloud cluster, finally obtaining a track tracing diagram of 8 cloud clusters with different levels between 0 and 30 minutes,
according to some embodiments of the present application, step S120 specifically includes step S210 and step S220:
step S210, converting the color values of the pixels in the radar echo image into radar echo intensity values to obtain an intensity matrix.
Step S220, carrying out hierarchical division on the intensity matrix by combining the radar echo intensity threshold value group to obtain a plurality of binarization hierarchical matrices.
In some embodiments, each radar echo image has a width of w pixels and a height of h pixels, the color values of the x-th column and the y-th row of the ith radar echo image are converted into radar echo intensity values and stored in the x-th column and the y-th row of an intensity matrix, and the intensity matrix of the ith radar echo image is recorded as N i Wherein, the mapping of the color value and the radar echo intensity is a bijective relation, i is [0, n ])∩N，x∈[0，w)∩N，y∈[0，h)∩N，N i ∈R w×h 。
The radar echo intensity threshold value group C is an ordered sequence, and the maximum value L, the minimum value Z and the total level Z of the radar echo intensity are given, and the radar echo intensity threshold value group C is expressed as:
C＝{C j ＝Z+j(L-Z)(z-1) -1 |j∈[0，z)∩N}
wherein the range of the radar echo intensity threshold value of the j-th level is represented as [ C ] j L). S is a hierarchical matrix set of all radar echo images, S i For all the level matrixes corresponding to the ith image, S i，j Is the i-th image, the j-th hierarchical matrix, S i，j ∈R w×h . N is recorded i，(x，y) Is N i Values of the x-th column and y-th row of S i，j，(x，y) Is S i，j The values of the x column and the y row are N i，(x，y) The value of (2) is within the range of the radar echo intensity threshold value of the j-th level, then the corresponding S i，j，(x，y) And the value of (2) is 1, otherwise 0. The expression of each element in the hierarchical matrix is:
specifically, referring to fig. 2 and 3, fig. 2 is an exemplary diagram of a radar echo image in a radar echo image sequence, in which the lighter the color is, the higher the radar echo intensity value is, the radar echo image shown in fig. 2 is converted into an intensity matrix, and a set of level matrices shown in fig. 3 can be obtained after the level division is performed by combining with a radar echo intensity threshold set.
According to some embodiments of the application, step S130 includes, but is not limited to, step S310 and step S320:
step S310, determining a connected set in the hierarchical matrix according to constraint rules of four or eight connections;
step S320, all echo blocks in the hierarchical matrix are determined according to the features of the connected set.
In some embodiments, a function dst (θ, λ) is defined to calculate the coordinates θ and λDistance between, where θ= (θ) x ，θ y )，λ＝(λ x ，λ y ). If a four-way rule is employed, dst=dst 4 The method comprises the steps of carrying out a first treatment on the surface of the If an eight-way rule is employed, dst=dst 8 The two different calculation modes of the communication rule coordinate distance are respectively as follows:
dst 4 (θ，λ)＝|θ x -λ x |+|θ y -λ y |
dst 8 (θ，λ)＝max(|θ x -λ x |，|θ y -λ y |)
omega-recording device i，j For being composed of a hierarchical matrix S i，j All the connected set sets obtained by calculation in the step (a), omega i，j，k Aggregation of omega for connected sets i，j The kth connected set, Ω i，j，k，l For communicating the set omega i，j，k The first coordinates of (a) are:
Ω i，j，k，l ＝(Ω i，j，k，l，x ，Ω i，j，k，l，y )
connected set omega i，j The following constraints exist for all connected sets in (a):
wherein a epsilon N, b epsilon N, c epsilon N, d epsilon N, f epsilon N, k epsilon N, l epsilon N.
At the level matrix S i，j If all the values in the matrix are not 0, one or more connected sets are needed to obtain a hierarchical matrix S according to the constraint conditions i，j All connected set omega in (1) i，j 。
Further, a connected set omega is calculated i，j Is set omega for each connection i，j，k Thereby determining a hierarchical matrix S based on the features of the connected set i，j The features of the connected set may include a shape matrix, a number of coordinates, a boundary value, a centroid, a divergence, and a divergence of the lateral and longitudinal coordinates, and each feature is referred to as a feature set of the echo block.
Record Q i，j，k For communicating the set omega i，j，k Shape matrix, Q i，j，k ∈R w×h ，Q i，j，k，(x，y) Is a shape matrix Q i，j，k Value of the x column, y row, shape matrix Q i，j，k The expression is:
connected omega i，j，k The area of (2) is denoted as G i，j，k According to the connected set Ω i，j，k Shape matrix Q i，j，k All non-0 elements in the list can be connected to a set omega i，j，k Area G of (2) i，j，k The method comprises the following steps:
record l i，j，k 、r i，j，k 、r i，j，k 、b i，j，k Respectively are connected together as a set omega i，j，k Left, right, upper, lower boundary, in combination with a connected set Ω i，j，k Left boundary l is obtained from all coordinates in the table i，j，k Right boundary r i，j，k Upper boundary r i，j，k Lower boundary b i，j，k The expressions of (2) are respectively:
l i，j，k ＝min(Ω i，j，k，l，x )
r i，j，k ＝max(Ω i，j，k，l，x )
t i，j，k ＝max(Ω i，j，k，l，y )
b i，j，k ＝min(Ω i，j k ，l，y )
coordinate (mu) is recorded i，j，k，x ，μ i，j，k，y ) For communicating the set omega i，j，k The centroid of (2) is calculated as:
μ i，j，k，x ＝G i，j，k -1 ∑ l＝0 Ω i，j，k，l，x
μ i，j，k，y ＝G i，j，k -1 ∑ l＝0 Ω i，j，k，l，y
record (sigma) i，j，k ，τ i，j，k ) For communicating the set omega i，j，k The calculated formula of the divergence in the transverse and longitudinal coordinates is as follows:
σ i，j，k ＝G i，j，k -1/2 (∑ l＝0 (Ω i，j，k，l，x -μ i，j，k，x ) 2 ) 1/2
τ i，j，k ＝G i，j，k -1/2 (∑ l＝0 (Ω i，j，k，l，y -μ i，j，k，y ) 2 ) 1/2
will connect to the collection omega i，j，k As echo block P i，j，k Is used for obtaining the echo block P i，j，k The feature set of (2) is:
P i，j，k ＝(Ω i，j，k ，Q i，j，k ，G i，j，k ，l i，j，k ，r i，j，k ，t i，j，k ，b i，j，k ，(μ i，j，k，x ，μ i，j，k，y )，(σ i，j，k ，τ i，j，k ))
and carrying out the steps on each image, all layers and all connected sets to obtain the characteristic sets of each image, all layers and all echo blocks.
According to some embodiments of the present application, step S140 includes, but is not limited to, step S410, step S420, step S430, and step S440:
step S410, a first echo block set is obtained according to echo blocks in a j-th level hierarchical matrix of the ith radar echo image.
Step S420, a second echo block set is obtained according to echo blocks in the j-th level of the level matrix of the i-1 th radar echo image.
Step S430, calculating the area similarity, the center similarity and the shape similarity of the e-th echo block in the first echo block set and the m-th echo block in the second echo block set;
step S440, the similarity of the two echo blocks is determined according to the area similarity, the center similarity and the shape similarity.
In some embodiments, note P i，j The echo block set for the ith image and the jth level is:
taking a first echo block set P of the same level in two radar echo images adjacent in time i，j And a second echo block set P i-1，j In the first echo block set P i，j One echo block is arbitrarily taken out and marked as P i，j，e In the second echo block set P i-1，j One echo block is arbitrarily taken out and marked as P i-1，j，m Wherein e.epsilon.N and m.epsilon.N.
Echo block P i，j，e And echo block P i-1，j，m The area similarity of (2) is the area ratio of two echo blocks, and the function F (i, j, e, m) is recorded as the calculation P i，j，e And P i-1，j，m The area similarity of (2) is calculated by the following formula:
F(i，j，e，m)＝min(G i-1，j，m /G i，j，e ，G i，j，e /G i-1，j，m )
echo block P i，j，e And echo block P i-1，j，m The center similarity of the echo block P is calculated by Euclidean distance and total discrete degree of the centroid of the echo block P i，j，e Centroid of (mu) i，j，e，x ，μ i，j，e，y ) The divergence of the transverse and longitudinal coordinates is (sigma i，j，e ，τ i，j，e ) Echo blockP i-1，j，m Centroid of (mu) i-1，j，m，x ，μ i-1，j，m，y ) The divergence of the transverse and longitudinal coordinates is (sigma i-1，j，m ，τ i-1，j，m ). Record W i，j，e，m For echo block P i，j，e And echo block P i-1，j，m Euclidean distance of centroid, V i，j，e，m For echo block P i，j，e And echo block P i-1，j，m The total discrete degree is as follows:
W i，j，e，m ＝((μ i，j，e，x -μ i-1，j，m，x ) 2 +(μ i，j，e，y -μ i-1，j，m，y ) 2 ) 1/2
V i，j，e，m ＝0.5((σ i，j，e +σ i-1，j，m ) 2 +(τ i，j，e +τ i-1，j，m ) 2 ) 1/2
recording the function U (i, j, e, m) as the calculated echo block P i，j，e And echo block P i-1，j，m The center similarity calculation formula is:
U(i，j，e，m)＝V i，j，e，m /(W i，j，e，m +V i，j，e，m )
echo block P i，j，e And echo block P i-1，j，m The shape similarity of the two echo blocks can be calculated according to a shape matrix, wherein the shape matrix of the two echo blocks is subjected to translational overlapping according to the centroid coordinates based on the centroid coordinates of the two echo blocks, and then the shape similarity of the two echo blocks is calculated. Echo block P i，j，e Shape matrix of Q i，j，e Echo block P i-1，j，m Shape matrix of Q i-1，j，m . Record d x Heel d y For the relative offset value between the mass centers of two echo blocks on the same coordinate system, the calculation formula is as follows:
d x ＝μ i，j，e，x -μ i-1，j，m，x
d y ＝μ i，j，e，y -μ i-1，j，m，y
defining a function cal (beta) to avoid operation subscript crossing, wherein beta is a real number, and the expression of the function cal (beta) is:
record A i，j，e，m For echo block P i，j，e And echo block P i-1，j，m Overlapping area after shape matrix translation, B i，j，e，m For echo block P i，j，e And echo block P i-1，j，m Sum of area of overlapping portion and non-overlapping portion after shape matrix translation of (a) i，j，e，m And B is connected with i，j，e，m The calculation modes of (a) are respectively as follows:
A i，j，e，m ＝∑ x，y (Q i-1，j，m， (cal x (x-d x )，cal y (y-d y ))×Q i，j，e，(x，y) )
B i，j，e，m ＝∑ x，y min(1，(Q i-1，j，m， (cal x (x-d x )，cal y (y-d y ))+Q i，j，e，(x，y) ))
the function O (i, j, e, m) is recorded as the calculated echo block P i，j，e And echo block P i-1，j，m The calculation formula of the shape similarity is:
O(i，j，e，m)＝A i，j，e，m /B i，j，e，m
echo block P i，j，e And echo block P i-1，j，m Is the product of the area similarity, the center similarity and the shape similarity, and a function mark (i, j, e, m) is defined as the calculation echo block P i，j，e And echo block P i-1，j，m The similarity calculation formula is as follows:
mark(i，j，e，m)＝F(i，j，e，m)U(i，j，e，m)O(i，j，e，m)
from the firstEcho block set P i，j And a second echo block set P i-1，j And (3) arbitrarily selecting one echo block for combination, calculating the similarity of two echo blocks in the combination, enabling the similarity of each echo block combination to be the maximum value of the current similarity combination in all obtained non-repeated echo block combinations, and finally obtaining the maximum similarity combination sequence in all echo block combinations. Record set E as first echo block set P i，j Is the second echo block P i-1，j The echo block sequence numbers of (1) are:
wherein u is N, v is N, e k ∈N，m u ∈N。
Let phi be the similarity threshold parameter, if v is present, such that mark (i, j, e v ，m v ) Echo block > phiAnd echo block->The two echo blocks are not the same cloud, and are reflected on adjacent time of the same cloud.
Furthermore, in order to make the matching result of the echo blocks in the level matrixes of different levels more accurate, the echo block combination of the level matrixes of the same level of the adjacent radar echo images can be performed based on the relation of the echo blocks in the level matrixes of different levels of one radar echo image, so that the similarity of the echo block combination is calculated. Specifically, according to the inclusion relationship of the coordinate sets corresponding to two echo blocks of adjacent layers of the same radar echo image, the relationship of the two echo blocks of the adjacent layers can be judged. If the coordinate set of the high-level echo block is included in the coordinate set of the low-level echo block, the high-level echo block is called a child echo block of the low-level echo block, and the low-level echo block is called a parent echo block of the high-level echo block; if the coordinate set does not have the inclusion relationship, two adjacent level echo blocks do not have the relationship.
For example, if there is a set of echo blocksThen compute echo block set P for the adjacent hierarchy i，j+1 Each echo block and echo block set P i，j Is associated with each echo block. Wherein if r epsilon N and q epsilon N exist, the echo block set P is formed i，j+1 Connected set Ω in (a) i，j+1，q Contained in echo block set P i，j Connected set Ω in (a) i，j，r Echo block P i，j+1，q For echo block P i，j，r Is a sub-echo block of (2), echo block P i，j，r For echo block P i，j+1，q Is included in the parent echo block of (a).
Defining a function parent (i, j, k) for calculating an echo block P i，j，k The function child (i, j, k) is used to calculate P i，j，k The sub echo block sets of (1) are:
m is recorded i，j，k For echo block P i，j，k The sub echo block sets of (1) are:
M i，j，k ＝children(i，j，k)
in order to record the relation among echo blocks conveniently, a tree-shaped relation diagram among echo blocks of adjacent level matrixes on the same radar echo image is established based on a tree analysis method. The tree root is a special echo block, and the sub of the tree rootThe echo block set is all echo blocks of the lowest level, and the parent echo block of all echo blocks of the lowest level is the tree root. Record P i Tree root P expressed as tree root of the ith tree corresponding to the ith image i Is set as all sub echo blocks of (1)Wherein s.epsilon.N. From each echo block P of the lowest hierarchy i，0，s Initially, according to each echo block P i，0，s Corresponding sub-echo block set M i，0，s Obtaining each P i，0，s And so on, thereby obtaining all the sub-echo blocks of each echo block in the next level layer by layer. Wherein, if the sub echo block set +.>Echo block P i，j，k Is tree root P i If the sub-echo block set is +.>Echo block P i，j，k Is of tree root P i There is a set of sub-echo blocks. And calculating a sub-echo block set of the next adjacent layer of each echo block layer by layer to obtain the tree-like relation of the ith tree corresponding to the ith radar echo image.
After all radar echo images establish a tree relationship, the similarity of echo combination can be calculated by combining echo blocks according to the tree relationship of adjacent trees. Specifically, the similarity of echo blocks in the same level is calculated from the lowest level to the higher level of two adjacent trees, and when an echo block is calculated in the j-th levelAnd echo block->Is greater than a first threshold, echo block +.>And echo block/>Is determined to be a similar echo block, and echo block +.>And echo block->The simultaneous presence of a set of sub-echo blocks, i.e. +.>And->In the j+1 level, only the echo block is +.>Sub-echo block set->Is>Sub-echo block set->And performing group sum and calculating the combined similarity calculation of the echo blocks. If echo block->Or echo block->There is no sub-echo block set and echo block +.>Or echo block->For similar echo blocks, then determine that the two echo blocks are reflections of the same cloud at adjacent times, then respond to the j in the same mannerAnd carrying out similarity calculation and judgment on the sub echo block set of the next group of similar echo blocks of the hierarchy. It should be noted that, the calculation of the similarity between echo blocks has been specifically described in the above embodiments, and will not be described herein.
According to some embodiments of the present application, the echo block tracking method of the radar echo image further includes step S510 and step S520:
step S510, when two echo blocks are judged to be the same cloud cluster, determining a previous echo block and a subsequent echo block according to the time sequence of two adjacent radar echo images in the radar echo image sequence;
step S520, the motion trail of the cloud cluster is displayed by plotting the previous echo block and the subsequent echo block.
In some embodiments, after determining that two echo blocks in the same hierarchical matrix are reflections of the same cloud on adjacent times according to the time sequence of two adjacent radar echo images in the radar echo image sequence, the echo block with the earlier time is referred to as a previous echo block of the echo block with the later time, and the echo block with the later time is referred to as a subsequent echo block of the echo block with the earlier time. For example, echo blocksAnd echo block->For the reflection of the same cloud on adjacent time, the echo block is called +.>For echo block->Before echo block of (2), echo block->For echo block->Is a subsequent echo block of (c). Then, the display is plotted according to the determined previous echo block and the subsequent echo blockThe motion track of the cloud cluster specifically acquires all echo blocks and centroids thereof of the j-th level of all images in the radar echo image sequence, for example, acquires the 8-th level echo blocks of 4 radar echo images in the radar echo image sequence shown in fig. 4 and calculates the barycenter coordinates of the echo blocks, then fills the acquired barycenter coordinates of the echo blocks into corresponding positions of a space-time tracking result record table, the space-time tracking result record table is shown in fig. 5, when the barycenter coordinates of the echo blocks are filled into the space-time tracking result record table, whether the echo blocks in a level matrix are similar or not needs to be analyzed first, and the sequence relationship between the similar echo blocks is analyzed, namely, the previous echo blocks and the subsequent echo blocks of the same level matrix in the adjacent radar echo images are analyzed, so that the echo block sequences which belong to a certain cloud cluster are identical in the previous continuous images are obtained, and the barycenter of the echo block sequences which belong to a certain cloud cluster are filled into the corresponding positions of the space-time tracking result record table. And displaying the echo block track images of each level of each radar echo image according to the space-time tracking result record table and the acquired echo blocks, wherein the echo block track images are radar echo image cloud cluster motion track tracking images in a radar echo image sequence as shown in fig. 6.
An embodiment of the present application further provides an echo block tracking device for radar echo images, including a memory and a processor, where the memory is configured to store at least one program, and the processor is configured to load the at least one program to perform the echo block tracking method for radar echo images in the above embodiment.
An embodiment of the present application also provides a computer-readable storage medium storing computer-executable instructions for execution by one or more control processors, e.g., to perform the steps described in the above embodiments.
Those of ordinary skill in the art will appreciate that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
Compared with the prior art, the embodiment of the application has the following advantages:
the embodiment of the application has the advantages of accurately acquiring and analyzing the echo block:
according to the embodiment of the application, the radar echo threshold value group is adopted to carry out hierarchical division on the hierarchical matrix corresponding to the radar echo image, and then the feature sets of all echo blocks of each level of one radar echo image are obtained from the hierarchical matrices of a plurality of levels through the communication rule and the constraint condition, so that various features of the echo blocks can be analyzed, and a basis is provided for subsequent similarity calculation of the echo blocks.
The embodiment of the application has the advantage of accurately tracking the echo blocks of the same hierarchy of adjacent images:
according to the embodiment of the application, the similarity of two echo blocks in the same layer of the adjacent image is calculated, and whether the two echo blocks are reflection of the same cloud cluster in adjacent time is judged based on similarity threshold comparison. The method can avoid inaccuracy of echo block matching caused by numerical value or shape change of echo blocks on adjacent layers, and can avoid loss of tracking results of the regional cloud cluster caused by splitting of echo blocks of a single layer by carrying out echo block similarity calculation and comparison in layers. The matching result of the low-level echo block also affects the similarity calculation mode of the high-level echo block, so that the matching object of the high-level echo block is restrained, and inaccurate matching of the high-level echo block is avoided.
The embodiment of the application has the advantage that the moving track of the echo block on a time sequence can be obtained:
each image and each layer corresponds to a space-time tracking result record table. And filling the centroid of each echo block of the current level of the current image in the corresponding position of the corresponding record table, and then analyzing the previous or subsequent relation of echo blocks of the same level in the previous continuous images of the current image to obtain an echo block sequence of the same level and belonging to a cloud cluster in the previous continuous images, and filling the centroid of the echo block sequence of the same level and belonging to the cloud cluster in the previous continuous images in the corresponding position of the record table. Finally, the space-time tracking result of the echo block corresponding to each layer of each image can be accurately obtained, so that the radar echo image echo block can be tracked, and the radar echo image echo block can be used as the basis for subsequent radar echo image extrapolation.
While the application has been described with reference to the preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions may be made therein without departing from the spirit of the application and that these changes and substitutions are intended in the scope of the application as defined by the appended claims.
Claims (8)
1. An echo block tracking method of radar echo images is characterized by comprising the following steps:
acquiring a plurality of radar echo images to form a radar echo image sequence, wherein the radar echo images in the radar echo image sequence are from the same radar and are arranged in time sequence;
performing hierarchical division on each radar echo image in the radar echo image sequence to obtain a plurality of binarized hierarchical matrixes;
determining all echo blocks in the hierarchical matrix according to the communication rule;
calculating the similarity between echo blocks in the level matrix of the same level of two adjacent radar echo images;
judging two echo blocks with similarity larger than a threshold value as the same cloud cluster;
the step of calculating the similarity between echo blocks in the same-level hierarchical matrix of two adjacent radar echo images comprises the following steps:
determining the relation of echo blocks in adjacent level matrixes according to the characteristics of the echo blocks in the level matrixes of the adjacent level of the same radar echo image, wherein when a first echo block coordinate set in a low level matrix comprises a plurality of second echo block coordinate sets in a high level matrix, the first echo block is a father echo block of the second echo block, and the second echo block is a child echo block of the first echo block;
calculating the similarity between a third echo block in a j-th level hierarchical matrix of the ith radar echo image and a fourth echo block in the j-th level hierarchical matrix of the i-1 th radar echo image, wherein the similarity between a father echo block of the third echo block and a father echo block of the fourth echo block is larger than a first threshold value;
wherein the determining that the two echo blocks with similarity larger than the threshold value are the same cloud cluster comprises the following steps:
judging whether the third echo block has a sub echo block according to the echo block relation;
judging whether the fourth echo block has a sub echo block according to the echo block relation;
and when the third echo block does not have the sub echo block or the fourth echo block does not have the sub echo block, and the similarity of the third echo block and the fourth echo block is larger than the threshold value, the third echo block and the fourth echo block are in the same cloud cluster.
2. The method for echo block tracking of radar echo images according to claim 1, wherein the step of hierarchically dividing each radar echo image in the radar echo image sequence to obtain a plurality of binarized hierarchical matrices includes the steps of:
converting the color values of pixels in the radar echo image into radar echo intensity values to obtain an intensity matrix;
and carrying out hierarchical division on the intensity matrix by combining the radar echo intensity threshold value group to obtain a plurality of binarization hierarchical matrices.
3. The method for echo block tracking of radar echo images according to claim 2, wherein the step of hierarchically dividing the intensity matrix in combination with a radar echo intensity threshold set to obtain a plurality of binarized hierarchical matrices includes the steps of:
acquiring a radar intensity echo threshold value of a j-th level according to the radar echo intensity threshold value group, wherein the radar echo intensity threshold value group is an ordered sequence, and the expression of the radar echo intensity threshold value group is as follows:
C＝{C j ＝Z+j(L-z)(z-1) -1 |j∈[0，z)∩N}，C j the radar intensity echo threshold value of the j-th level is represented, L represents the set maximum radar echo intensity value, Z represents the set minimum radar echo intensity value, and Z represents the total level of division;
the ith intensity matrix N according to the radar echo intensity threshold value of the jth level i Performing numerical comparison and assignment to obtain a j-th hierarchical matrix, wherein the element position of the intensity matrix, which is larger than the radar intensity echo threshold value of the j-th hierarchical matrix, is 1, and the element position of the intensity matrix, which is not larger than the radar intensity echo threshold value of the j-th hierarchical matrix, is 0, so as to obtain a j-th binary hierarchical matrix S i,j 。
4. The method for echo block tracking of radar echo images according to claim 1, wherein the calculating of the similarity between echo blocks in the same hierarchical matrix of two adjacent radar echo images includes the steps of:
obtaining a first echo block set according to echo blocks in a j-th hierarchical matrix of the ith radar echo image;
obtaining a second echo block set according to echo blocks in a j-th hierarchical matrix of the i-1 th radar echo image;
calculating the area similarity, the center similarity and the shape similarity of an e-th echo block in the first echo block set and an m-th echo block in the second echo block set;
and determining the similarity of the two echo blocks according to the area similarity, the center similarity and the shape similarity.
5. The method for echo block tracking of radar echo images according to claim 1, wherein the determining all echo blocks in the hierarchical matrix according to the connectivity rule comprises the steps of:
determining a connected set in the hierarchical matrix according to constraint rules of four or eight connections;
and determining all echo blocks in the hierarchical matrix according to the characteristics of the connected set.
6. The method for echo block tracking of a radar echo image according to claim 1, wherein the method for echo block tracking of a radar echo image further comprises the steps of:
when two echo blocks are judged to be the same cloud cluster, determining a previous echo block and a subsequent echo block according to the time sequence of two adjacent radar echo images in the radar echo image sequence;
the motion trajectories of the cloud are displayed by plotting the previous echo block and the subsequent echo block.
7. An echo block tracking device for radar echo images, comprising: a memory for storing at least one program and a processor for loading the at least one program to perform the echo block tracking method of radar echo images of any one of claims 1-6.
8. A computer storage medium in which a processor-executable program is stored, characterized in that the processor-executable program, when executed by the processor, is for implementing an echo block tracking method of a radar echo image according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110332840.8A CN113109816B (en) | 2021-03-29 | 2021-03-29 | Echo block tracking method, device and storage medium of radar echo image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110332840.8A CN113109816B (en) | 2021-03-29 | 2021-03-29 | Echo block tracking method, device and storage medium of radar echo image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113109816A CN113109816A (en) | 2021-07-13 |
| CN113109816B true CN113109816B (en) | 2023-10-03 |
Family
ID=76713127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110332840.8A Active CN113109816B (en) | 2021-03-29 | 2021-03-29 | Echo block tracking method, device and storage medium of radar echo image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113109816B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113466856A (en) * | 2021-08-04 | 2021-10-01 | 广州市气象台 | Forest fire early stage identification and early warning method based on X-band dual-polarization phased array radar |
| CN114691918B (en) * | 2022-04-01 | 2023-06-02 | 北京百度网讯科技有限公司 | Radar image retrieval method and device based on artificial intelligence and electronic equipment |
| CN114814991A (en) * | 2022-06-28 | 2022-07-29 | 成都信息工程大学 | Method and device for identifying spatial distribution area of meteorological element field |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003270357A (en) * | 2002-03-13 | 2003-09-25 | Mitsubishi Electric Corp | Visibility learning device |
| JP2009075017A (en) * | 2007-09-21 | 2009-04-09 | Central Res Inst Of Electric Power Ind | Surface flow rate estimation method, device, and program |
| CN101937078A (en) * | 2009-06-30 | 2011-01-05 | 深圳市气象局 | Nowcasting method and system of thunder cloud cluster based on boundary recognition and tracer technique |
| CN106093933A (en) * | 2016-08-22 | 2016-11-09 | 电子科技大学 | A kind of multipaths restraint method of multiple target side wall after through-wall radar imaging |
| CN110632574A (en) * | 2019-11-12 | 2019-12-31 | 徐州市气象局 | Hailstorm monomer echo identification, tracking and short-time forecasting system |
| CN110942111A (en) * | 2019-12-31 | 2020-03-31 | 北京弘象科技有限公司 | Method and device for identifying strong convection cloud cluster |
| CN111625993A (en) * | 2020-05-25 | 2020-09-04 | 中国水利水电科学研究院 | Small watershed surface rainfall interpolation method based on mountainous terrain and rainfall characteristic prediction |
| CN112558022A (en) * | 2020-11-02 | 2021-03-26 | 广东工业大学 | Radar echo image processing method, system, device and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7391358B2 (en) * | 2005-06-30 | 2008-06-24 | Massachusetts Institute Of Technology | Weather radar echo tops forecast generation |
-
2021
- 2021-03-29 CN CN202110332840.8A patent/CN113109816B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003270357A (en) * | 2002-03-13 | 2003-09-25 | Mitsubishi Electric Corp | Visibility learning device |
| JP2009075017A (en) * | 2007-09-21 | 2009-04-09 | Central Res Inst Of Electric Power Ind | Surface flow rate estimation method, device, and program |
| CN101937078A (en) * | 2009-06-30 | 2011-01-05 | 深圳市气象局 | Nowcasting method and system of thunder cloud cluster based on boundary recognition and tracer technique |
| CN106093933A (en) * | 2016-08-22 | 2016-11-09 | 电子科技大学 | A kind of multipaths restraint method of multiple target side wall after through-wall radar imaging |
| CN110632574A (en) * | 2019-11-12 | 2019-12-31 | 徐州市气象局 | Hailstorm monomer echo identification, tracking and short-time forecasting system |
| CN110942111A (en) * | 2019-12-31 | 2020-03-31 | 北京弘象科技有限公司 | Method and device for identifying strong convection cloud cluster |
| CN111625993A (en) * | 2020-05-25 | 2020-09-04 | 中国水利水电科学研究院 | Small watershed surface rainfall interpolation method based on mountainous terrain and rainfall characteristic prediction |
| CN112558022A (en) * | 2020-11-02 | 2021-03-26 | 广东工业大学 | Radar echo image processing method, system, device and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| An Automatic Tracking and Recognition Algorithm for Thunderstorm Cloud-Cluster;Lan Hongping等;《24th Conference on Severe Local Storms》;第1-4页 * |
| 基于雷达数据云团外推的降雨预测算法研究;李亚;《中国优秀硕士学位论文全文数据库信息科技辑(月刊)》(第6期);第I138-1715页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113109816A (en) | 2021-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113109816B (en) | Echo block tracking method, device and storage medium of radar echo image | |
| CN111191732B (en) | Target detection method based on full-automatic learning | |
| CN109829399B (en) | Vehicle-mounted road scene point cloud automatic classification method based on deep learning | |
| CN111784685A (en) | Power transmission line defect image identification method based on cloud edge cooperative detection | |
| US10275683B2 (en) | Clustering-based person re-identification | |
| CN111160469B (en) | Active learning method of target detection system | |
| CN111626128A (en) | Improved YOLOv 3-based pedestrian detection method in orchard environment | |
| CN111602138B (en) | Object detection system and method based on artificial neural network | |
| CN113012177A (en) | Three-dimensional point cloud segmentation method based on geometric feature extraction and edge perception coding | |
| CN112734931B (en) | Method and system for assisting point cloud target detection | |
| CN114972759A (en) | Remote sensing image semantic segmentation method based on hierarchical contour cost function | |
| CN115147798A (en) | Method, model and device for predicting travelable area and vehicle | |
| CN109118518B (en) | Double-layer multi-Bernoulli random finite ant colony multi-cell tracking method | |
| CN116433050A (en) | Abnormality alarm method and system applied to agricultural big data management system | |
| CN115880533A (en) | Bridge apparent crack identification method based on adaptive subset search and deep learning | |
| CN115984640A (en) | Target detection method, system and storage medium based on combined distillation technology | |
| CN113298822B (en) | Point cloud data selection method and device, equipment and storage medium | |
| CN114819325A (en) | Score prediction method, system, device and storage medium based on graph neural network | |
| CN114972429A (en) | Target tracking method and system for cloud edge collaborative self-adaptive inference path planning | |
| CN113093798B (en) | Unmanned aerial vehicle inspection risk avoiding method and device based on electric field distribution and storage medium | |
| Yasarla et al. | Futuredepth: Learning to predict the future improves video depth estimation | |
| CN111310511A (en) | Method and device for identifying objects | |
| CN112994944B (en) | Network state prediction method | |
| CN115687953B (en) | Target clustering method, medium and device based on Hilbert coding | |
| CN117649635B (en) | Method, system and storage medium for detecting shadow eliminating point of narrow water channel scene |
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 |