CN108959355B - Ship classification method and device and electronic equipment - Google Patents
Ship classification method and device and electronic equipment Download PDFInfo
- Publication number
- CN108959355B CN108959355B CN201810424901.1A CN201810424901A CN108959355B CN 108959355 B CN108959355 B CN 108959355B CN 201810424901 A CN201810424901 A CN 201810424901A CN 108959355 B CN108959355 B CN 108959355B
- Authority
- CN
- China
- Prior art keywords
- ship
- analyzed
- range profile
- dimensional range
- dimensional
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Electromagnetism (AREA)
- Image Analysis (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention provides a ship classification method, a ship classification device and electronic equipment, which are used for determining the running state of a ship to be analyzed, calculating the attitude angle of the ship to be analyzed when the running state is a unidirectional stable running state, acquiring a one-dimensional distance image of the ship to be analyzed, selecting a target one-dimensional distance image with the maximum similarity to the one-dimensional distance image from a pre-constructed one-dimensional distance image simulation database according to the attitude angle and the one-dimensional distance image, and taking the ship type of the target one-dimensional distance image with the maximum similarity as the ship type of the ship to be analyzed. By the method, the classification of the ships can be realized, and the problem that the type of the ship is determined based on the one-dimensional distance image in the prior art is solved.
Description
Technical Field
The invention relates to the field of shape recognition, in particular to a ship classification method, a ship classification device and electronic equipment.
Background
Nowadays, radar is often used to acquire one-dimensional range images of ship targets. Among the vessels are ships and commercial vessels. The one-dimensional distance image is the vector sum of the scattering point sub-echo obtained by the broadband radar signal projected on the radar sight line, the structural distribution and the geometric shape of the ship target are displayed, the distribution of the scattering points of the ship target along the distance direction is shown, and important information is provided for the ship category identification.
When a ship runs on the sea, in order to ensure the safety of the sea, the ship is often identified, for example, whether the ship is a ship or a commercial ship is judged, but how to determine the type of the ship based on a one-dimensional distance image is a problem that needs to be solved urgently by a person skilled in the art.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus and an electronic device for classifying ships, so as to solve the urgent need for determining the type of the ship based on a one-dimensional distance image.
In order to solve the technical problems, the invention adopts the following technical scheme:
a method of sorting vessels comprising:
determining the running state of a ship to be analyzed;
when the running state is a one-way stable running state, calculating the attitude angle of the ship to be analyzed;
acquiring a one-dimensional distance image of the ship to be analyzed;
selecting a target one-dimensional range profile with the maximum similarity to the one-dimensional range profile from a pre-constructed one-dimensional range profile simulation database according to the attitude angle and the one-dimensional range profile;
taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed;
the construction process of the one-dimensional range profile simulation database comprises the following steps:
acquiring a plurality of ship photos under different attitude angles;
identifying a first target area in each of the ship photos; wherein the first target area is an imaging area of a ship in a ship photograph;
constructing a one-dimensional range profile simulation sample corresponding to each first target area;
and summarizing the one-dimensional range profile simulation samples to obtain the one-dimensional range profile simulation database.
Preferably, constructing a one-dimensional range-image simulation sample corresponding to each first target region includes:
carrying out binarization on the first target area and the non-first target area in each ship photo, and carrying out projection processing in the horizontal direction to obtain a structural distribution amplitude image projected by a plurality of ships in the horizontal direction;
denoising each structural distribution amplitude image to obtain a simulated oscillogram;
calculating the distance resolution of the monitoring radar;
calculating a ship projection length of each ship on a radar azimuth line of the monitoring radar;
calculating the number of sampling points of each of the analog oscillograms based on the ship projection length of each of the ships and the distance resolution;
and resampling the corresponding analog oscillogram by using the calculated sampling points to obtain a one-dimensional range profile simulation sample corresponding to each first target area.
Preferably, selecting a target one-dimensional range profile with the maximum similarity to the one-dimensional range profile from a pre-constructed one-dimensional range profile simulation database according to the attitude angle and the one-dimensional range profile, includes:
identifying a second target region in the one-dimensional range profile; wherein the second target area is an imaging area of the ship to be analyzed in the one-dimensional range image;
denoising the second target region to obtain a denoised second target region;
calculating an attitude angle range within a preset value different from the attitude angle;
selecting one-dimensional range profile simulation samples with corresponding attitude angles within the range of the attitude angles from the one-dimensional range profile simulation database as one-dimensional range profile simulation samples to be analyzed;
calculating the correlation degree of each one-dimensional range profile simulation sample to be analyzed and the denoised second target area;
selecting a corresponding one-dimensional distance image simulation sample with the maximum correlation degree to be analyzed;
and taking the selected one-dimensional range profile simulation sample to be analyzed as the target one-dimensional range profile.
Preferably, determining the driving state of the vessel to be analyzed comprises:
acquiring running data of the ship to be analyzed within preset time;
determining a relative value of the distance change between the ship to be analyzed and the monitoring radar per unit time based on the running data;
determining the relative driving direction of the ship to be analyzed and the monitoring radar according to the relative distance change value per unit time;
calculating the average course change value of the ship to be analyzed in unit time according to the running data;
and determining whether the ship to be analyzed is in a stable running result or not based on the average course change value.
Preferably, when the driving state is a one-way stable driving state, calculating an attitude angle of the ship to be analyzed includes:
and when the relative driving direction is unidirectional driving and the ship to be analyzed stably drives, calculating to obtain the attitude angle according to the driving data and a preset attitude angle calculation formula.
Preferably, identifying a first target area in each of the ship photographs comprises:
carrying out graying and denoising processing on each ship photo to obtain a grayed image of each ship photo;
detecting the contour part of the ship in each gray image by adopting an edge detection algorithm;
and segmenting a first target area from each grayed image based on the detected contour part of the ship in each grayed image.
A watercraft sorting device comprising:
the running state determining module is used for determining the running state of the ship to be analyzed;
the attitude angle calculation module is used for calculating the attitude angle of the ship to be analyzed when the running state is a unidirectional stable running state;
the data acquisition module is used for acquiring a one-dimensional distance image of the ship to be analyzed;
the selecting module is used for selecting a target one-dimensional range profile with the maximum similarity to the one-dimensional range profile from a one-dimensional range profile simulation database constructed by the database construction module according to the attitude angle and the one-dimensional range profile;
the type determining module is used for taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed;
wherein the database construction module comprises:
the photo acquisition sub-module is used for acquiring a plurality of ship photos under different attitude angles;
a first identification submodule for identifying a first target area in each of the ship photographs; wherein the first target area is an imaging area of a ship in a ship photograph;
the sample construction submodule is used for constructing a one-dimensional distance image simulation sample corresponding to each first target area;
and the data summarizing submodule is used for summarizing the one-dimensional distance image simulation samples to obtain the one-dimensional distance image simulation database.
Preferably, the sample construction submodule comprises:
the image processing unit is used for carrying out binarization on the first target area and the non-first target area in each ship photo and carrying out projection processing in the horizontal direction to obtain structural distribution amplitude images projected by a plurality of ships in the horizontal direction;
the de-noising processing unit is used for de-noising each structural distribution amplitude image to obtain a simulated oscillogram;
the resolution calculation unit is used for calculating the distance resolution of the monitoring radar;
a length calculation unit for calculating a ship projection length of each of the ships on a radar azimuth line of the monitoring radar;
a point number calculation unit for calculating the number of sampling points of each of the analog oscillograms based on a ship projection length of each of the ships and the distance resolution;
and the resampling unit is used for resampling the corresponding analog oscillogram by using the calculated sampling points to obtain a one-dimensional range profile simulation sample corresponding to each first target area.
Preferably, the selecting module includes:
the second identification submodule is used for identifying a second target area in the one-dimensional range profile; wherein the second target area is an imaging area of the ship to be analyzed in the one-dimensional range image;
the denoising processing submodule is used for denoising the second target region to obtain a denoised second target region;
the range calculation submodule is used for calculating an attitude angle range within a preset value of the difference between the range and the attitude angle;
the first selection submodule is used for selecting a one-dimensional range profile simulation sample of which the corresponding attitude angle is within the range of the attitude angle from the one-dimensional range profile simulation database as a one-dimensional range profile simulation sample to be analyzed;
the correlation degree calculation operator module is used for calculating the correlation degree of each one-dimensional range profile simulation sample to be analyzed and the denoised second target area;
the second selection submodule is used for selecting the corresponding one-dimensional distance image simulation sample with the maximum correlation degree to be analyzed;
and the sample determining submodule is used for taking the selected one-dimensional range profile simulation sample to be analyzed as the target one-dimensional range profile.
An electronic device, comprising: a memory and a processor;
wherein the memory is used for storing programs;
the processor calls a program and is used to:
determining the running state of a ship to be analyzed;
when the running state is a one-way stable running state, calculating the attitude angle of the ship to be analyzed;
acquiring a one-dimensional distance image of the ship to be analyzed;
selecting a target one-dimensional range profile with the maximum similarity to the one-dimensional range profile from a pre-constructed one-dimensional range profile simulation database according to the attitude angle and the one-dimensional range profile;
taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed;
the construction process of the one-dimensional range profile simulation database comprises the following steps:
acquiring a plurality of ship photos under different attitude angles;
identifying a first target area in each of the ship photos; wherein the first target area is an imaging area of a ship in a ship photograph;
constructing a one-dimensional range profile simulation sample corresponding to each first target area;
and summarizing the one-dimensional range profile simulation samples to obtain the one-dimensional range profile simulation database.
Compared with the prior art, the invention has the following beneficial effects:
the invention provides a ship classification method, a ship classification device and electronic equipment, which are used for determining the running state of a ship to be analyzed, calculating the attitude angle of the ship to be analyzed when the running state is a unidirectional stable running state, acquiring a one-dimensional distance image of the ship to be analyzed, selecting a target one-dimensional distance image with the maximum similarity with the one-dimensional distance image from a pre-constructed one-dimensional distance image simulation database according to the attitude angle and the one-dimensional distance image, and taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed. By the method, the classification of the ships can be realized, and the problem that the type of the ship is determined based on the one-dimensional distance image in the prior art is solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of a method of the present invention for classifying ships;
FIG. 2 is a flow chart of a method of another method of classifying vessels according to the present invention;
FIG. 3 is a flow chart of a method of another method of classifying ships according to the present invention;
FIG. 4 is a schematic view of a scenario for calculating the projected length of a ship according to the present invention;
FIG. 5 is a flow chart of a method of another method of classifying ships in accordance with the present invention;
fig. 6 is a schematic structural diagram of a ship classification device provided by the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
An embodiment of the present invention provides a ship classification method, and with reference to fig. 1, the method may include:
s11, determining the running state of the ship to be analyzed;
the driving state refers to a relative driving direction to the monitoring radar and a state of maneuvering, disturbance, or stable driving. The relative driving direction comprises three conditions of the ship to be analyzed far away from the monitoring radar, the ship to be analyzed close to the monitoring radar and the motion direction is not clear.
Optionally, on the basis of this embodiment, step S11 may include:
1) acquiring running data of the ship to be analyzed within preset time;
the preset time can be five minutes, the running data of the ship to be analyzed can comprise distance information R (i), and the distance information refers to a distance value of the distance monitoring radar. Time t (i), where i 1.., N, indicates the order in which the positions were acquired.
In addition, the travel data may further include heading information a (i) and azimuth information h (i). The heading information refers to an included angle between the driving direction of the ship to be analyzed and the due north direction, and the azimuth information refers to an included angle between the radar sight line direction and the due north direction.
2) Determining a relative value of the distance change between the ship to be analyzed and the monitoring radar per unit time based on the running data;
in this embodiment, the definition of the relative value of the distance change is:
for example, taking the ship running in the fourth second and the fifth second as an example, the relative value of the distance change in the fourth second is defined as:
and when the difference between the distance value of the fifth second and the distance value of the fourth second is greater than zero, setting the distance change relative value to be 1, if the difference is equal to zero, setting the distance change relative value to be 0, and if the difference is less than zero, setting the distance change relative value to be-1. According to the above rules, the relative value of the distance change between the ship to be analyzed and the monitoring radar per unit time can be calculated.
3) Determining the relative driving direction of the ship to be analyzed and the monitoring radar according to the relative distance change value per unit time;
specifically, set thNUM，thNUMThe threshold is set by a technician according to a specific scenario. If | ∑ Ri|＞thNUMThe direction of the track motion of the ship to be analyzed is considered to be monotonous. On this basis, if ∑ RiIf the ship moves more than 0, the ship moves in the direction away from the monitoring radar, sigma RiIf the ship moving direction is less than 0, the ship moving direction is considered as the ship to be analyzed approaching the monitoring radar.
If | ∑ Ri|＞thNUMIf the situation is not true, the movement direction is considered to have the uncertain situation, and the movement direction of the ship to be analyzed is unknown.
4) Calculating the average course change value of the ship to be analyzed in unit time according to the running data;
the calculation formula of the average course change value is as follows:
5) and determining whether the ship to be analyzed is in a stable running result or not based on the average course change value.
Specifically, a heading metric indicative function is given
Sign function giving change of orientation
If: sigma Ri＞thNUM1Or Σ Ri_sigh＞thNUM2If yes, then judging the course is a maneuvering or disorderly state, wherein thF、ΔFTo set threshold value, thNUM1、thNUM2For doors in abnormal conditionsA threshold limit value.
Otherwise, the state is judged to be a stable state.
S12, when the driving state is a one-way stable driving state, calculating the attitude angle of the ship to be analyzed;
optionally, on the basis of this embodiment, step S12 may include:
and when the relative driving direction is unidirectional driving and the ship to be analyzed stably drives, calculating to obtain the attitude angle according to the driving data and a preset attitude angle calculation formula.
Specifically, the unidirectional travel relative to the travel direction may include that the ship to be analyzed is far from the monitoring radar and the ship to be analyzed is close to the monitoring radar, that is, the ship to be analyzed does not belong to an ambiguous movement direction. The stable running of the ship to be analyzed means that the ship is not in a course maneuvering or disorder state.
Set the attitude angle at
And then the preset attitude angle calculation formula is as follows:
wherein the content of the first and second substances,
the ship to be analyzed is in a one-way stable running state, so that the attitude angle is fixed, and the attitude angle is calculated at any time and is used as the attitude angle of the ship to be analyzed.
S13, acquiring a one-dimensional distance image of the ship to be analyzed;
s14, selecting a target one-dimensional range profile with the maximum similarity to the one-dimensional range profile from a pre-constructed one-dimensional range profile simulation database according to the attitude angle and the one-dimensional range profile;
a large number of one-dimensional range profile simulation samples of various ships are stored in a pre-constructed one-dimensional range profile simulation database.
S15, taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed;
specifically, since the similarity between the one-dimensional range profile of the ship to be analyzed and the target one-dimensional range profile is the largest, the structure of the ship to be analyzed and the structure of the target one-dimensional range profile can be considered to be similar, and the type of the ship of the target one-dimensional range profile can be used as the type of the ship to be analyzed.
On the basis of the present embodiment, referring to fig. 2, the building process of the one-dimensional distance image simulation database may include:
s21, acquiring a plurality of ship photos under different attitude angles;
specifically, a camera may be used to take a picture of a ship, where the ship may be a commercial ship, a ship, or the like.
S22, identifying a first target area in each ship picture; wherein the first target area is an imaging area of a ship in a ship photograph;
specifically, step S22 may include:
1) carrying out graying and denoising processing on each ship photo to obtain a grayed image of each ship photo;
among them, an image captured using a camera is a color image.
In the RGB model, if R ═ G ═ B, the color represents a gray color, where the value of R ═ G ═ B is called the gray value, so that each pixel of the gray image only needs one byte to store the gray value (also called the intensity value and the brightness value), and the gray range is 0 to 255. The color image is grayed by four methods, namely a component method, a maximum value method, an average value method and a weighted average method.
The denoising method may use any denoising method.
2) Detecting the contour part of the ship in each gray image by adopting an edge detection algorithm;
3) and segmenting a first target area from each grayed image based on the detected contour part of the ship in each grayed image.
Specifically, the contour of the ship in the ship picture can be determined by adopting an edge detection algorithm, and after the contour of the ship is determined, the first target area is segmented from the gray-scale image based on the position of the contour.
In addition, the first target region finally obtained may be a color image without performing the graying process.
S23, constructing a one-dimensional distance image simulation sample corresponding to each first target area;
specifically, the first target area is obtained, and this step is to obtain a one-dimensional distance image simulation sample corresponding to the first target area.
And S24, summarizing the one-dimensional range profile simulation samples to obtain the one-dimensional range profile simulation database.
In this embodiment, a running state of a ship to be analyzed is determined, when the running state is a unidirectional stable running state, an attitude angle of the ship to be analyzed is calculated, a one-dimensional distance image of the ship to be analyzed is obtained, a target one-dimensional distance image with the maximum similarity to the one-dimensional distance image is selected from a one-dimensional distance image simulation database established in advance according to the attitude angle and the one-dimensional distance image, and a ship type of the target one-dimensional distance image is used as a ship type of the ship to be analyzed. By the method, the classification of the ships can be realized, and the problem that the type of the ship is determined based on the one-dimensional distance image in the prior art is solved.
Optionally, on the basis of the above embodiment of the process for constructing the one-dimensional range profile simulation database, referring to fig. 3, constructing a one-dimensional range profile simulation sample corresponding to each first target region may include:
s31, binarizing the first target area and the non-first target area in each ship photo, and performing projection processing in the horizontal direction to obtain structural distribution amplitude images of a plurality of ships projected in the horizontal direction;
specifically, the binarization of the image is to set the gray value of a pixel point on the image to be 0 or 255, that is, the whole image has an obvious visual effect of only black and white.
The non-first target area in the ship photo is the background of the first target area, the first target area and the background are subjected to binarization processing and projected in the horizontal direction, and a structure distribution amplitude image projected in the horizontal direction of the ship can be obtained, and the structure distribution amplitude image can represent the structure of the ship.
S32, denoising each structural distribution amplitude image to obtain a simulated oscillogram;
specifically, denoising may be performed by using a minimum value of 20 consecutive mean values as a numerical value of the current coordinate.
S33, calculating the distance resolution of the monitoring radar;
specifically, the range resolution of the monitoring radar depends on the bandwidth and the electromagnetic wave propagation speed, and if the radar bandwidth is B, the resolution is Δ R, and the electromagnetic wave propagation speed is c, the range resolution is: Δ R ═ c/2B, i.e., the size represented by a distance cell.
S34, calculating the ship projection length of each ship on the radar azimuth line of the monitoring radar;
specifically, referring to FIG. 4, the attitude angle, i.e., θ in FIG. 4, the actual size L of the vessel is known at the time the picture is takenRAccording to Lp＝LRCalculating the ship projection length L of the ship on the radar azimuth line of the monitoring radar by cos thetap。
S35, calculating the sampling point number of each analog oscillogram based on the ship projection length of each ship and the distance resolution;
specifically, LPThe integral part of the/delta R is the number of sampling points.
And S36, resampling the corresponding analog oscillogram by using the calculated sampling points to obtain the specific one-dimensional distance image simulation sample corresponding to each first target area, and determining the position of each sampling point by using the calculated sampling points, wherein the position of each sampling point can be determined by using an average value method and other methods.
After the position of each sampling point is determined, the analog oscillogram is sampled, so that numerical values the same as the number of the sampling points can be obtained, and a series of arrays are formed, namely the one-dimensional range profile simulation sample. One-dimensional range images hrrp (i), i 1, 2.
In this embodiment, the number of sampling points can be calculated, and the de-simulation oscillogram is sampled to obtain a one-dimensional range profile simulation sample corresponding to each first target area.
Optionally, on the basis of any of the foregoing embodiments, referring to fig. 5, step S14 may include:
s41, identifying a second target area in the one-dimensional range profile;
wherein the second target area is an imaging area of the ship to be analyzed in the one-dimensional range image.
In particular, the second target region may be referred to as ROI. The process of determining the second target area is as follows:
1) p (i) represents a one-dimensional range profile HRRP, where i is 1.
2) And (3) calculating expectation of the normalized one-dimensional distance image:
3) and calculating the starting position and the ending position of the second target area, and recording as follows: k1,...,M1. Wherein, K1,...,M1Is a starting value and an ending value of the distance unit number selected from i 1.
Specifically, the selection satisfies
The maximum value of the selected distance cell number is M1Taking the minimum value of the selected distance unit number as K1. Where L is set in advance by the technician.
For example, there are a total of 12345 five points, i.e., five distance units, if satisfied
Is the 4 th and 5 th points, then K1Is 4, M1Is 5.
4) For the second target area K1,...,M1And expanding a certain distance unit number to ensure that the second target area is completely included, and recording as: k,.., M.
In particular, if K1To the left boundary, only for M1Performing right expansion; if M is1To the right boundary, then only for K1Carry out left expansion if K1、M1If none are boundary values, then K is1Performing expansion on M1And performing right expansion.
Specifically, taking 12345 as an example, if the left and right are extended by 1, K is 3, and M is also the same1Already at the right boundary, M is 5.
After K and M are obtained, the imaging area of the ship, i.e. the position of the second target area, can be determined in the one-dimensional range image.
S42, denoising the second target region to obtain a denoised second target region;
specifically, since the one-dimensional range profile of the ship to be analyzed is easily affected by sea clutter, weather and radar signal errors, certain noise is easily generated, and the ship is subjected to noise elimination
Performing rolling smoothing treatment, and recording the smoothed target as: t (i), the method is as follows:
where M represents a smooth scale constant. The obtained T (i) is the denoised second target area.
S43, calculating an attitude angle range within a preset value of the difference between the calculated attitude angle and the preset value;
specifically, the preset value is set by a technician according to a specific use scenario. The preset value may be 5 degrees.
After the attitude angle is determined, a range of 5 degrees up and down of the attitude angle is used as the attitude angle range.
S44, selecting a one-dimensional range profile simulation sample with a corresponding attitude angle within the range of the attitude angle from the one-dimensional range profile simulation database as a one-dimensional range profile simulation sample to be analyzed;
specifically, each one-dimensional range profile simulation sample corresponds to an attitude angle, and the one-dimensional range profile simulation sample with the corresponding attitude angle within the range of the attitude angle is selected.
S45, calculating the correlation degree between each one-dimensional range profile simulation sample to be analyzed and the denoised second target area;
specifically, p (i) represents the denoised second target region, Sj(i) Representing a simulation sample of the jth one-dimensional range profile to be analyzed, i is 1.
Through the formula, the correlation degree between each one-dimensional range profile simulation sample to be analyzed and the denoised second target area can be calculated.
S46, selecting a corresponding one-dimensional distance image simulation sample with the maximum correlation degree to be analyzed;
and S47, taking the selected one-dimensional range profile simulation sample to be analyzed as the target one-dimensional range profile.
Specifically, the ship corresponding to the one-dimensional distance image simulation sample to be analyzed with the largest correlation degree is considered to be the closest to the ship to be analyzed.
In this embodiment, the ship closest to the ship to be analyzed is determined by calculating the correlation between each one-dimensional range profile simulation sample to be analyzed and the one-dimensional range profile.
Alternatively, on the basis of the above embodiment of the ship classification method, another embodiment of the present invention provides a ship classification apparatus, referring to fig. 6, which may include:
a driving state determination module 101, configured to determine a driving state of a ship to be analyzed;
an attitude angle calculation module 102, configured to calculate an attitude angle of the ship to be analyzed when the driving state is a unidirectional stable driving state;
a data acquisition module 103, configured to acquire a one-dimensional range profile of the ship to be analyzed;
a selecting module 104, configured to select, according to the attitude angle and the one-dimensional range profile, a target one-dimensional range profile with the largest similarity to the one-dimensional range profile from a one-dimensional range profile simulation database constructed by a database construction module;
a type determination module 105, configured to use a ship type of the target one-dimensional range image as a ship type of the ship to be analyzed;
wherein the database construction module comprises:
the photo acquisition sub-module is used for acquiring a plurality of ship photos under different attitude angles;
a first identification submodule for identifying a first target area in each of the ship photographs; wherein the first target area is an imaging area of a ship in a ship photograph;
the sample construction submodule is used for constructing a one-dimensional distance image simulation sample corresponding to each first target area;
and the data summarizing submodule is used for summarizing the one-dimensional distance image simulation samples to obtain the one-dimensional distance image simulation database.
Further, the driving state determining module 101 is configured to, when determining the driving state of the ship to be analyzed, specifically:
acquiring running data of the ship to be analyzed within preset time;
determining a relative value of the distance change between the ship to be analyzed and the monitoring radar per unit time based on the running data;
determining the relative driving direction of the ship to be analyzed and the monitoring radar according to the relative distance change value per unit time;
calculating the average course change value of the ship to be analyzed in unit time according to the running data;
and determining whether the ship to be analyzed is in a stable running result or not based on the average course change value.
Further, the attitude angle calculation module 102 is configured to, when the driving state is a unidirectional stable driving state, specifically:
and when the relative driving direction is unidirectional driving and the ship to be analyzed stably drives, calculating to obtain the attitude angle according to the driving data and a preset attitude angle calculation formula.
Further, when the first identification sub-module is configured to identify the first target area in each of the ship photos, it is specifically configured to:
carrying out graying and denoising processing on each ship photo to obtain a grayed image of each ship photo;
detecting the contour part of the ship in each gray image by adopting an edge detection algorithm;
and segmenting a first target area from each grayed image based on the detected contour part of the ship in each grayed image.
In this embodiment, a running state of a ship to be analyzed is determined, when the running state is a unidirectional stable running state, an attitude angle of the ship to be analyzed is calculated, a one-dimensional distance image of the ship to be analyzed is obtained, a target one-dimensional distance image with the maximum similarity to the one-dimensional distance image is selected from a one-dimensional distance image simulation database established in advance according to the attitude angle and the one-dimensional distance image, and a ship type of the target one-dimensional distance image is used as a ship type of the ship to be analyzed. By the method, the classification of the ships can be realized, and the problem that the type of the ship is determined based on the one-dimensional distance image in the prior art is solved.
It should be noted that, for the working processes of each module and sub-module in this embodiment, please refer to the corresponding description in the above embodiments, which is not described herein again.
Optionally, on the basis of any one of the above embodiments, the sample construction sub-module may include:
the image processing unit is used for carrying out binarization on the first target area and the non-first target area in each ship photo and carrying out projection processing in the horizontal direction to obtain structural distribution amplitude images projected by a plurality of ships in the horizontal direction;
the de-noising processing unit is used for de-noising each structural distribution amplitude image to obtain a simulated oscillogram;
the resolution calculation unit is used for calculating the distance resolution of the monitoring radar;
a length calculation unit for calculating a ship projection length of each of the ships on a radar azimuth line of the monitoring radar;
a point number calculation unit for calculating the number of sampling points of each of the analog oscillograms based on a ship projection length of each of the ships and the distance resolution;
and the resampling unit is used for resampling the corresponding analog oscillogram by using the calculated sampling points to obtain a one-dimensional range profile simulation sample corresponding to each first target area.
In this embodiment, the number of sampling points can be calculated, and the analog oscillogram is sampled to obtain a one-dimensional range profile simulation sample corresponding to each first target area.
It should be noted that, for the working processes of each module, sub-module, and unit in this embodiment, please refer to the corresponding description in the above embodiments, which is not described herein again.
Optionally, on the basis of any of the above embodiments, the selecting module includes:
the selecting module comprises:
the second identification submodule is used for identifying a second target area in the one-dimensional range profile; wherein the second target area is an imaging area of the ship to be analyzed in the one-dimensional range image;
the denoising processing submodule is used for denoising the second target region to obtain a denoised second target region;
the range calculation submodule is used for calculating an attitude angle range within a preset value of the difference between the range and the attitude angle;
the first selection submodule is used for selecting a one-dimensional range profile simulation sample of which the corresponding attitude angle is within the range of the attitude angle from the one-dimensional range profile simulation database as a one-dimensional range profile simulation sample to be analyzed;
the correlation degree calculation operator module is used for calculating the correlation degree of each one-dimensional range profile simulation sample to be analyzed and the denoised second target area;
the second selection submodule is used for selecting the corresponding one-dimensional distance image simulation sample with the maximum correlation degree to be analyzed;
and the sample determining submodule is used for taking the selected one-dimensional range profile simulation sample to be analyzed as the target one-dimensional range profile.
In this embodiment, the ship closest to the ship to be analyzed is determined by calculating the correlation between each one-dimensional range profile simulation sample to be analyzed and the one-dimensional range profile.
It should be noted that, for the working processes of each module and sub-module in this embodiment, please refer to the corresponding description in the above embodiments, which is not described herein again.
Optionally, on the basis of the embodiments of the ship classification method and apparatus, another embodiment of the present invention provides an electronic device, including: a memory and a processor;
wherein the memory is used for storing programs;
the processor calls a program and is used to:
determining the running state of a ship to be analyzed;
when the running state is a one-way stable running state, calculating the attitude angle of the ship to be analyzed;
acquiring a one-dimensional distance image of the ship to be analyzed;
selecting a target one-dimensional range profile with the maximum similarity to the one-dimensional range profile from a pre-constructed one-dimensional range profile simulation database according to the attitude angle and the one-dimensional range profile;
taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed;
the construction process of the one-dimensional range profile simulation database comprises the following steps:
acquiring a plurality of ship photos under different attitude angles;
identifying a first target area in each of the ship photos; wherein the first target area is an imaging area of a ship in a ship photograph;
constructing a one-dimensional range profile simulation sample corresponding to each first target area;
and summarizing the one-dimensional range profile simulation samples to obtain the one-dimensional range profile simulation database.
In this embodiment, a running state of a ship to be analyzed is determined, when the running state is a unidirectional stable running state, an attitude angle of the ship to be analyzed is calculated, a one-dimensional distance image of the ship to be analyzed is obtained, a target one-dimensional distance image with the maximum similarity to the one-dimensional distance image is selected from a one-dimensional distance image simulation database established in advance according to the attitude angle and the one-dimensional distance image, and a ship type of the target one-dimensional distance image is used as a ship type of the ship to be analyzed. By the method, the classification of the ships can be realized, and the problem that the type of the ship is determined based on the one-dimensional distance image in the prior art is solved.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A method of sorting vessels, comprising:
determining the running state of a ship to be analyzed;
when the running state is a one-way stable running state, calculating the attitude angle of the ship to be analyzed;
acquiring a one-dimensional distance image of the ship to be analyzed;
selecting a target one-dimensional range profile with the maximum similarity to the one-dimensional range profile from a pre-constructed one-dimensional range profile simulation database according to the attitude angle and the one-dimensional range profile, wherein the method comprises the following steps: identifying a second target region in the one-dimensional range profile; wherein the second target area is an imaging area of the ship to be analyzed in the one-dimensional range image; denoising the second target region to obtain a denoised second target region; calculating an attitude angle range within a preset value different from the attitude angle; selecting one-dimensional range profile simulation samples with corresponding attitude angles within the range of the attitude angles from the one-dimensional range profile simulation database as one-dimensional range profile simulation samples to be analyzed; calculating the correlation degree of each one-dimensional range profile simulation sample to be analyzed and the denoised second target area; selecting a corresponding one-dimensional distance image simulation sample with the maximum correlation degree to be analyzed; taking the selected one-dimensional range profile simulation sample to be analyzed as the target one-dimensional range profile;
taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed;
the construction process of the one-dimensional range profile simulation database comprises the following steps:
acquiring a plurality of ship photos under different attitude angles;
identifying a first target area in each of the ship photos; wherein the first target area is an imaging area of a ship in a ship photograph;
constructing a one-dimensional range profile simulation sample corresponding to each first target area;
and summarizing the one-dimensional range profile simulation samples to obtain the one-dimensional range profile simulation database.
2. The method of claim 1, wherein constructing one-dimensional range-image simulation samples for each of the first target areas comprises:
carrying out binarization on the first target area and the non-first target area in each ship photo, and carrying out projection processing in the horizontal direction to obtain a structural distribution amplitude image projected by a plurality of ships in the horizontal direction;
denoising each structural distribution amplitude image to obtain a simulated oscillogram;
calculating the distance resolution of the monitoring radar;
calculating a ship projection length of each ship on a radar azimuth line of the monitoring radar;
calculating the number of sampling points of each of the analog oscillograms based on the ship projection length of each of the ships and the distance resolution;
and resampling the corresponding analog oscillogram by using the calculated sampling points to obtain a one-dimensional range profile simulation sample corresponding to each first target area.
3. The ship classification method according to claim 1, wherein determining the driving state of the ship to be analyzed comprises:
acquiring running data of the ship to be analyzed within preset time;
determining a relative value of the distance change between the ship to be analyzed and the monitoring radar per unit time based on the running data;
determining the relative driving direction of the ship to be analyzed and the monitoring radar according to the relative distance change value per unit time;
calculating the average course change value of the ship to be analyzed in unit time according to the running data;
and determining whether the ship to be analyzed is in a stable running result or not based on the average course change value.
4. The method of claim 3, wherein calculating the attitude angle of the vessel to be analyzed when the driving state is a one-way steady driving state comprises:
and when the relative driving direction is unidirectional driving and the ship to be analyzed stably drives, calculating to obtain the attitude angle according to the driving data and a preset attitude angle calculation formula.
5. The method of claim 1, wherein identifying a first target area in each of the photographs of the vessel comprises:
carrying out graying and denoising processing on each ship photo to obtain a grayed image of each ship photo;
detecting the contour part of the ship in each gray image by adopting an edge detection algorithm;
and segmenting a first target area from each grayed image based on the detected contour part of the ship in each grayed image.
6. A watercraft sorting device, comprising:
the running state determining module is used for determining the running state of the ship to be analyzed;
the attitude angle calculation module is used for calculating the attitude angle of the ship to be analyzed when the running state is a unidirectional stable running state;
the data acquisition module is used for acquiring a one-dimensional distance image of the ship to be analyzed;
a selecting module, configured to select a target one-dimensional range profile with the largest similarity to the one-dimensional range profile from a one-dimensional range profile simulation database constructed by a database construction module according to the attitude angle and the one-dimensional range profile, where the selecting module includes: the second identification submodule is used for identifying a second target area in the one-dimensional range profile; wherein the second target area is an imaging area of the ship to be analyzed in the one-dimensional range image; the denoising processing submodule is used for denoising the second target region to obtain a denoised second target region; the range calculation submodule is used for calculating an attitude angle range within a preset value of the difference between the range and the attitude angle; the first selection submodule is used for selecting a one-dimensional range profile simulation sample of which the corresponding attitude angle is within the range of the attitude angle from the one-dimensional range profile simulation database as a one-dimensional range profile simulation sample to be analyzed; the correlation degree calculation operator module is used for calculating the correlation degree of each one-dimensional range profile simulation sample to be analyzed and the denoised second target area; the second selection submodule is used for selecting the corresponding one-dimensional distance image simulation sample with the maximum correlation degree to be analyzed; the sample determination submodule is used for taking the selected one-dimensional range profile simulation sample to be analyzed as the target one-dimensional range profile;
the type determining module is used for taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed;
wherein the database construction module comprises:
the photo acquisition sub-module is used for acquiring a plurality of ship photos under different attitude angles;
a first identification submodule for identifying a first target area in each of the ship photographs; wherein the first target area is an imaging area of a ship in a ship photograph;
the sample construction submodule is used for constructing a one-dimensional distance image simulation sample corresponding to each first target area;
and the data summarizing submodule is used for summarizing the one-dimensional distance image simulation samples to obtain the one-dimensional distance image simulation database.
7. The watercraft sorting device of claim 6, wherein the sample construction sub-module comprises:
the image processing unit is used for carrying out binarization on the first target area and the non-first target area in each ship photo and carrying out projection processing in the horizontal direction to obtain structural distribution amplitude images projected by a plurality of ships in the horizontal direction;
the de-noising processing unit is used for de-noising each structural distribution amplitude image to obtain a simulated oscillogram;
the resolution calculation unit is used for calculating the distance resolution of the monitoring radar;
a length calculation unit for calculating a ship projection length of each of the ships on a radar azimuth line of the monitoring radar;
a point number calculation unit for calculating the number of sampling points of each of the analog oscillograms based on a ship projection length of each of the ships and the distance resolution;
and the resampling unit is used for resampling the corresponding analog oscillogram by using the calculated sampling points to obtain a one-dimensional range profile simulation sample corresponding to each first target area.
8. An electronic device, comprising: a memory and a processor;
wherein the memory is used for storing programs;
the processor calls a program and is used to:
determining the running state of a ship to be analyzed;
when the running state is a one-way stable running state, calculating the attitude angle of the ship to be analyzed;
acquiring a one-dimensional distance image of the ship to be analyzed;
selecting a target one-dimensional range profile with the maximum similarity to the one-dimensional range profile from a pre-constructed one-dimensional range profile simulation database according to the attitude angle and the one-dimensional range profile, wherein the method comprises the following steps: identifying a second target region in the one-dimensional range profile; wherein the second target area is an imaging area of the ship to be analyzed in the one-dimensional range image; denoising the second target region to obtain a denoised second target region; calculating an attitude angle range within a preset value different from the attitude angle; selecting one-dimensional range profile simulation samples with corresponding attitude angles within the range of the attitude angles from the one-dimensional range profile simulation database as one-dimensional range profile simulation samples to be analyzed; calculating the correlation degree of each one-dimensional range profile simulation sample to be analyzed and the denoised second target area; selecting a corresponding one-dimensional distance image simulation sample with the maximum correlation degree to be analyzed; taking the selected one-dimensional range profile simulation sample to be analyzed as the target one-dimensional range profile;
taking the ship type of the target one-dimensional distance image as the ship type of the ship to be analyzed;
the construction process of the one-dimensional range profile simulation database comprises the following steps:
acquiring a plurality of ship photos under different attitude angles;
identifying a first target area in each of the ship photos; wherein the first target area is an imaging area of a ship in a ship photograph;
constructing a one-dimensional range profile simulation sample corresponding to each first target area;
and summarizing the one-dimensional range profile simulation samples to obtain the one-dimensional range profile simulation database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810424901.1A CN108959355B (en) | 2018-05-04 | 2018-05-04 | Ship classification method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810424901.1A CN108959355B (en) | 2018-05-04 | 2018-05-04 | Ship classification method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108959355A CN108959355A (en) | 2018-12-07 |
| CN108959355B true CN108959355B (en) | 2021-04-20 |
Family
ID=64499041
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810424901.1A Active CN108959355B (en) | 2018-05-04 | 2018-05-04 | Ship classification method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108959355B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112098965B (en) * | 2020-11-04 | 2021-02-19 | 长沙莫之比智能科技有限公司 | Carrier attitude angle resolving method based on millimeter wave radar |
| CN112257811B (en) * | 2020-11-11 | 2022-04-19 | 珠海大横琴科技发展有限公司 | Ship classification method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102636782A (en) * | 2012-04-29 | 2012-08-15 | 西安电子科技大学 | Super-resolution one-dimensional distance imaging method of step frequency radar |
| CN104459663A (en) * | 2014-11-27 | 2015-03-25 | 中国船舶重工集团公司第七二四研究所 | Naval vessel and cargo vessel classification method based on high-resolution one-dimensional range profile |
| CN104865562A (en) * | 2015-06-12 | 2015-08-26 | 西安电子科技大学 | Identification method for radar disoperative target based on mixed model |
| CN107765240A (en) * | 2017-10-20 | 2018-03-06 | 中国人民解放军海军七〇工厂 | A kind of determination methods of motion state, device and electronic equipment |
-
2018
- 2018-05-04 CN CN201810424901.1A patent/CN108959355B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102636782A (en) * | 2012-04-29 | 2012-08-15 | 西安电子科技大学 | Super-resolution one-dimensional distance imaging method of step frequency radar |
| CN104459663A (en) * | 2014-11-27 | 2015-03-25 | 中国船舶重工集团公司第七二四研究所 | Naval vessel and cargo vessel classification method based on high-resolution one-dimensional range profile |
| CN104865562A (en) * | 2015-06-12 | 2015-08-26 | 西安电子科技大学 | Identification method for radar disoperative target based on mixed model |
| CN107765240A (en) * | 2017-10-20 | 2018-03-06 | 中国人民解放军海军七〇工厂 | A kind of determination methods of motion state, device and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 舰船目标一维距离像多姿态角相关匹配识别;孙剑波等;《船舶电子工程》;20111130;第31卷(第11期);第49-51页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108959355A (en) | 2018-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109635685B (en) | Target object 3D detection method, device, medium and equipment | |
| US10346720B2 (en) | Rotation variant object detection in Deep Learning | |
| CN108805904B (en) | Moving ship detection and tracking method based on satellite sequence image | |
| US5081689A (en) | Apparatus and method for extracting edges and lines | |
| US9031285B2 (en) | Detection of floating objects in maritime video using a mobile camera | |
| CN101609504B (en) | Method for detecting, distinguishing and locating infrared imagery sea-surface target | |
| US20120328161A1 (en) | Method and multi-scale attention system for spatiotemporal change determination and object detection | |
| CN108573499B (en) | Visual target tracking method based on scale self-adaption and occlusion detection | |
| CN102598057A (en) | Method and system for automatic object detection and subsequent object tracking in accordance with the object shape | |
| CN106228570B (en) | A kind of Truth data determines method and apparatus | |
| CN113850848A (en) | Unmanned boat-mounted marine radar and visual image cooperative marine multi-target long-term detection and tracking method | |
| CN109063669B (en) | Bridge area ship navigation situation analysis method and device based on image recognition | |
| CN116310837B (en) | SAR ship target rotation detection method and system | |
| CN108959355B (en) | Ship classification method and device and electronic equipment | |
| CN111160477A (en) | Image template matching method based on feature point detection | |
| Benedek et al. | Moving target analysis in ISAR image sequences with a multiframe marked point process model | |
| CN114821358A (en) | Optical remote sensing image marine ship target extraction and identification method | |
| CN108919284B (en) | Ship classification method and device and electronic equipment | |
| CN111914627A (en) | Vehicle identification and tracking method and device | |
| EP4323952A1 (en) | Semantically accurate super-resolution generative adversarial networks | |
| Bisht et al. | Integration of hough transform and inter-frame clustering for road lane detection and tracking | |
| Shi et al. | Obstacle type recognition in visual images via dilated convolutional neural network for unmanned surface vehicles | |
| Hashmani et al. | A survey on edge detection based recent marine horizon line detection methods and their applications | |
| Mou et al. | Image-based maritime obstacle detection using global sparsity potentials | |
| CN108985216B (en) | Pedestrian head detection method based on multivariate logistic regression feature fusion |
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 |