CN117111072A - Ship target detection method and device, electronic equipment and storage medium - Google Patents

Ship target detection method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN117111072A
CN117111072A CN202310869027.3A CN202310869027A CN117111072A CN 117111072 A CN117111072 A CN 117111072A CN 202310869027 A CN202310869027 A CN 202310869027A CN 117111072 A CN117111072 A CN 117111072A
Authority
CN
China
Prior art keywords
remote sensing
space
based remote
detection
sensing detection
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310869027.3A
Other languages
Chinese (zh)
Inventor
牟方厉
樊子德
朱可卿
葛蕴萍
耿莹
王磊
李肖赫
李新明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Aerospace Information Research Institute of CAS
Original Assignee
Aerospace Information Research Institute of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Aerospace Information Research Institute of CAS filed Critical Aerospace Information Research Institute of CAS
Priority to CN202310869027.3A priority Critical patent/CN117111072A/en
Publication of CN117111072A publication Critical patent/CN117111072A/en
Pending legal-status Critical Current

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
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/02—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems using reflection of acoustic waves
    • G01S15/50—Systems of measurement, based on relative movement of the target
    • G01S15/58—Velocity or trajectory determination systems; Sense-of-movement determination 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
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/86—Combinations of sonar systems with lidar systems; Combinations of sonar systems with systems not using wave reflection

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)

Abstract

The invention provides a ship target detection method, a device, electronic equipment and a storage medium, and relates to the technical field of target detection, wherein the method comprises the following steps: determining a first day-base remote sensing detection point set contained in a sonar detection area of a target ship and a sonar detection point set contained in the sonar detection area; determining a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, wherein each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to a space-based remote sensing detection point of the target ship; and optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second space-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as a track of the target ship. According to the invention, the space-based remote sensing data and the sonar information are effectively fused, so that the marine ship target is rapidly and accurately detected.

Description

Ship target detection method and device, electronic equipment and storage medium
Technical Field
The present invention relates to the field of target detection technologies, and in particular, to a method and apparatus for detecting a ship target, an electronic device, and a storage medium.
Background
The automatic detection of the marine ship target has important significance for guaranteeing the safety of the deep sea navigation path, implementing the supervision of the sea area, fighting the marine crime and the like. The current detection of the targets of the marine ships is mostly realized through a sonar, an automatic ship identification system (Automatic Identification System, AIS), a shore radar, a satellite and the like, and under a single information source, the sonar and the sonar networking have the defect of lower detection precision, the precision and coverage capacity of the shore radar are limited, the AIS provides cooperation requirements for the ships, and the sampling and revisiting period of the satellite is longer. Therefore, it is difficult to achieve fast and high-precision tracking of offshore vessels with a single information source.
Because space-based remote sensing has the characteristics of visualization and high precision, sonar information has the advantages of all weather and full time, and therefore, the method has good application value for realizing automatic detection and tracking of marine ship targets by fusing space-based remote sensing and sonar information by utilizing the complementarity between space-based remote sensing data and sea-based sonar data.
However, in a deep sea environment, a large distance exists between a ship target and a deployed sonar array, and the adopted sonar form is mainly passive sonar, so that the sonar detection result of the ship has low precision and large uncertainty. The current space-based remote sensing detection is mainly divided into two types of synthetic aperture radars (Synthetic Aperture Radar, SAR) and optical images, the precision and the resolution of the space-based remote sensing detection are strong, but the ship targets in the space-based remote sensing detection images have the characteristic of very weak and small, and cloud layers, islands and noise points can bring about larger interference, so that false alarms or omission detection are easily caused. In addition, the frequency of the space-based remote sensing detection is significantly lower than that of the sonar detection (the difference is at least 600 times), so that the fusion of the space-based remote sensing data and the sonar information becomes more difficult.
Therefore, how to realize a method for detecting the marine ship target by fusing the space-based remote sensing data and the sonar information, so as to improve the detection efficiency and the detection precision of the marine ship target, and the method is a problem to be solved in the industry.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a ship target detection method, a device, electronic equipment and a storage medium.
In a first aspect, the present invention provides a method for detecting a ship target, including:
determining a first day-based remote sensing detection point set contained in a sonar detection area of a target ship and a sonar detection point set contained in the sonar detection area;
determining a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, wherein each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to the space-based remote sensing detection point of the target ship;
and optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second day-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as the track of the target ship.
Optionally, according to the method for detecting a ship target provided by the present invention, the determining the second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the set of sonar detection points includes:
based on the sonar detection point set, carrying out hypothesis test on the first day-based remote sensing detection point set, and eliminating the first day-based remote sensing detection point from the first day-based remote sensing detection point set to obtain the second day-based remote sensing detection point set under the condition that a first mathematical distribution obeyed by detection errors of the first day-based remote sensing detection points in the first day-based remote sensing detection point set is different from a second mathematical distribution obeyed by detection errors of the second day-based remote sensing detection points in the first day-based remote sensing detection point set;
The second space-based remote sensing detection points are the rest space-based remote sensing detection points except the first space-based remote sensing detection points in the first space-based remote sensing detection point set, and the number of the second space-based remote sensing detection points is larger than that of the first space-based remote sensing detection points.
Optionally, according to the method for detecting a ship target provided by the present invention, the performing, based on the sonar detection point set, hypothesis test on the first day-based remote sensing detection point set includes:
under the condition that the expected value of the false alarm space-based detection points in the first space-based remote sensing detection point set is smaller than the expected value of the real space-based detection points and the number of the real space-based detection points is larger than a set value, carrying out hypothesis testing on the first space-based remote sensing detection point set based on the sonar detection point set;
the real space-based detection points are space-based remote sensing detection points belonging to the target ship and included in the first space-based remote sensing detection point set, and the false alarm space-based detection points are space-based remote sensing detection points not belonging to the target ship in the first space-based remote sensing detection point set.
Optionally, according to the method for detecting a ship target provided by the present invention, the determining the second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the set of sonar detection points further includes:
Generating a space-based remote sensing detection track matrix based on each space-based remote sensing detection point in the first space-based remote sensing detection point set at different sampling moments;
determining a target space-based remote sensing detection track section based on the space-based remote sensing detection track matrix, wherein the track similarity between the target space-based remote sensing detection track section and a target sonar detection track section is maximum, and the target sonar detection track section is a sonar detection track section formed between two adjacent sonar detection points in the sonar detection point set;
determining a complete space-based remote sensing detection track of the target ship based on the target space-based remote sensing detection track section;
and determining the second space-based remote sensing detection point set based on the space-based remote sensing detection points included in the complete space-based remote sensing detection track.
Optionally, according to the method for detecting a ship target provided by the present invention, the determining the second set of space-based remote sensing detection points based on space-based remote sensing detection points included in the complete space-based remote sensing detection track includes:
determining a third day-based remote sensing point, a fourth day-based remote sensing point and a fifth day-based remote sensing point which are sequentially adjacent and are included in the complete day-based remote sensing detection track;
Determining a first track similarity corresponding to a first day-based remote sensing detection track segment formed between the third day-based remote sensing detection point and the fourth day-based remote sensing detection point, determining a second track similarity corresponding to a second day-based remote sensing detection track segment formed between the fourth day-based remote sensing detection point and the fifth day-based remote sensing detection point, and determining a third track similarity corresponding to a third day-based remote sensing detection track segment formed between the third day-based remote sensing detection point and the fifth day-based remote sensing detection point;
when the first track similarity and the third track similarity are both larger than or equal to a preset track similarity threshold value and the second track similarity is smaller than the preset track similarity threshold value, eliminating the fourth space-based remote sensing detection point from space-based remote sensing detection points included in the complete space-based remote sensing detection track to obtain a new complete space-based remote sensing detection track;
and forming the second space-based remote sensing detection point set by the space-based remote sensing detection points included in the new complete space-based remote sensing detection track.
Optionally, according to the method for detecting a ship target provided by the present invention, the optimizing a sonar detection track formed by sonar detection points in the sonar detection point set based on the second space-based remote sensing detection point set, to obtain an optimized sonar detection track includes:
Determining a space-based remote sensing detection track formed by space-based remote sensing detection points in the second space-based remote sensing detection point set;
determining a gravitational field of the space-based remote sensing detection track based on the sonar detection track;
and optimizing the sonar detection track based on the gravitational field to obtain an optimized sonar detection track.
Optionally, according to the method for detecting a ship target provided by the invention, the confidence level of each sonar detection point in the sonar detection area of the target ship is a preset value.
In a second aspect, the present invention also provides a device for detecting a ship target, including:
the first determining module is used for determining a first day-based remote sensing detection point set contained in a sonar detection area of the target ship and a sonar detection point set contained in the sonar detection area;
the second determining module is configured to determine a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, where each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to a space-based remote sensing detection point of the target ship;
And the track optimization module is used for optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second day-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as the track of the target ship.
In a third aspect, the present invention also provides an electronic device, including a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method for detecting a ship target according to the first aspect when executing the program.
In a fourth aspect, the invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of detecting a ship target according to the first aspect.
According to the ship target detection method, device, electronic equipment and storage medium, a first day-based remote sensing detection point set contained in a sonar detection area of a target ship and a sonar detection point set contained in the sonar detection area are firstly determined, then a second day-based remote sensing detection point set is determined based on the first day-based remote sensing detection point set and the sonar detection point set, each day-based remote sensing detection point in the second day-based remote sensing detection point set belongs to the day-based remote sensing detection point of the target ship, and further, a sonar detection track formed by the sonar detection points in the sonar detection point set is optimized based on the second day-based remote sensing detection point set, so that an optimized sonar detection track is obtained, and the optimized sonar detection track is used as the track of the target ship; the method realizes the rapid and accurate detection of the marine ship target by effectively fusing the space-based remote sensing data and the sonar information, and has simple algorithm and easy realization.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow diagram of a method for detecting a ship target provided by the invention;
FIG. 2 is a schematic illustration of an isolated ship track provided by the present invention;
FIG. 3 is a schematic diagram of the relationship between the space-based detection threshold and the space-based false alarm rate provided by the invention;
FIG. 4 is a schematic diagram of the relationship between the number of space-based detections and the space-based detection rate provided by the invention;
FIG. 5 is a schematic diagram of a space-based detection result before space-based and sonar fusion;
FIG. 6 is a schematic diagram of a space-based detection result after the space-based detection result is fused with sonar;
FIG. 7 is a schematic diagram of the optimized sonar flight path based on the space-based detection flight path and the radial base potential field method provided by the invention;
FIG. 8 is a schematic diagram of a sonar track optimization result of the fusion of the space base and the sonar;
FIG. 9 is a schematic structural diagram of a ship target detection device provided by the invention;
fig. 10 is a schematic diagram of the physical structure of the electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that, in the description of the present invention, the terms "first," "second," and the like are used for distinguishing between similar objects and not for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged, as appropriate, such that embodiments of the present invention may be implemented in sequences other than those illustrated or described herein, and that the objects identified by "first," "second," etc. are generally of a type, and are not limited to the number of objects, such as the first object may be one or more.
The method, the device, the electronic equipment and the storage medium for detecting the ship target are described in an exemplary mode by combining the drawings.
Fig. 1 is a schematic flow chart of a method for detecting a ship target, provided by the invention, as shown in fig. 1, the method comprises the following steps:
step 100, determining a first day-based remote sensing detection point set contained in a sonar detection area of a target ship and a sonar detection point set contained in the sonar detection area;
step 110, determining a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, wherein each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to the space-based remote sensing detection point of the target ship;
and 120, optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second space-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as the track of the target ship.
It should be noted that, the execution body of the ship target detection method provided by the embodiment of the invention may be an electronic device, a component in the electronic device, an integrated circuit, or a chip. The electronic device may be a mobile electronic device or a non-mobile electronic device. Illustratively, the mobile electronic device may be a mobile phone, a tablet computer, a notebook computer, a palm computer, a wearable device, an Ultra mobile personal computer (Ultra-mobile Personal Computer, UMPC), a netbook or a personal digital assistant (Personal Digital Assistant, PDA), etc., and the non-mobile electronic device may be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (Personal Computer, PC), a Television (Television, TV), a self-service machine, etc., which is not particularly limited by the embodiments of the present invention.
The technical scheme of the embodiment of the invention is described in detail below by taking a method for detecting the ship target provided by the invention executed by a computer as an example.
Specifically, in order to overcome the defect that the prior art is difficult to realize accurate detection of a marine ship target, the invention optimizes a sonar detection track formed by sonar detection points in a sonar detection area of a target ship to obtain an optimized sonar detection track, and then determines a second space-based remote sensing detection point set based on the first space-based remote sensing detection point set and the sonar detection point set, wherein each space-based remote sensing detection point in the second space-based remote sensing detection point set belongs to the space-based remote sensing detection point of the target ship, and further optimizes the sonar detection track formed by the sonar detection points in the sonar detection point set based on the second space-based remote sensing detection point set to obtain the optimized sonar detection track; the method realizes the rapid and accurate detection of the marine ship target by effectively fusing the space-based remote sensing data and the sonar information, and has simple algorithm and easy realization.
It should be noted that, the track mode of the target ship in the embodiment of the invention is an isolated track mode on water, wherein the isolated track is represented as a discrete track in sonar, and is characterized in that the sonar detection areas of different ship targets are not aliased. For example, fig. 2 is a schematic diagram of an isolated ship track provided by the present invention, and as shown in fig. 2, the sonar detection area of the ship 1 and the sonar detection area of the ship 2 are not aliased, and in this case, the track modes of the ship 1 and the ship 2 are considered to be the isolated track modes.
Alternatively, in the case where the track similarity of the ship 1 and the ship 2 satisfies the following formula (1), the track pattern of the ship 1 and the ship 2 is determined to be the isolated track pattern:
wherein dis (t) represents the track similarity of ship 1 and ship 2 at time t, [ t ] a ，t b ]Representing the analysis period, fh, of the tracks of the vessels 1 and 2 rship Representing the track similarity threshold.
Alternatively, the track similarity dis (t) of the ship 1 and the ship 2 is calculated as shown in formula (2):
wherein,representing the course of the ship 1 +.>Represents the track of the ship 2 and epsilon represents the slack factor.
In the actual space-based remote sensing, the directly obtained space-based remote sensing result cannot determine which ship corresponds to the space-based remote sensing result. Therefore, it is necessary to screen out the space-based remote sensing detection data corresponding to the target ship from the acquired space-based remote sensing detection results.
Because considerable transmission time and processing time are consumed in actual space-based remote sensing detection, the obtained space-based remote sensing detection result is always embodied in the form of historical data. Let S sp|T (t) represents the space-based remote sensing positive sample at time t, { S so (t, α) } represents the sonar detection area at a confidence level α (α may be 95%), then S sp|T (t) and { S ] so (t, α) } has a relationship as shown in the formula (3):
S sp|T (t)∈{S so (t，α)} (3)
alternatively, the directly acquired space-based remote sensing detection result { S }, may be first used sp (t) } and S so And (t, alpha) pre-fusing to identify the space-based remote sensing detection data corresponding to the target ship.
Alternatively, { S sp (t) } and S so (t, alpha) Algorithm input to pre-blend is { S sp (t)}、S so (t, alpha), detection threshold k 1 ，k 2 (k 1 >k 2 )、{S sp The number n of the space-based remote sensing points included in (T) } and the time set T corresponding to each space-based remote sensing point 0 ＝{t 0 ，...，t m The algorithm is output as S in the target ship set and sonar detection area on water so (t，α) And (5) a set of space-based remote sensing detection points. The basic flow of the pre-fusion algorithm comprises the following steps (1) to (6):
(1) For each space-based remote sensing detection point, corresponding to each sonar detection target, checking S sp (t 0 )∈{S so (t 0 Whether alpha) is established or not, and marking the sonar detection targets with the corresponding space-based remote sensing detection points as a set { ID } 0 }；
(2) For the set { ID } 0 Each sonar detection target in, test S was performed within i=1 sp (t 0 )∈{S so (t 0 Whether or not α) is established, and the number of times of establishment of the check is denoted as k 0 ；
(3) Since sonar detection cannot distinguish the above-water and underwater states of the ship targets, and space-based remote sensing can only detect the above-water ship targets, the state of the sonar detection targets can be determined by the space-based remote sensing detection results. If k is 0 ≥k 1 Determining the corresponding sonar detection target as a water ship target, otherwise, when k 0 ≤k 2 When the underwater ship target is determined, the corresponding sonar detection target is determined to be the underwater ship target;
(4) If i=1 for all tests, the term, n, none of the sonar detection targets belongs to { ID } 0 Determining a corresponding sonar detection target as an underwater ship target;
(5) When the space-based detection corresponding to a new sonar detection target appears at the time t new Let t 0 ＝t new Repeatedly executing the steps (2) to (4);
(6) Outputting the set { S of the target ships on water surface And outputs that have been excludedIs a set of space-based remote sensing detection points.
Wherein the threshold k is detected 1 ，k 2 And the number n of the space-based remote sensing detection points is selected by the space-based detection rate p T And space-based false alarm rate p TF Desired detection confidence P R And (5) determining. P is p T Defined as a ship Probability of being detected, p TF The probability that the intersection exists between the space-based remote sensing detection result area and the real ship area is defined. Assuming that each time the space-based remote sensing is independent, the space-based remote sensing can be modeled as a Bernoulli experiment of the sequence, and P can be calculated by the following formula (4) R ：
P R ＝(1-α 1 )(1-α 2 ) (4)
The above formula (5) represents p TF And alpha is 1 ，p T And alpha is 2 The relation to be satisfied is that, wherein, alpha 1 And alpha 2 Representing different significance levels, k 1 Representing the space-based detection threshold.
Fig. 3 is a schematic diagram of the relationship between the space-based detection threshold and the space-based false alarm rate provided by the present invention, wherein the vertical axis in fig. 3 represents the minimum observed number K, namely the space-based detection threshold, and fig. 4 is a schematic diagram of the relationship between the space-based detection number and the space-based detection rate provided by the present invention, wherein k=2 in fig. 4 represents the space-based detection threshold of 2, n represents the space-based detection number, when the significance level α is 1 ＝0.01，α 2 ＝0.05，p TF ≤22.3％,p T When the content is more than or equal to 68.4%, k can be selected 1 =1, n=2, when 22.3%<p TF ≤36.8％,p T When the content is more than or equal to 86.5%, k can be selected 1 ＝2,n＝3。
Similarly, the false alarm rate P of the space-based detection can be calculated by the following formula (6) FR False alarm rate P here FR The ratio between false positive samples and all false samples in the day-based detection:
optionally, after obtaining the first set of space-based remote sensing detection points included in the sonar detection area of the target ship and the set of sonar detection points included in the sonar detection area of the target ship, a second set of space-based remote sensing detection points may be determined based on the first set of space-based remote sensing detection points and the set of sonar detection points, where each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to a space-based remote sensing detection point of the target ship.
Optionally, the confidence level of each sonar detection point in the sonar detection area of the target ship is a preset value, where the preset value may be adaptively set according to practical applications, and the embodiment of the present invention is not specifically limited to this, and for example, the preset value may be 98%, 95% or 90%.
Optionally, after the second space-based remote sensing detection point set is obtained, a sonar detection track formed by the sonar detection points in the sonar detection point set can be optimized based on the second space-based remote sensing detection point set, so that an optimized sonar detection track is obtained, and the optimized sonar detection track is used as a track of the target ship.
According to the ship target detection method, a first day-based remote sensing detection point set contained in a sonar detection area of a target ship and a sonar detection point set contained in the sonar detection area are firstly determined, then a second day-based remote sensing detection point set is determined based on the first day-based remote sensing detection point set and the sonar detection point set, each day-based remote sensing detection point in the second day-based remote sensing detection point set belongs to the day-based remote sensing detection point of the target ship, and further, a sonar detection track formed by the sonar detection points in the sonar detection point set is optimized based on the second day-based remote sensing detection point set, so that an optimized sonar detection track is obtained, and the optimized sonar detection track is used as the track of the target ship; the method realizes the rapid and accurate detection of the marine ship target by effectively fusing the space-based remote sensing data and the sonar information, and has simple algorithm and easy realization.
Optionally, the determining the second set of day-based remote sensing detection points based on the first set of day-based remote sensing detection points and the set of sonar detection points includes:
based on the sonar detection point set, carrying out hypothesis test on the first day-based remote sensing detection point set, and eliminating the first day-based remote sensing detection point from the first day-based remote sensing detection point set to obtain the second day-based remote sensing detection point set under the condition that a first mathematical distribution obeyed by detection errors of the first day-based remote sensing detection points in the first day-based remote sensing detection point set is different from a second mathematical distribution obeyed by detection errors of the second day-based remote sensing detection points in the first day-based remote sensing detection point set;
the second space-based remote sensing detection points are the rest space-based remote sensing detection points except the first space-based remote sensing detection points in the first space-based remote sensing detection point set, and the number of the second space-based remote sensing detection points is larger than that of the first space-based remote sensing detection points.
Specifically, in the embodiment of the present invention, in order to determine a second set of space-based remote sensing points based on the first set of space-based remote sensing points and the sonar set of detection points, the first set of space-based remote sensing points may be first subjected to hypothesis test based on the sonar set of detection points, and if a first mathematical distribution to which detection errors of the first set of space-based remote sensing points obey is determined and a second mathematical distribution to which detection errors of the second set of space-based remote sensing points obey is different, the first set of space-based remote sensing points are removed from the first set of space-based remote sensing points to obtain the second set of space-based remote sensing points, where the second set of space-based remote sensing points is the rest of space-based remote sensing points in the first set of space-based remote sensing points except the first set of space-based remote sensing points, and the number of the second set of space-based remote sensing points is greater than the number of the first set of space-based remote sensing points.
It should be noted that, under the condition that the number of space-based remote sensing detection points which are detected correctly in probability is greater than that of space-based remote sensing detection points which are detected incorrectly, and the constructed space-based remote sensing detection statistics can ensure that the space-based remote sensing detection has significance, the following algorithm 1 can be used for performing space-based remote sensing track association (fusion of space-based remote sensing and sonar information).
The input of algorithm 1 is the first set of day-based remote sensing detection points { S ] sp (t) } and a set of sonar detection points { S } at corresponding times so (t)}({S so Each sonar detection point in (t) is a sonar track { S } at the sampling moment so The point with the highest probability in (t, alpha) } is output as a set { S } of associated second-day-based remote sensing detection points sp|S (t) }. The basic flow of the algorithm 1 comprises the following steps (1) and (2):
(1) Based on { S ] so (t) } for { S sp (t) } performing hypothesis test based on { S } sp (t) } whether or not the same mathematical distribution is followed, if the test is rejected, continuously decreasing { S } sp The number of detection points in (t) until the same mathematical distribution exists;
(2) Outputting the processed second-day-based remote sensing detection point set { S } sp|S (t)}。
Let X i Representing the target ship trackSpace-based remote sensing test sample, Y i Representing the corresponding sonar detection samples, and making the following assumptions: sonar detection error obeys two-dimensional Gaussian distribution +. >The space-based remote sensing detection error follows a two-dimensional gaussian distribution N (0,0,50,50,0), at which time,
from the deductions of the above formulas (7), (8) and (9)Error variable Z between space-based remote sensing detection and sonar detection i Also in the form of a two-dimensional normal distribution, the following hypothesis testing problem can be constructed:
H 0 :{Z i -has a normal distribution; h 1 :{Z i No normal distribution.
Based on the Henze-zirke test method, the following formula (10) can be obtained:
wherein,p is the dimension of the space-based remote sensing detection sample, D i Is the square Mahalanobis distance (Mahalanobis) between sample i and the center point of all samples.
When the data has a multidimensional normal distribution, the statistic HZ has an approximate log-normal distribution form, the mean μ and variance σ of the distribution 2 As shown in the formula (11) and the formula (12):
wherein a=1+2 β 2 ，w β ＝(1+β 2 )(1+3β 2 ). Thereafter, the mean and variance of the log-normal distribution of the statistic HZ can be defined as the following formulas (13) and (14):
therefore, the significance of a multidimensional normal distribution can be checked by constructing a log-normal distribution parameter as shown in the following equation (15) for the Wald (Wald) statistic (Wald statistic is generally applicable to a constraint for checking nonlinearity, and by estimating the original equation, a test statistic is constructed, which obeys the chi-square distribution under a large sample):
It should be noted that, the updating criteria of the space-based remote sensing detection data are as follows: when the same distribution test is not rejected, the space-based remote sensing detection data with higher probability is selected.
Optionally, the performing, based on the sonar detection point set, hypothesis testing on the first day-based remote sensing detection point set includes:
under the condition that the expected value of the false alarm space-based detection points in the first space-based remote sensing detection point set is smaller than the expected value of the real space-based detection points and the number of the real space-based detection points is larger than a set value, carrying out hypothesis testing on the first space-based remote sensing detection point set based on the sonar detection point set;
the real space-based detection points are space-based remote sensing detection points belonging to the target ship and included in the first space-based remote sensing detection point set, and the false alarm space-based detection points are space-based remote sensing detection points not belonging to the target ship in the first space-based remote sensing detection point set.
Specifically, in the embodiment of the present invention, before performing hypothesis testing on the first set of space-based remote sensing points based on the sonar detection point set, it is required to determine that an expected value of a false alarm space-based detection point in the first set of space-based remote sensing points is smaller than an expected value of a real space-based detection point, and the number of real space-based detection points is greater than a set value, and then performing hypothesis testing on the first set of space-based remote sensing points based on the sonar detection point set, where the real space-based detection point is a space-based remote sensing point belonging to a target ship included in the first set of space-based remote sensing points, and the false alarm space-based detection point is a space-based remote sensing point not belonging to the target ship in the first set of space-based remote sensing points.
Optionally, under the condition that the first day base remote sensing point set is subjected to hypothesis test based on two-dimensional normal distribution based on the sonar detection point set, the set value in the embodiment of the invention is 4, that is, before the first day base remote sensing point set is subjected to hypothesis test based on two-dimensional normal distribution based on the sonar detection point set, it is required to determine that the expected value of the false alarm day base detection point in the first day base remote sensing point set is smaller than the expected value of the real day base detection point, the number of the real day base detection points is greater than 4, and then the first day base remote sensing point set is subjected to hypothesis test based on the sonar detection point set, wherein the real day base detection point is a day base remote sensing point belonging to a target ship and included in the first day base remote sensing point set, and the false alarm day base remote sensing point is a day base remote sensing point not belonging to the target ship.
It should be noted that, since the assumption test based on the two-dimensional normal distribution needs to satisfy the sample size required for the test, but the sample size of 5 or less does not have significance for the test, the number of real space-based detection points needs to be larger than 4.
It can be understood that the embodiment of the invention effectively realizes the fusion of the space-based remote sensing detection data and the sonar detection data, and can improve the space-based remote sensing detection precision of the target ship.
Optionally, the determining the second set of day-based remote sensing detection points based on the first set of day-based remote sensing detection points and the set of sonar detection points further includes:
generating a space-based remote sensing detection track matrix based on each space-based remote sensing detection point in the first space-based remote sensing detection point set at different sampling moments;
determining a target space-based remote sensing detection track section based on the space-based remote sensing detection track matrix, wherein the track similarity between the target space-based remote sensing detection track section and a target sonar detection track section is maximum, and the target sonar detection track section is a sonar detection track section formed between two adjacent sonar detection points in the sonar detection point set;
determining a complete space-based remote sensing detection track of the target ship based on the target space-based remote sensing detection track section;
and determining the second space-based remote sensing detection point set based on the space-based remote sensing detection points included in the complete space-based remote sensing detection track.
Specifically, in the embodiment of the invention, in order to determine the second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, a space-based remote sensing detection track matrix can be generated based on each space-based remote sensing detection point in the first set of space-based remote sensing detection points at different sampling moments, then a target space-based remote sensing detection track section is determined based on the space-based remote sensing detection track matrix, the track similarity between the target space-based remote sensing detection track section and the target sonar detection track section is the sonar detection track section formed between two adjacent sonar detection points in the sonar detection point set, further, based on the target space-based remote sensing track section, the complete space-based remote sensing track of the target ship is determined, and finally the second set of space-based remote sensing detection points are determined based on the space-based remote sensing points included in the complete space-based remote sensing track.
Optionally, the determining the second set of space-based remote sensing detection points based on the space-based remote sensing detection points included in the complete space-based remote sensing detection track includes:
determining a third day-based remote sensing point, a fourth day-based remote sensing point and a fifth day-based remote sensing point which are sequentially adjacent and are included in the complete day-based remote sensing detection track;
determining a first track similarity corresponding to a first day-based remote sensing detection track segment formed between the third day-based remote sensing detection point and the fourth day-based remote sensing detection point, determining a second track similarity corresponding to a second day-based remote sensing detection track segment formed between the fourth day-based remote sensing detection point and the fifth day-based remote sensing detection point, and determining a third track similarity corresponding to a third day-based remote sensing detection track segment formed between the third day-based remote sensing detection point and the fifth day-based remote sensing detection point;
when the first track similarity and the third track similarity are both larger than or equal to a preset track similarity threshold value and the second track similarity is smaller than the preset track similarity threshold value, eliminating the fourth space-based remote sensing detection point from space-based remote sensing detection points included in the complete space-based remote sensing detection track to obtain a new complete space-based remote sensing detection track;
And forming the second space-based remote sensing detection point set by the space-based remote sensing detection points included in the new complete space-based remote sensing detection track.
Specifically, in the embodiment of the invention, in order to determine the second set of space-based remote sensing detection points based on the space-based remote sensing detection points included in the whole space-based remote sensing detection track, a third space-based remote sensing detection point, a fourth space-based remote sensing detection point and a fifth space-based remote sensing detection point which are sequentially adjacent and included in the whole space-based remote sensing detection track can be determined first track similarity corresponding to a first space-based remote sensing detection track segment formed between the third space-based remote sensing detection point and the fourth space-based remote sensing detection point, second track similarity corresponding to a second space-based remote sensing detection track segment formed between the fourth space-based remote sensing detection point and the fifth space-based remote sensing detection point is determined, third track similarity corresponding to a third space-based remote sensing detection track segment formed between the third space-based remote sensing detection point and the fifth space-based remote sensing detection point is determined, then the first track similarity and the third track similarity are both greater than or equal to a preset track similarity threshold, the second track similarity is smaller than the preset track similarity, the second track similarity is calculated by eliminating the second track similarity corresponding to the second track segment formed between the third space-based remote sensing detection point and the fifth space-based remote sensing detection point and the third track is further removed from the whole space-based remote sensing track.
It should be noted that, the fusion of the space-based remote sensing and the sonar data by using the algorithm 1 above requires that at least 5 space-based remote sensing samples are performed, but in some cases, only sparse space-based remote sensing detection results (only 3 or 4 space-based remote sensing detections) in the time domain can be obtained, and it is not desirable to wait for a long space-based remote sensing detection interval (usually more than 10 minutes) due to limitation of revisit time and other factors. In order to solve the problem of fusion of space-based remote sensing and sonar data in this case, the following algorithm 2 may be used to perform fusion of space-based remote sensing and sonar data.
The input of the algorithm 2 is a first day base remote sensing detection point set { S } sp (t) } and corresponding sonar detection set { S } so (t) } output as the set of correlated second-day-based remote sensing detection points { S } sp| S (t). The basic flow of algorithm 2 includes the following steps (1) to (6):
(1) Generating a space-based remote sensing detection track matrix G according to space-based detection targets at different sampling moments;
(2) Calculating the track similarity of each track segment in G, and marking the track segment with the maximum similarity as G 0 ；
(3) Based on g 0 Determining the track with the maximum similarity in the grown complete space-based remote sensing detection tracks, and marking the track as G 0 ；
(4) Circularly detecting and rejecting the space-based remote sensing detection result, repeating the steps (1) to (3), and sorting in descending order according to the track similarity to obtain { G } 0 Let i=1, get G * ＝G 0 (i)；
(5) Inspection G * Whether the similarity of adjacent track segments meets the consistency requirement, if so, according to G * Generate { S ] sp|S (t) }, otherwise, let i++, G * ＝G 0 (i) Repeating step (5);
assume that there is a track similarity vector [ dist 1 ，dist 2 ，dist 3 ]If it meetsThen confirmDetermining that the similarity of the corresponding adjacent track segments does not meet the consistency requirement, and eliminating the corresponding track segments;
(6) Outputting the correlated space-based remote sensing detection point set { S } sp|s (t) } (i.e., the second set of space-based remote sensing detection points).
Wherein, space-based remote sensing detection track matrixn is the number of space-based remote sensing points, k is the number of possible space-based remote sensing detection track combinations, and the non-zero element in G represents the space-based remote sensing points, and the zero element represents the non-space-based remote sensing points.
Track similarity dist corresponding to space-based remote sensing track segment 1 The calculation of the track similarity dist (trace) corresponding to the complete space-based remote sensing detection track is shown in the following formulas (16) and (17):
wherein v is 1 Average speed, theta, of track segment for space-based remote sensing detection 1 For detecting average course angle, v of track section by space-based remote sensing 0 To correspond to the average speed of sonar track section, theta 0 The average course angle of the corresponding sonar track section is set; Δn is the dimension of the processed dataset, and this value reflects the information loss during processing; a, a i ,b i And lambda are weight parameters.
It should be noted that, the data updating criterion of the algorithm 2 is that the track formed by the new space-based remote sensing detection should have a larger track similarity, when the similarity of the newly added track segment is greater than a certain threshold, and when the assumption of the algorithm 1 is established, the association of the space-based remote sensing detection can be performed by using the algorithm 1.
Fig. 5 is a schematic diagram of a space-based detection result before space-based and sonar fusion, and fig. 6 is a schematic diagram of a space-based detection result after space-based and sonar fusion, and it can be known by comparing fig. 5 with fig. 6 that the embodiment of the invention realizes effective elimination of space-based false alarm targets, effectively correlates space-based remote sensing detection with sonar tracks, and can improve detection accuracy of target ships.
Optionally, the optimizing the sonar detection track formed by the sonar detection points in the sonar detection point set based on the second day-based remote sensing detection point set to obtain an optimized sonar detection track includes:
Determining a space-based remote sensing detection track formed by space-based remote sensing detection points in the second space-based remote sensing detection point set;
determining a gravitational field of the space-based remote sensing detection track based on the sonar detection track;
and optimizing the sonar detection track based on the gravitational field to obtain an optimized sonar detection track.
Specifically, in the embodiment of the invention, in order to optimize a sonar detection track formed by sonar detection points in a sonar detection point set based on a second space-based remote sensing detection point set, an optimized sonar detection track is obtained, the space-based remote sensing detection track formed by the space-based remote sensing detection points in the second space-based remote sensing detection point set can be determined first, then, the gravitational field of the space-based remote sensing detection track is determined based on the sonar detection track, and then, the sonar detection track is optimized based on the gravitational field of the space-based remote sensing detection track, so that the optimized sonar detection track is obtained.
It should be noted that, for ship detection in an isolated ship track scene, the embodiment of the invention optimizes the sonar detection track based on the radial base potential field method, and fig. 7 is a schematic diagram of optimizing the sonar track based on the space base detection track and the radial base potential field method, as shown in fig. 7. For a group of space-based remote sensing detection results S sp (t 1 ),S sp (t 2 ),t 1 <t 2 Assume that the ship track is only consistent with the current trackThe time-of-day remote sensing detection tracks are related to other time-of-day remote sensing detection results. Gravitational field U capable of constructing space-based remote sensing detection track att The following formula (18) shows:
wherein d (So (t), sp (t)) is Euclidean distance between sonar detection point and space-based remote sensing detection point, d 0 Is a constant of distance, k att The value of the positive field parameter is determined by the accuracy of the space-based remote sensing detection.
Gravitational field U att The gradient calculation of (2) is shown in the formula (19):
wherein, the space-based remote sensing detection track Sp (t) is a set { S } of space-based remote sensing detection points sp As obtained by spline interpolation of (t) }, it is apparent that the longer the time interval between the space-based remote sensing samples, the lower the confidence of the space-based remote sensing detection track Sp (t) difference. Thus, the confidence of the interpolated track can be described by a radial basis function (radial basis function, RBF), as shown in equation (20):
ψ i (t)＝exp(-h(t-t i ) 2 )，i＝1，2 (20)
wherein h is a variance parameter, and the value of the variance parameter and the space-based remote sensing sampling interval T=t 2 -t 1 Positive correlation, ψ i (t) represents the ith basis function.
Attraction force F att Can be weighted by RBF to obtain gravitational field U att A negative gradient is obtained as shown in formula (21):
wherein sat is a normalized saturation function, and the calculation is shown in formula (22):
Fig. 8 is a schematic diagram of a sonar track optimization result of fusion of a space-based and a sonar provided by the invention, as shown in fig. 8, in which a fusion track 1 is a sonar track obtained by using a maximum likelihood optimization mean value, and a fusion track 2 is a sonar track obtained by fusion optimization of a space-based remote sensing and a sonar provided by the embodiment of the invention.
It can be understood that the embodiment of the invention constructs the information fusion frame by utilizing the complementarity between the sea-based sonar data and the space-based remote sensing data, and realizes the effective detection and tracking of the sea ship target under the isolated track aiming at the characteristics of the multi-source heterogeneous information source, thereby having the advantages of high detection precision, high fusion accuracy, strong tracking capability and the like.
According to the ship target detection method, a first day-based remote sensing detection point set contained in a sonar detection area of a target ship and a sonar detection point set contained in the sonar detection area are firstly determined, then a second day-based remote sensing detection point set is determined based on the first day-based remote sensing detection point set and the sonar detection point set, each day-based remote sensing detection point in the second day-based remote sensing detection point set belongs to the day-based remote sensing detection point of the target ship, and further, a sonar detection track formed by the sonar detection points in the sonar detection point set is optimized based on the second day-based remote sensing detection point set, so that an optimized sonar detection track is obtained, and the optimized sonar detection track is used as the track of the target ship; the method realizes the rapid and accurate detection of the marine ship target by effectively fusing the space-based remote sensing data and the sonar information, and has simple algorithm and easy realization.
The detection device of the ship target provided by the invention is described below, and the detection device of the ship target described below and the detection method of the ship target described above can be correspondingly referred to each other.
Fig. 9 is a schematic structural diagram of a detection device for ship targets, provided by the invention, as shown in fig. 9, the device includes: a first determination module 910, a second determination module 920, and a track optimization module 930; wherein:
the first determining module 910 is configured to determine a first set of day-based remote sensing detection points included in a sonar detection area of a target ship, and a set of sonar detection points included in the sonar detection area;
the second determining module 920 is configured to determine a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the set of sonar detection points, where each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to a space-based remote sensing detection point of the target ship;
the track optimization module 930 is configured to optimize a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second space-based remote sensing detection point set, obtain an optimized sonar detection track, and use the optimized sonar detection track as the track of the target ship.
It should be noted that, the detection device for the ship target provided by the embodiment of the present invention can implement all the method steps implemented by the method embodiment for detecting the ship target, and can achieve the same technical effects, and specific details of the same parts and beneficial effects as those of the method embodiment in the embodiment are not repeated here.
Fig. 10 is a schematic diagram of an entity structure of an electronic device according to the present invention, as shown in fig. 10, the electronic device may include: a processor 1010, a communication interface (Communications Interface) 1020, a memory 1030, and a communication bus 1040, wherein the processor 1010, the communication interface 1020, and the memory 1030 communicate with each other via the communication bus 1040. Processor 1010 may invoke logic instructions in memory 1030 to perform the method of detecting a ship target provided by the methods described above, the method comprising:
the first determining module is used for determining a first day-based remote sensing detection point set contained in a sonar detection area of the target ship and a sonar detection point set contained in the sonar detection area;
the second determining module is configured to determine a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, where each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to a space-based remote sensing detection point of the target ship;
And the track optimization module is used for optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second day-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as the track of the target ship.
Further, the logic instructions in the memory 1030 described above may be implemented in the form of software functional units and stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method of detecting a ship target provided by the methods described above, the method comprising:
the first determining module is used for determining a first day-based remote sensing detection point set contained in a sonar detection area of the target ship and a sonar detection point set contained in the sonar detection area;
the second determining module is configured to determine a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, where each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to a space-based remote sensing detection point of the target ship;
and the track optimization module is used for optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second day-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as the track of the target ship.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the above provided methods of detecting a ship target, the method comprising:
the first determining module is used for determining a first day-based remote sensing detection point set contained in a sonar detection area of the target ship and a sonar detection point set contained in the sonar detection area;
the second determining module is configured to determine a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, where each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to a space-based remote sensing detection point of the target ship;
and the track optimization module is used for optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second day-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as the track of the target ship.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. The ship target detection method is characterized by comprising the following steps of:
determining a first day-based remote sensing detection point set contained in a sonar detection area of a target ship and a sonar detection point set contained in the sonar detection area;
determining a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, wherein each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to the space-based remote sensing detection point of the target ship;
and optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second day-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as the track of the target ship.
2. The method of claim 1, wherein the determining the second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the set of sonar detection points comprises:
based on the sonar detection point set, carrying out hypothesis test on the first day-based remote sensing detection point set, and eliminating the first day-based remote sensing detection point from the first day-based remote sensing detection point set to obtain the second day-based remote sensing detection point set under the condition that a first mathematical distribution obeyed by detection errors of the first day-based remote sensing detection points in the first day-based remote sensing detection point set is different from a second mathematical distribution obeyed by detection errors of the second day-based remote sensing detection points in the first day-based remote sensing detection point set;
The second space-based remote sensing detection points are the rest space-based remote sensing detection points except the first space-based remote sensing detection points in the first space-based remote sensing detection point set, and the number of the second space-based remote sensing detection points is larger than that of the first space-based remote sensing detection points.
3. The method for detecting a ship target according to claim 2, wherein the performing hypothesis testing on the first day-based remote sensing detection point set based on the sonar detection point set comprises:
under the condition that the expected value of the false alarm space-based detection points in the first space-based remote sensing detection point set is smaller than the expected value of the real space-based detection points and the number of the real space-based detection points is larger than a set value, carrying out hypothesis testing on the first space-based remote sensing detection point set based on the sonar detection point set;
the real space-based detection points are space-based remote sensing detection points belonging to the target ship and included in the first space-based remote sensing detection point set, and the false alarm space-based detection points are space-based remote sensing detection points not belonging to the target ship in the first space-based remote sensing detection point set.
4. The method of claim 1, wherein the determining the second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the set of sonar detection points further comprises:
Generating a space-based remote sensing detection track matrix based on each space-based remote sensing detection point in the first space-based remote sensing detection point set at different sampling moments;
determining a target space-based remote sensing detection track section based on the space-based remote sensing detection track matrix, wherein the track similarity between the target space-based remote sensing detection track section and a target sonar detection track section is maximum, and the target sonar detection track section is a sonar detection track section formed between two adjacent sonar detection points in the sonar detection point set;
determining a complete space-based remote sensing detection track of the target ship based on the target space-based remote sensing detection track section;
and determining the second space-based remote sensing detection point set based on the space-based remote sensing detection points included in the complete space-based remote sensing detection track.
5. The method of claim 4, wherein the determining the second set of space-based remote sensing points based on space-based remote sensing points included in the complete space-based remote sensing detection track comprises:
determining a third day-based remote sensing point, a fourth day-based remote sensing point and a fifth day-based remote sensing point which are sequentially adjacent and are included in the complete day-based remote sensing detection track;
Determining a first track similarity corresponding to a first day-based remote sensing detection track segment formed between the third day-based remote sensing detection point and the fourth day-based remote sensing detection point, determining a second track similarity corresponding to a second day-based remote sensing detection track segment formed between the fourth day-based remote sensing detection point and the fifth day-based remote sensing detection point, and determining a third track similarity corresponding to a third day-based remote sensing detection track segment formed between the third day-based remote sensing detection point and the fifth day-based remote sensing detection point;
when the first track similarity and the third track similarity are both larger than or equal to a preset track similarity threshold value and the second track similarity is smaller than the preset track similarity threshold value, eliminating the fourth space-based remote sensing detection point from space-based remote sensing detection points included in the complete space-based remote sensing detection track to obtain a new complete space-based remote sensing detection track;
and forming the second space-based remote sensing detection point set by the space-based remote sensing detection points included in the new complete space-based remote sensing detection track.
6. The method for detecting a ship target according to claim 1, wherein optimizing a sonar detection track formed by sonar detection points in the sonar detection point set based on the second space-based remote sensing detection point set to obtain an optimized sonar detection track comprises:
Determining a space-based remote sensing detection track formed by space-based remote sensing detection points in the second space-based remote sensing detection point set;
determining a gravitational field of the space-based remote sensing detection track based on the sonar detection track;
and optimizing the sonar detection track based on the gravitational field to obtain an optimized sonar detection track.
7. The method of claim 1-6, wherein a confidence level of each sonar detection point in the sonar detection area of the target ship is a preset value.
8. A ship target detection device, comprising:
the first determining module is used for determining a first day-based remote sensing detection point set contained in a sonar detection area of the target ship and a sonar detection point set contained in the sonar detection area;
the second determining module is configured to determine a second set of space-based remote sensing detection points based on the first set of space-based remote sensing detection points and the sonar detection point set, where each space-based remote sensing detection point in the second set of space-based remote sensing detection points belongs to a space-based remote sensing detection point of the target ship;
And the track optimization module is used for optimizing a sonar detection track formed by the sonar detection points in the sonar detection point set based on the second day-based remote sensing detection point set to obtain an optimized sonar detection track, and taking the optimized sonar detection track as the track of the target ship.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of detecting a ship target according to any one of claims 1 to 7 when the program is executed by the processor.
10. A non-transitory computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements a method of detecting a ship target according to any one of claims 1 to 7.
CN202310869027.3A 2023-07-14 2023-07-14 Ship target detection method and device, electronic equipment and storage medium Pending CN117111072A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310869027.3A CN117111072A (en) 2023-07-14 2023-07-14 Ship target detection method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310869027.3A CN117111072A (en) 2023-07-14 2023-07-14 Ship target detection method and device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN117111072A true CN117111072A (en) 2023-11-24

Family

ID=88801067

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310869027.3A Pending CN117111072A (en) 2023-07-14 2023-07-14 Ship target detection method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN117111072A (en)

Similar Documents

Publication Publication Date Title
Ao et al. Detection and discrimination of ship targets in complex background from spaceborne ALOS-2 SAR images
CN110598730A (en) Flight path association algorithm based on decision tree
Wang et al. Ship detection based on fused features and rebuilt YOLOv3 networks in optical remote-sensing images
KR101817011B1 (en) Clutter reduction Method and Apparatus for active sonar based on clustering characteristics
Heiselberg et al. Ship velocity estimation in SAR images using multitask deep learning
Fowdur et al. Tracking targets with known spatial extent using experimental marine radar data
Sun et al. Ship detection in SAR images based on multi-level superpixel segmentation and fuzzy fusion
Jiang et al. Ship detection in RADARSAT SAR imagery
CN111707999B (en) Sea surface floating small target detection method based on combination of multiple features and ensemble learning
Wu et al. Priori information-based feature extraction method for small target detection in sea clutter
CN117115436A (en) Ship attitude detection method and device, electronic equipment and storage medium
CN117111072A (en) Ship target detection method and device, electronic equipment and storage medium
Villar et al. Efficient approach for OS‐CFAR 2D technique using distributive histograms and breakdown point optimal concept applied to acoustic images
CN116699662A (en) GNSS-R-based marine target positioning method and system
Zhang et al. FRS-Net: An efficient ship detection network for thin-cloud and FOG-covered high-resolution optical satellite imagery
Hadzagic et al. A Bayesian inference approach for batch trajectory estimation
Liu et al. A nonuniform clutter intensity estimation algorithm for random finite set filters
Chabbi et al. CFAR-CNN detector of ships from SAR image using generalized gamma distribution and real dataset
EP3647994B1 (en) Automated generation of training images
Deng et al. Ship target detection based on CFAR and deep learning SAR image
Xie et al. A Convolution Neural Network-based Method for Sea Ice Remote Sensing using GNSS-R Data
Wu Two-parameter CFAR ship detection algorithm based on Rayleigh distribution in SAR images
Yan et al. An efficient plot fusion method for high resolution radar based on contour tracking algorithm
Zhang et al. SAR image ship detection based on a superpixel non-window fast cfar strategy
CN113009470B (en) Processing method, system, device and medium for target situation characteristic data

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