CN116106850A - Method for automatically identifying oil stain stealing and removing of ship by combining radar satellite image and AIS - Google Patents
Method for automatically identifying oil stain stealing and removing of ship by combining radar satellite image and AIS Download PDFInfo
- Publication number
- CN116106850A CN116106850A CN202310340565.3A CN202310340565A CN116106850A CN 116106850 A CN116106850 A CN 116106850A CN 202310340565 A CN202310340565 A CN 202310340565A CN 116106850 A CN116106850 A CN 116106850A
- Authority
- CN
- China
- Prior art keywords
- ship
- ais
- oil stain
- pixel
- oil
- 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.)
- Granted
Links
Images
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
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N22/00—Investigating or analysing materials by the use of microwaves or radio waves, i.e. electromagnetic waves with a wavelength of one millimetre or more
-
- 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
- G01S13/86—Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/20—Controlling water pollution; Waste water treatment
- Y02A20/204—Keeping clear the surface of open water from oil spills
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Electromagnetism (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Chemical & Material Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Biochemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Artificial Intelligence (AREA)
- Astronomy & Astrophysics (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Radar Systems Or Details Thereof (AREA)
- Image Processing (AREA)
Abstract
The invention provides a method for automatically identifying oil stealing and removing oil stains of a ship by combining radar satellite images and AIS, which comprises the following steps: acquiring oil stain and oil stain-like pixels in a radar satellite image; linearly identifying the greasy dirt and the greasy dirt-like pixels to obtain linear greasy dirt bands, wherein the linear greasy dirt bands are distributed in a manner of being narrow at one end and wide at the other end; and identifying a source ship for stealing oil and dirt based on the linear oil stain belt, and tracing the source ship. According to the method, the marine monitoring capability of the oil stain stealing behavior of the ship covering the middle and high sea water area can be established based on the radar satellite and the ship AIS, and the marine law enforcement force is guided to conduct targeted accurate inspection at the ship destination port.
Description
Technical Field
The invention relates to the technical field of recognition of oil stealing and discharging behaviors of ships, in particular to a method for automatically recognizing oil stealing and discharging behaviors of ships by combining radar satellite images and AIS.
Background
The compliance of ship oily water goes to include emission up to standard after the ship is handled and on shore recovery treatment, because the processing cost is higher, leads to the steal oil dirt phenomenon of ship during the offshore navigation often to be continuous. The method can strengthen supervision in a mode that an airplane or a sea patrol boat regularly cruises in a relatively closed or offshore sea area. For open middle and far sea water areas, the ship density is low, the offshore is far, and the cost of the monitoring mode is too high, so that the monitoring of illegal actions such as oil and dirt stealing actions of the ship in the middle and far sea areas is always a difficult point. Even because the middle and open sea water area is far away from the coastal zone with dense population, the environmental pollution behavior cannot be widely focused, so that the environment pollution behavior becomes a blind point for ship pollution prevention supervision. The radar satellite has the advantages of being free from interference of cloud layers and capable of identifying oil stains on water, and is widely applied to tracking and monitoring of marine ship accident oil spill and offshore oil platform oil spill. The radar wave is reflected strongly by the ship, and the satellite, the airplane or the ship radar is widely applied to the active identification of the marine ship, complements the advantages of the AIS passive identification system, and becomes one of the means for monitoring the marine ship. However, neither radar satellite images nor AIS are applied to monitoring of oil and dirt stealing behaviors of ships. The method is an essential environmental protection measure for preventing the ship pollution in the middle and high sea water areas, and has important significance in monitoring the oil and dirt stealing behaviors of the ship in the middle and high sea water areas. Therefore, it is needed to propose a method for automatically identifying the oil and dirt stealing behavior of a ship by combining a radar satellite image and an AIS, analyze the characteristics of the oil and dirt stealing behavior of the ship in the radar satellite image, and automatically identify the oil and dirt stealing behavior of the ship by combining the radar satellite image and the AIS, which is technically feasible.
Disclosure of Invention
In order to solve the technical problems, in order to establish the marine supervision capability of the behavior of oil stain in the open sea water area covered by the ship, and guide the marine law enforcement forces to carry out targeted accurate inspection at the ship destination port, the invention provides a method for automatically identifying the oil stain in the ship by combining radar satellite images and AIS, which comprises the following steps: oil stain and oil stain-like pixel image recognition algorithm, linear oil stain band image recognition algorithm, source ship recognition algorithm and source ship identity tracing method. According to the method, the characteristics that the oil stain zone in the radar satellite image is specific to the ship radar reflection signal and the tight spatial correlation exists between the oil stain stealing and discharging ship and the oil stain zone are utilized, a method for automatically identifying the oil stain stealing and discharging behavior of the ship is provided, and the identity information of the ship is traced by combining with the ship AIS data. According to the method, the marine monitoring capability of the oil stain stealing behavior of the ship covering the middle and high sea water area can be established based on the radar satellite and the ship AIS, and the marine law enforcement force is guided to conduct targeted accurate inspection at the ship destination port.
In order to achieve the above purpose, the invention provides a method for automatically identifying oil theft and grease removal of a ship by combining radar satellite images and AIS, which comprises the following steps:
acquiring oil stain and oil stain-like pixels in a radar satellite image;
linearly identifying the greasy dirt and the greasy dirt-like pixels to obtain linear greasy dirt bands, wherein the linear greasy dirt bands are distributed in a manner of being narrow at one end and wide at the other end;
and identifying a source ship for stealing oil and dirt based on the linear oil stain belt, and tracing the source ship.
Optionally, acquiring the greasy dirt and greasy dirt-like pixels in the radar satellite image includes:
acquiring an oil stain index when the neighborhood width of each pixel in the radar satellite image is z;
the oil stain index is matched with a preset specific threshold value i threshold And judging that the pixel is smaller than the oil stain pixel and the oil stain-like pixel if the pixel is smaller than the oil stain pixel.
Optionally, acquiring the oil stain index when the neighborhood width of each pixel in the radar satellite image is z includes:
acquiring an average value of n values of the inner neighborhood pixels, standard deviation of n values of the inner neighborhood pixels and an average value of n values of the outer neighborhood pixels when the neighborhood width is z;
acquiring an oil stain index of the pixel based on the average value of the n values of the inner neighborhood pixels, the standard deviation of the n values of the inner neighborhood pixels and the average value of the n values of the outer neighborhood pixels;
gradually increasing z from 1 to a constant, repeatedly calculating the oil stain index, and finally, only keeping the oil stain index with the smallest positive number as the oil stain index of the pixel.
Optionally, the oil stain index is:
i x,y,z =sd x,y,z /(ao x,y,z -ai x,y,z )
wherein i is x,y,z Ai is the oil stain index when the neighborhood width is z x,y,z Is the average value of n values of pixels in the inner neighborhood, sd x,y,z Is the standard deviation of n values of the pixels in the inner neighborhood, ao x,y,z And x, y and z are respectively the row number, the column number and the neighborhood width of the pixel in the image for the average value of the n values of the pixels in the outer neighborhood.
Optionally, the linear recognition process of the greasy dirt and greasy dirt-like pixels adopts Hough transformation.
Optionally, identifying source vessel PSS for oil theft removal x,y Comprising the following steps:
acquiring ship pixel PS in radar satellite image x,y ；
Obtaining the linear greasy dirt belt L k , b Narrow Head pixel po_head of (a) x,y ；
And acquiring the distance and the slope between the ship pixel and the narrow-head pixel, and identifying the source ship.
Optionally, acquiring the narrow-head pixel of the linear greasy dirt belt comprises:
comparing the z values of the oil stain indexes on the oil stain and oil stain-like pixels, wherein the smaller the z value is, the closer the oil stain and oil stain-like pixels are to the narrow head of the linear oil stain band, and the larger the z value is, the closer the oil stain and oil stain-like pixels are to the wide head of the linear oil stain band;
and calculating the x average value x_zmin and the y average value y_zmin of the greasy dirt and greasy dirt like pixels with the minimum z value, and calculating the x average value x_zmax and the y average value y_zmax of the greasy dirt and greasy dirt like pixels with the maximum z value. If x_zmin is less than x_zmax, x of the narrow-head pixel is the minimum value of x in the greasy dirt and greasy dirt-like pixel with the minimum z value, otherwise, x is the maximum value; if y_zmin is less than y_zmax, y of the narrow-head pixel is the minimum y value of the greasy dirt and the greasy dirt-like pixel with the minimum z value, otherwise, the y is the maximum value.
Optionally, acquiring the distance and the slope between the ship pixel and the narrow-head pixel includes:
when the distance is smaller than a preset specific threshold dso threshold And the absolute value of the difference between the slope and the included angle between the straight line represented by k and the horizontal direction is smaller than a preset specific threshold k bias The ship is not far from the oil stain belt and on the extension line of the linear oil stain belt, the ship pixel is judged to be the source ship of the linear oil stain belt, otherwise, the ship pixel is not; where k is the slope of the straight line identified by the greasy dirt pixel using hough variation.
Optionally, tracing the source vessel includes:
according to the shooting time t of the radar satellite image, according to the coordinate system information of the radar satellite image, converting the line number positions x and y of the source ship in the image into longitude and latitude x s , y s ；
Calling specific time length, [ x ] before and after t time from ship AIS database s , y s ]All ship AIS data [ t ] within a specific distance range around a location ais , x ais , y ais , v ais ]The method comprises the steps of carrying out a first treatment on the surface of the Wherein t is ais , x ais , y ais , v ais The AIS data recording time, the ship longitude position, the ship latitude position and the ship navigational speed are respectively;
for each AIS data, calculate |v ais ×(t-t ais ) 2 -(x s -x ais ) 2 -(y s -y ais ) 2 And the AIS data with the smallest value is AIS data of the source ship, and unique identity information is extracted from the AIS data.
Compared with the prior art, the invention has the following advantages and technical effects:
according to the method, the characteristics that the oil stain zone in the radar satellite image is specific to the ship radar reflection signal and the tight spatial correlation exists between the oil stain stealing and discharging ship and the oil stain zone are utilized, a method for automatically identifying the oil stain stealing and discharging behavior of the ship is provided, and the identity information of the ship is traced by combining with the ship AIS data. According to the method, the marine monitoring capability of the oil stain stealing behavior of the ship covering the middle and high sea water area can be established based on the radar satellite and the ship AIS, and the marine law enforcement force is guided to conduct targeted accurate inspection at the ship destination port.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application, illustrate and explain the application and are not to be construed as limiting the application. In the drawings:
FIG. 1 is a schematic flow chart of a method for automatically identifying oil theft and oil pollution of a ship by combining radar satellite images and AIS according to an embodiment of the invention;
FIG. 2 is a schematic diagram of the result of recognizing greasy dirt and greasy dirt-like pixels under different neighborhood widths z according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a source vessel identified in accordance with an embodiment of the present invention;
FIG. 4 is a diagram showing an embodiment of the invention for automatically identifying a ship's oil theft and oil fouling behavior in combination with radar satellite images and AIS;
wherein 201, a ship; 202. greasy dirt-like dark pixels in the region are automatically identified as PO when z is 1 or 2 x,y The method comprises the steps of carrying out a first treatment on the surface of the 203. The most central pixel of the nearly circular greasy dirt like dark pixels in the area has the largest greasy dirt index when z is 5 or 6 and is automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the 204. Dark pixels in the region have the greatest greasy dirt index when z is 1 and are automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the 205. Dark pixels in the region have the greatest greasy dirt index at z of 2 and are automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the 206. Dark pixels in the region have the greatest greasy dirt index at z of 3 and are automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the 207. Dark pixels in the region have the greatest greasy dirt index at z of 4 and are automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the 208. Dark pixels within the area are not automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the 301. Linear greasy dirt band pixels; 302. line type greasy dirt belt central axis L k , b ；303、L k , b In the narrow-Head pixel po_head x,y The method comprises the steps of carrying out a first treatment on the surface of the 304. And PO_head x,y The distance and the slope of the connecting line meet the requirements of the source ship; 305. and PO_head x,y A non-source ship with a connecting line with a distance which does not meet the requirements and a slope which meets the requirements; 306. and PO_head x,y A non-source ship with a connecting line with a distance meeting the requirements and a slope not meeting the requirements; 307. and PO_head x,y The distance and the slope of the connecting line are not in accordance with the non-source ship; 401. a source vessel; 402. a linear greasy dirt belt; 403. and AIS data of nearby ships.
Detailed Description
It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
The invention provides a method for automatically identifying oil stealing and removing oil stains of a ship by combining radar satellite images and AIS, which comprises the following steps:
acquiring oil stain and oil stain-like pixels in a radar satellite image;
linearly identifying the greasy dirt and greasy dirt-like pixels to obtain linear greasy dirt bands which are distributed in a manner of being narrow at one end and wide at the other end;
based on the linear greasy dirt belt, identify the source ship of steal oil extraction dirt to trace to the source ship.
Further, acquiring the greasy dirt and greasy dirt-like pixels in the radar satellite image comprises:
acquiring an oil stain index when the neighborhood width of each pixel in the radar satellite image is z;
the oil stain index is matched with a preset specific threshold value i threshold And judging that the pixel is less than the oil stain pixel and the oil stain pixel if the pixel is judged to be the oil stain pixel and the oil stain pixel.
Further, acquiring the oil stain index when the neighborhood width of each pixel in the radar satellite image is z comprises:
acquiring an average value of n values of the inner neighborhood pixels, standard deviation of n values of the inner neighborhood pixels and an average value of n values of the outer neighborhood pixels when the neighborhood width is z;
acquiring an oil stain index of the pixel based on the average value of the n values of the inner neighborhood pixels, the standard deviation of the n values of the inner neighborhood pixels and the average value of the n values of the outer neighborhood pixels;
and gradually increasing z from 1 to a constant, repeatedly calculating the oil stain index, and finally, only keeping the oil stain index with the minimum positive number as the oil stain index of the pixel.
Further, the method for linearly identifying the greasy dirt and the greasy dirt-like pixels comprises the following steps: a Hough transform is used.
Further, identifying source ship PSS for oil theft and oil removal x,y Comprising the following steps:
acquiring ship pixel PS in radar satellite image x,y ；
Acquiring linear greasy dirt belt L k , b Narrow Head pixel po_head of (a) x,y ；
And acquiring the distance and the slope between the ship pixel and the narrow-head pixel, and identifying the source ship.
Further, acquiring the narrow-head pixels of the linear greasy dirt belt comprises:
comparing z values of all oil indexes on the oil stain and oil stain like pixels, wherein the smaller the z value is, the closer the oil stain and oil stain like pixels are to the narrow head of the linear oil stain band, and the larger the z value is, the closer the oil stain and oil stain like pixels are to the wide head of the linear oil stain band;
and calculating the x average value x_zmin and the y average value y_zmin of the greasy dirt and greasy dirt like pixels with the minimum z value, and calculating the x average value x_zmax and the y average value y_zmax of the greasy dirt and greasy dirt like pixels with the maximum z value. If x_zmin is less than x_zmax, x of the narrow-head pixel is the minimum value of x in the greasy dirt and greasy dirt-like pixel with the minimum z value, otherwise, x is the maximum value; if y_zmin is less than y_zmax, y of the narrow-head pixel is the minimum y value of the greasy dirt and the greasy dirt-like pixel with the minimum z value, otherwise, the y is the maximum value.
Further, acquiring the distance and the slope between the ship pixel and the narrow-head pixel comprises:
when the distance is smaller than the preset distanceSet a specific threshold dso threshold And the absolute value of the difference between the slope and the included angle between the straight line represented by k and the horizontal direction is smaller than a preset specific threshold k bias And the ship is not far from the oil stain belt and is judged to be the source ship of the linear oil stain belt on the extension line of the linear oil stain belt, otherwise, the ship is not judged to be the source ship of the linear oil stain belt.
Further, tracing the source vessel includes:
according to the shooting time t of the radar satellite image, according to the coordinate system information of the radar satellite image, converting the line number positions x and y of the source ship in the image into longitude and latitude x s , y s ；
Calling specific time length, [ x ] before and after t time from ship AIS database s , y s ]All ship AIS data [ t ] within a specific distance range around a location ais , x ais , y ais , v ais ]The method comprises the steps of carrying out a first treatment on the surface of the Wherein t is ais , x ais , y ais , v ais The AIS data recording time, the ship longitude position, the ship latitude position and the ship navigational speed are respectively;
for each AIS data, calculate |v ais ×(t-t ais ) 2 -(x s -x ais ) 2 -(y s -y ais ) 2 And the AIS data with the smallest value is AIS data of the source ship, and unique identity information is extracted from the AIS data.
As shown in fig. 1, the method for automatically identifying the oil theft and the oil removal of the ship by combining the radar satellite image and the AIS provided by the embodiment specifically comprises the following steps:
the description above: the radar satellite image is l pixels long and w pixels wide. Any one pixel is named P x,y The brightness value of the pixel is n x,y X is an integer between 1 and l, y is an integer between 1 and w, and n is a positive integer. P (P) x,y The neighborhood width of (a) is z pixels, z is a positive integer, then the neighboring pixels are P [(x-z)~(x+z)],[(y-z)~(y+z)] Comprises P x,y Self-sum (2 Xz+1) 2 A pixel, and the neighboring pixels are P [ (x-2 z)/(x-z) and (x+z)/(x+2 z)][ (y-2 z)/(y-z) and (y+z)/(y+2 z)] Together (4 Xz + 1) 2 -(2×z+1) 2 And each pixel.
The following is noted: the land area in the radar satellite image is not considered, and only the water area is considered. The elimination method is based on geographic information systems such as a world map, and pixels located on the land area of the map are eliminated.
3 typical characteristics of oil and dirt stealing behavior of ships: the radar reflection signal of the greasy dirt is weaker than the water body, so that the n value of the greasy dirt band pixel is smaller than the n value of the surrounding water body pixel, and naturally, the situation that the radar reflection signal is weaker than the water body can be generated due to some natural phenomena on the sea, which is generally called greasy dirt; the ship is sailed and discharged, so that the greasy dirt with pixels are arranged linearly, and the closer the greasy dirt with pixels is to the ship, the narrower the greasy dirt with pixels is, and the farther the greasy dirt with pixels is to the ship, the wider the greasy dirt with pixels is; the radar reflection signal of the ship is extremely strong, so that pixels with extremely large n values are necessarily arranged near the narrow end of the oil stain band pixels in a linear arrangement.
S1, identifying greasy dirt and greasy dirt-like pixel PO x,y
For each pixel, an average ai of n values of pixels in the inner neighborhood when the neighborhood width is z is calculated x,y,z Standard deviation sd of n values of pixels in the inner neighborhood x,y,z Average value ao of n values of pixels in outer neighborhood x,y,z Calculating oil stain index i when the neighborhood width is z x,y,z =sd x,y,z /(ao x,y,z -ai x,y,z ). z is gradually increased from 1 to a constant, and i is repeatedly calculated x,y,z Finally, only the i of the least positive number is reserved x,y,z As an oil stain index for the pixel; e.g. i x,y,z All negative numbers, the pixel is not likely to be an oil and oil-like pixel. The constant is related to the resolution of the image, the higher the resolution, the greater the number of pixels in the oil stain band, and the greater the constant should be. (Note: use of sd above) x,y,z Taking into account noise of the image, taking ao into account x,y,z -ai x,y,z Is to highlight the difference between greasy and greasy-like pixels and the surrounding background
For each pixel, judging the oil stain index i x,y Whether or not it is smaller than a specific threshold i threshold If yes, the pixel PO is oil stain and oil stain like pixel x,y Otherwise, other pixels. i.e threshold Depending on the image, it is suggested that the patient is not fixed0.1 to 0.3. In general, the greasy dirt belt presents a line shape with a narrow end and a wide end, PO x,y Essentially consisting of pixels with a central axis of oil stain.
The result of the identified greasy dirt and greasy dirt-like pixel is shown in fig. 2, and in fig. 2: the bright pixels in the rectangle 201 are the ship; the greasy dirt-like dark pixels in the 202 area are automatically identified as PO when z is 1 or 2 x,y The method comprises the steps of carrying out a first treatment on the surface of the The most central pixel of the nearly circular greasy dirt-like dark pixels in 203 area has the largest greasy dirt index when z is 5 or 6 and is automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the Dark pixels in region 204 have the greatest greasy dirt index at z of 1 and are automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the Dark pixels in region 205 have the greatest greasy dirt index at z of 2 and are automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the Dark pixels in region 206 have the greatest greasy dirt index at z of 3 and are automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the Dark pixels in region 207 have the greatest greasy dirt index at z of 4 and are automatically identified as PO x,y The method comprises the steps of carrying out a first treatment on the surface of the Dark pixels in region 208 are not automatically identified as PO x,y 。
S2, identifying linear greasy dirt belt L k , b
For all POs in the image x,y Linear recognition can be performed by adopting methods such as Hough transformation and the like which are mature in the field of image processing. The calculation process of the Hough transformation method can be briefly summarized as that a straight line L with a slope of k and an intercept of b is drawn on a radar satellite image k , b Both k and b take values within a specific range. When a PO x,y Fall to L k , b When the number is up, give L k , b Adding 1 min, and finally obtaining L with higher score k , b Representing more PO x,y Fall to L k , b On, the score is greater than a specific threshold number hough L of (2) k , b As an oil stain band. number of number hough Depending on the resolution of the image, the higher the resolution, the greater the threshold should be.
S3, identifying source ship PSS x,y
S3.1 identifying ship pixel PS x,y . The value of n is greater than a specific threshold value n threshold Is defined as ship pixel PS x,y 。n threshold The greater the difference in signal strength, the greater the value may be, depending on the radar signal strength. When the resolution of radar satellite images is high, a ship is often composed of a group of adjacent PS x,y Composition is prepared.
S3.2 identifying linear greasy dirt belt L k , b Narrow Head pixel po_head of (a) x,y . Pair-type greasy dirt belt L k , b All PO on x,y I of (2) x,y,z The smaller the z value is, the smaller the PO is x,y The closer to the narrow head of the linear greasy dirt belt, the larger the z value is, the more PO is x,y The closer to the wide head of the linear greasy dirt belt. Calculation of PO with minimum z value x,y X-zmin and y-zmin, and calculating the PO with the largest z value x,y X_zmax and y average y_zmax. If x_zmin<x_zmax, then the narrow-Head pixel PO_head x,y X is PO with minimum z value x,y If not, taking x maximum value; if y_zmin<y_zmax, then the narrow-Head pixel PO_head x,y Y is PO with minimum z value x,y And if not, taking the y maximum value.
S3.3 determining a ship pixel PS x,y Whether or not it is a linear greasy dirt belt L k , b Is a source vessel of (a). For each PS x,y Connection PS x,y And PO_Head x,y The distance dso and slope kso between two pixels are calculated. When dso is less than a certain threshold dso threshold And the absolute value of the difference between the included angles of the straight lines represented by kso and k and the horizontal direction is smaller than a specific threshold k bias The PS is determined by indicating that the ship is not far from the oil stain belt and on the extension line of the linear oil stain belt x,y Is a linear greasy dirt belt L k , b Source vessel PSS of (a) x,y Otherwise, not. dso threshold Depending on whether the oil stain band can remain linear after the duration of the ship sailing the distance has elapsed, it is recommended to be 10-50 km. k (k) bias The death is not fixed, and 5-10 degrees are recommended.
The identified source vessel is shown in fig. 3, wherein 301 line-type greasy dirt belt pixels; 302 automatic identification linear greasy dirt belt central axis L k , b ；303L k , b In the narrow-Head pixel po_head x,y The method comprises the steps of carrying out a first treatment on the surface of the 304 and PO_Head x,y The distance and the slope of the connecting line meet the requirements of the source ship; 305 and PO_Head x,y A non-source ship with a connecting line with a distance which does not meet the requirements and a slope which meets the requirements; 306 and PO_head x,y A non-source ship with a connecting line with a distance meeting the requirements and a slope not meeting the requirements; 307 and PO_Head x,y And the distance and the slope of the connecting line are not in accordance with the non-source ship.
S4, tracing source ship AIS identity information
According to the shooting time t of the radar satellite image, according to the coordinate system information of the radar satellite image, the PSS is processed x,y The line number positions x and y in the image are converted into longitude and latitude x s , y s . Calling specific time length, [ x ] before and after t time from ship AIS database s , y s ]All ship AIS data [ t ] within a specific distance range around a location ais , x ais , y ais , v ais ]The duration and distance range can be empirically set with the objective of reducing the traceability range by prescreening. t is t ais , x ais , y ais , v ais The AIS data record time, ship longitude position, ship latitude position and ship navigational speed are respectively. For each AIS data, calculate |v ais ×(t-t ais ) 2 -(x s -x ais ) 2 -(y s -y ais ) 2 And the AIS data with the smallest value is AIS data of the source ship, and unique identity information such as ship name, MMSI number and the like is extracted from the AIS data.
The implementation of the invention is further illustrated by taking a Sentinel 1 radar satellite image of the sea area near Indonesia Peureuloak as an example;
the radar satellite image used: on 15 days 2 and 2019, the resolution of the seninel 1 radar satellite image (e.g., fig. 4, wherein the vessel AIS data is in the vicinity of the source vessel 402 line type greasy dirt band 403, 401) in the sea area around indonesia peureluak between the maxajia strais and the manglauca bay is 5m.
S1, identifying greasy dirt and greasy dirt-like pixel PO x,y
The width z of the neighborhood is 1-10, the rootThe recognized oil stain band width is considered to be 15 to 105m wide in terms of image resolution. i.e threshold Set to 0.1. Through identification, the number of oil-like pixels on the sea surface is small in the imaging time range of the image, and most of the oil-like pixels are oil-like pixels. PO was identified when z was 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10 x,y 532, 486, 612, 544, 567, 488, 498, 413, 115, 12 pixels, respectively. When the satellite shoots an oil stain band, the ship is stealing and discharging the oil stain, the oil stain band presents a linear characteristic of being narrow at one end and wide at the other end, and the diffusion width of the oil stain band at the first discharge position is approximately 85m wide.
S2, identifying linear greasy dirt belt L k , b
For all POs in the image x,y And (3) taking the lower left corner in the image as an origin, and adopting a Hough transformation method to perform linear recognition. Discovery of PO after identification x,y Mainly concentrate on the straight line with 128 degrees included angle of horizontal line (corresponding to k value of-1.26) and intercept of 4310-4320 pixels, and total 3593 pixels, which is far larger than other straight lines with slope and intercept. Thus, the linear greasy dirt belt L is identified k , b The k value of-1.26, and the b value of 4315.
S3, identifying source ship PSS x,y
S3.1 setting a threshold value n of the ship pixel threshold 200, the value of n of the background pixel and the oil stain pixel is far smaller than the value, and the ship pixel PS x,y Is approximately 132 in number and is basically in a gathered shape, and is mainly positioned at positions of 918-932 in x and 3160-3172 in y, so that 1 ship can be effectively identified.
S3.2 is obvious, line type greasy dirt belt L k , b PO with upper z value of 1 x,y PO with z value of 10 is upward and leftward x,y Downward and rightward. PO with z value of 1 is counted x,y The x minimum value of (3) is 1013 and the y maximum value is 3101. Linear greasy dirt belt L k , b Narrow Head pixel po_head of (a) x,y X and y of (3) are 1013 and 3101, respectively.
S3.3 for each PS x,y Connection PS x,y And PO_Head x,y The distance dso and slope kso between two pixels are calculated. Setting dso threshold 10km, k bias Is 10 deg.. Calculated, dso is about 98-120 m, much less than 10km. Calculated, the slope kso is about-1.38 to-1.32 (corresponding to an included angle of about 126 to 127 degrees with the horizontal line), and L k , b The angular deviation of (2) is about 1 to 2 degrees, much smaller than 10 degrees. Determining that the ship is oil stain zone L k , b Source vessel PSS of (a) x,y 。
S4, tracing source ship AIS identity information
According to the coordinate system information of the radar satellite image, PSS is processed x,y Centers 925, 3166 of row and column positions x (918-932) and y (3160-3172) in the image are converted into longitude and latitude x s , y s About 4 deg. 52.393',97 deg. 57.813'. AIS data within 10km about the longitude and latitude for the first 30min are called from a ship AIS database, and a total of 2 AIS data (little because of far offshore, AIS data of marine satellites can be received only, and the revisit period is long) are respectively [ 29min before, 4 DEG 48.291',98 DEG 0.975',5.2m/s]And [ 18min before, 4℃ 49.903',97℃ 59.708',5.3m/s ]]. Extracting ship name information from AIS data, wherein the AIS data are recorded by the same ship at different moments, and the ship name is Perkasa, so that the calculation of |v is omitted ais ×(t-t ais ) 2 -(x s -x ais ) 2 -(y s -y ais ) 2 And (3) confirming the ship identity information.
The foregoing is merely a preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions easily conceivable by those skilled in the art within the technical scope of the present application should be covered in the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (9)
1. The method for automatically identifying the oil stain stealing and removing of the ship by combining the radar satellite image and the AIS is characterized by comprising the following steps of:
acquiring oil stain and oil stain-like pixels in a radar satellite image;
linearly identifying the greasy dirt and the greasy dirt-like pixels to obtain linear greasy dirt bands, wherein the linear greasy dirt bands are distributed in a manner of being narrow at one end and wide at the other end;
and identifying a source ship for stealing oil and dirt based on the linear oil stain belt, and tracing the source ship.
2. The method for automatically identifying oil theft and oil removal of a ship by combining a radar satellite image and an AIS according to claim 1, wherein acquiring oil and oil-like pixels in the radar satellite image comprises:
acquiring an oil stain index when the neighborhood width of each pixel in the radar satellite image is z;
the oil stain index is matched with a preset specific threshold value i threshold And judging that the pixel is smaller than the oil stain pixel and the oil stain-like pixel if the pixel is smaller than the oil stain pixel.
3. The method for automatically identifying oil theft and grease removal of a ship by combining a radar satellite image and an AIS according to claim 2, wherein obtaining an oil stain index when a neighborhood width of each pixel in the radar satellite image is z comprises:
acquiring an average value of n values of the inner neighborhood pixels, standard deviation of n values of the inner neighborhood pixels and an average value of n values of the outer neighborhood pixels when the neighborhood width is z;
acquiring an oil stain index of the pixel based on the average value of the n values of the inner neighborhood pixels, the standard deviation of the n values of the inner neighborhood pixels and the average value of the n values of the outer neighborhood pixels;
gradually increasing z from 1 to a constant, repeatedly calculating the oil stain index, and finally, only keeping the oil stain index with the smallest positive number as the oil stain index of the pixel.
4. The method for automatically identifying oil theft and removal of a ship by combining radar satellite images and AIS according to claim 2, wherein the oil contamination index is:
i x,y,z =sd x,y,z /(ao x,y,z -ai x,y,z )
wherein i is x,y,z For a neighborhood width zOil stain index ai x,y,z Is the average value of n values of pixels in the inner neighborhood, sd x,y,z Is the standard deviation of n values of the pixels in the inner neighborhood, ao x,y,z And x, y and z are respectively the row number, the column number and the neighborhood width of the pixel in the image for the average value of the n values of the pixels in the outer neighborhood.
5. The method for automatically identifying oil stain stealing and removing by combining radar satellite images and AIS according to claim 1, wherein the linear identification process of oil stains and oil stain-like pixels adopts Hough transformation.
6. The method for automatically identifying a ship from which oil is stolen and discharged by combining radar satellite images and AIS according to claim 4, wherein the source ship PSS from which oil is stolen and discharged is identified x,y Comprising the following steps:
acquiring ship pixel PS in radar satellite image x,y ；
Obtaining the linear greasy dirt belt L k , b Narrow Head pixel po_head of (a) x,y ；
And acquiring the distance and the slope between the ship pixel and the narrow-head pixel, and identifying the source ship.
7. The method for automatically identifying a ship from which oil is stolen and removed by combining radar satellite images and AIS according to claim 6, wherein obtaining the narrow-head pixels of the linear oil stain band comprises:
comparing the z values of the oil stain indexes on the oil stain and oil stain-like pixels, wherein the smaller the z value is, the closer the oil stain and oil stain-like pixels are to the narrow head of the linear oil stain band, and the larger the z value is, the closer the oil stain and oil stain-like pixels are to the wide head of the linear oil stain band;
calculating an x average value x_zmin and a y average value y_zmin of the greasy dirt and greasy dirt like pixels with the minimum z value, and calculating an x average value x_zmax and a y average value y_zmax of the greasy dirt and greasy dirt like pixels with the maximum z value; if x_zmin is less than x_zmax, x of the narrow-head pixel is the minimum value of x in the greasy dirt and greasy dirt-like pixel with the minimum z value, otherwise, x is the maximum value; if y_zmin is less than y_zmax, y of the narrow-head pixel is the minimum y value of the greasy dirt and the greasy dirt-like pixel with the minimum z value, otherwise, the y is the maximum value.
8. The method for automatically identifying oil theft and removal from a ship in combination with radar satellite imaging and AIS according to claim 6, wherein the step of obtaining the distance and slope between the ship pixel and the narrow head pixel comprises the steps of:
when the distance is smaller than a preset specific threshold dso threshold And the absolute value of the difference between the slope and the included angle between the straight line represented by k and the horizontal direction is smaller than a preset specific threshold k bias The ship and the oil stain belt are positioned in a preset distance threshold value and on an extension line of the linear oil stain belt, the ship pixel is judged to be a source ship of the linear oil stain belt, otherwise, the ship pixel is not the source ship of the linear oil stain belt; where k is the slope of the straight line identified by the greasy dirt pixel using hough variation.
9. The method for automatically identifying oil theft and grease removal of a ship by combining radar satellite images and AIS according to claim 1, wherein tracing the source ship comprises:
according to the shooting time t of the radar satellite image, according to the coordinate system information of the radar satellite image, converting the line number positions x and y of the source ship in the image into longitude and latitude x s , y s ；
Calling specific time length, [ x ] before and after t time from ship AIS database s , y s ]All ship AIS data [ t ] within a specific distance range around a location ais , x ais , y ais , v ais ]The method comprises the steps of carrying out a first treatment on the surface of the Wherein t is ais , x ais , y ais , v ais The AIS data recording time, the ship longitude position, the ship latitude position and the ship navigational speed are respectively;
for each AIS data, calculate |v ais ×(t-t ais ) 2 -(x s -x ais ) 2 -(y s -y ais ) 2 And the AIS data with the smallest value is AIS data of the source ship, and unique identity information is extracted from the AIS data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310340565.3A CN116106850B (en) | 2023-04-03 | 2023-04-03 | Method for automatically identifying oil stain stealing and removing of ship by combining radar satellite image and AIS |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310340565.3A CN116106850B (en) | 2023-04-03 | 2023-04-03 | Method for automatically identifying oil stain stealing and removing of ship by combining radar satellite image and AIS |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116106850A true CN116106850A (en) | 2023-05-12 |
| CN116106850B CN116106850B (en) | 2023-07-07 |
Family
ID=86262340
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310340565.3A Active CN116106850B (en) | 2023-04-03 | 2023-04-03 | Method for automatically identifying oil stain stealing and removing of ship by combining radar satellite image and AIS |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116106850B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915910A (en) * | 2010-07-07 | 2010-12-15 | 大连海事大学 | Method and system for identifying marine oil spill object by marine radar |
| CN206627156U (en) * | 2017-03-28 | 2017-11-10 | 中国人民解放军镇江船艇学院 | A kind of ship spill monitoring system |
| KR20190000151A (en) * | 2017-06-22 | 2019-01-02 | (주)인스페이스 | Method for oil spill detection based vessel tracking |
| KR20190133442A (en) * | 2018-05-23 | 2019-12-03 | 한국해양과학기술원 | Method for detecting oil spills on satellite sar images using artificial neural network |
| CN211036935U (en) * | 2019-10-17 | 2020-07-17 | 嘉兴南洋职业技术学院 | Linear oil stain recovery device |
| CN114236490A (en) * | 2021-12-07 | 2022-03-25 | 大连海事大学 | X-band navigation radar oil spill detection system based on water surface echo model |
| CN114488163A (en) * | 2022-01-07 | 2022-05-13 | 武汉理工大学 | Ship garbage stealing and draining water entering positioning method and device |
-
2023
- 2023-04-03 CN CN202310340565.3A patent/CN116106850B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915910A (en) * | 2010-07-07 | 2010-12-15 | 大连海事大学 | Method and system for identifying marine oil spill object by marine radar |
| CN206627156U (en) * | 2017-03-28 | 2017-11-10 | 中国人民解放军镇江船艇学院 | A kind of ship spill monitoring system |
| KR20190000151A (en) * | 2017-06-22 | 2019-01-02 | (주)인스페이스 | Method for oil spill detection based vessel tracking |
| KR20190133442A (en) * | 2018-05-23 | 2019-12-03 | 한국해양과학기술원 | Method for detecting oil spills on satellite sar images using artificial neural network |
| CN211036935U (en) * | 2019-10-17 | 2020-07-17 | 嘉兴南洋职业技术学院 | Linear oil stain recovery device |
| CN114236490A (en) * | 2021-12-07 | 2022-03-25 | 大连海事大学 | X-band navigation radar oil spill detection system based on water surface echo model |
| CN114488163A (en) * | 2022-01-07 | 2022-05-13 | 武汉理工大学 | Ship garbage stealing and draining water entering positioning method and device |
Non-Patent Citations (6)
| Title |
|---|
| 宋莎莎;赵宇鹏;苏腾飞;马佑军;安伟;孟俊敏;赵朝方;: "海上溢油SAR卫星遥感监测系统研发", 遥感技术与应用, no. 05 * |
| 无: "瑞典航空海洋环境监测系统", 海洋技术, no. 01 * |
| 王彦情;马雷;田原;: "光学遥感图像舰船目标检测与识别综述", 自动化学报, no. 09 * |
| 艾加秋;齐向阳;禹卫东;刘凡;: "一种基于图像分割和归一化灰度Hough变换的SAR图像舰船尾迹CFAR检测算法", 电子与信息学报, no. 11 * |
| 邴磊;邹娜娜;过杰;丛旭东;林勐;史湘君;: "基于星载SAR和AIS构建海面溢油监测方法", 广西科学院学报, no. 02 * |
| 马龙;李颖;兰国新;侯利民;高超;: "基于可分离性指数的溢油图像定量分析", 海洋环境科学, no. 02 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116106850B (en) | 2023-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109409283B (en) | Method, system and storage medium for tracking and monitoring sea surface ship | |
| Galceran et al. | A real-time underwater object detection algorithm for multi-beam forward looking sonar | |
| KR102067242B1 (en) | Method for detecting oil spills on satellite sar images using artificial neural network | |
| CN110275153B (en) | Water surface target detection and tracking method based on laser radar | |
| CN108229433B (en) | Method for detecting ship landing on shore based on straight-line segment detection and shape characteristics | |
| CN101303735A (en) | Method for detecting moving objects in a blind spot region of a vehicle and blind spot detection device | |
| CN110852992B (en) | Ship trail detection method based on top hat transformation and Radon transformation | |
| CN115456970B (en) | Method for identifying carrier line of topside-cut ship based on dimension-reduction regression | |
| CN105787886A (en) | Multi-beam image sonar-based real-time image processing method | |
| CN110807424B (en) | Port ship comparison method based on aerial image | |
| Sato et al. | Study of a collision-avoidance system for ships | |
| CN110147716A (en) | Wave method for detecting area in a kind of SAR image combined based on frequency domain with airspace | |
| Gladstone et al. | Distance estimation for marine vehicles using a monocular video camera | |
| CN116106850B (en) | Method for automatically identifying oil stain stealing and removing of ship by combining radar satellite image and AIS | |
| CN108288268A (en) | Inshore ship detection method in high-resolution remote sensing image based on Projection Analysis | |
| CN110618685A (en) | Unmanned surface vessel obstacle detection error correction and safe collision avoidance method | |
| CN111105419B (en) | Vehicle and ship detection method and device based on polarized SAR image | |
| CN112435249B (en) | Dynamic small target detection method based on circumferential scanning infrared search system | |
| CN114236490B (en) | X-band navigation radar oil spill detection system based on water surface echo model | |
| KR102269384B1 (en) | support system for vessel operation and ship having the same | |
| CN115019561A (en) | External collision risk early warning system of ship towing system under mutual-seeing condition | |
| CN112070786B (en) | Method for extracting warning radar PPI image target and interference | |
| KR20090074927A (en) | Detection method of road vehicles | |
| CN105929379A (en) | Method for improving VTS radar resolution | |
| CN112505683A (en) | Radar and electronic chart information fusion detection method |
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 |