CN113963171A - Method and system for automatically identifying submarine line of shallow stratum section sonar image - Google Patents
Method and system for automatically identifying submarine line of shallow stratum section sonar image Download PDFInfo
- Publication number
- CN113963171A CN113963171A CN202111058860.7A CN202111058860A CN113963171A CN 113963171 A CN113963171 A CN 113963171A CN 202111058860 A CN202111058860 A CN 202111058860A CN 113963171 A CN113963171 A CN 113963171A
- Authority
- CN
- China
- Prior art keywords
- edge response
- image
- edge
- function
- shallow stratum
- 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
- 238000000034 method Methods 0.000 title claims abstract description 55
- 230000001629 suppression Effects 0.000 claims abstract description 15
- 230000004044 response Effects 0.000 claims description 79
- 238000012216 screening Methods 0.000 claims description 4
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 230000014509 gene expression Effects 0.000 claims description 3
- 230000008439 repair process Effects 0.000 claims description 3
- 238000009795 derivation Methods 0.000 claims description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 abstract description 14
- 230000008569 process Effects 0.000 abstract description 13
- 238000000605 extraction Methods 0.000 abstract description 5
- 238000001914 filtration Methods 0.000 abstract description 5
- 230000036039 immunity Effects 0.000 abstract description 4
- 230000002085 persistent effect Effects 0.000 abstract description 2
- 238000002592 echocardiography Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- 239000004480 active ingredient Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- 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
- G01S15/88—Sonar systems specially adapted for specific applications
- G01S15/89—Sonar systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/539—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Acoustics & Sound (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
- Geophysics And Detection Of Objects (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention provides a method and a system for automatically identifying a sea bottom line of a shallow stratum profile sonar image, wherein the method comprehensively considers the edge characteristic attribute and the spatial distribution attribute of the sea bottom line on the shallow stratum profile sonar image, utilizes a specific edge characteristic extraction operator to highlight the edge characteristic on the shallow stratum profile image, adopts a maximum suppression method with horizontal direction tendency to refine the edge characteristic, and realizes the automatic and accurate identification and tracking of the sea bottom line of the shallow stratum profile image based on a density clustering algorithm and a specific search strategy. The problem that a traditional threshold value method and a time sequence filtering algorithm are low in noise immunity and difficult to process persistent water body interference is solved, and the submarine line characteristics of the shallow stratum profile image can be automatically identified and tracked in an aviation, high-efficiency and high-reliability mode.
Description
Technical Field
The invention relates to the technical field of underwater sonar mapping methods and data automatic processing, in particular to a method and a system for automatically identifying a shallow stratum section sonar image sea bottom line.
Background
The shallow stratum profile sonar vertically emits narrow-open-angle sound pulses downwards, sound waves are expanded to a few meters to dozens of meters of shallow stratum below a bed surface through a water body, strong reflection echoes are generated at places where the impedance of a propagation medium is obviously changed, and then the reflected echoes are received by a transducer. The single sound wave is generated and received to form an echo sequence signal which is distributed along the vertical direction, and the shallow stratum profile sonar continuously transmits sound waves and receives echoes in the sailing process, so that a shallow stratum profile echo image along the sailing track direction is formed. The sound wave is obviously strongly reflected at the interface of the water body and the bed surface, and is represented as a boundary between a water body weak echo region and a bed surface strong echo region on a shallow stratum profile image, which is called a seabed line. The position of the seabed line on the shallow stratum section image is positioned, and the method is the basis for vertical geometric correction and intensity gain of the shallow stratum section image.
At present, shallow stratum profile image seabed line extraction is mainly based on a threshold method, in each vertical echo sequence, one echo point which is higher than a set threshold intensity is searched from top to bottom, a time sequence position corresponding to the echo is calibrated to be a boundary position of a water body and the seabed, and the positions determined in each echo sequence are sequentially connected in the horizontal direction to be the seabed line of a shallow-section sonar image.
Influenced by factors such as complex marine environment, water body suspended matters and the like, a large amount of noise or strong interference echoes exist in the water body before the echoes return from the water body and bed surface interface, and the echoes are easily judged as boundary points of the water body and the bed surface by a threshold value method, so that seabed point identification errors, seabed line fracture and tracking failures are caused, and further, the subsequent vertical distance correction and the strength gain have local errors. Filtering processing such as median and trend is carried out on the seabed line extracted by the threshold method, the seabed line extraction quality can be improved to a certain extent, but the seabed line extraction quality is only limited to eliminating scattered noise interference, and the robustness and universality are still poor for continuous operation environment interference, multiple reflection and the like.
Disclosure of Invention
The invention solves the main problems that the traditional threshold value method and the time sequence filtering algorithm have low noise immunity and are difficult to process continuous water body interference.
According to one aspect of the invention, the invention provides a method for automatically identifying a shallow stratum section sonar image sea bottom line, which comprises the following steps:
acquiring a sonar image of a shallow stratum section;
representing the sonar image of the shallow section as an image function:
I(x,y)
wherein x is the horizontal direction, y is the vertical direction, and I represents an image function;
acquiring the edge response characteristic of the sonar image according to the image function;
denoising the edge response characteristic to obtain a denoising edge response characteristic;
determining an edge response characteristic function according to the denoising edge response characteristic, wherein the edge response characteristic function is as follows:
wherein R isxIs a gradient in the x direction, RyThe gradient in the y direction is shown, and R is an edge response characteristic;
converting the edge response characteristic function based on a convolution function exchange law, and constructing an edge response characteristic convolution template of the sonar image;
carrying out non-maximum suppression on the edge response characteristics by using the edge response characteristic convolution template to obtain a binary image of the edge response characteristics;
clustering the binary image by adopting a horizontal density clustering algorithm to obtain an edge feature chain set;
repeatedly screening longest chains from the characteristic chain set in the same time sequence section, and removing short chains which have time sequence overlapping with the selected longest chains;
and repairing the default part between the longest chains according to a trend principle to obtain a complete seabed line.
Further, the repairing the default part between the longest chains according to the trend principle further comprises:
and when the clustered long chains are broken, performing short-term undersea line deletion repair according to the end trend between adjacent chains.
Further, before the step of determining an edge response feature function according to the noise reduction edge response value, the method further includes:
and denoising the sonar image by a Gaussian function.
Further, the performing non-maximum suppression on the edge response feature by using the edge response feature convolution template further includes:
according to the gradient RxAnd RyCalculating the gradient direction angle of the edge response characteristic;
dividing the neighborhood of each pixel center of the sonar image into a preset number of direction areas according to a preset angle;
and comparing the edge response value of the pixel center with the edge response value in the direction area corresponding to the gradient direction angle to determine whether the pixel position is the maximum value.
Further, the determining whether the pixel position is a maximum value further includes:
and if the edge response value of the pixel center is not less than the maximum edge response value in the direction neighborhood, the result of non-maximum suppression of the pixel center is 1.
Further, the determining whether the pixel position is a maximum value further includes:
and if the edge response value of the pixel center is smaller than the maximum edge response value in the direction neighborhood, the non-maximum suppression result of the pixel center is 0.
Further, the edge response value is a first derivative maximum of the image function at the sea bottom line along the gradient direction.
Further, the obtaining the edge response value according to the image function further includes:
by a first expression:
wherein,
representing the derivation operation of the image function I (x, y) in the x and y directions respectively;
and constructing a first-order partial derivative vector of the image function in the x and y directions, wherein the modulus of the first-order partial derivative vector is an edge response value.
Further, clustering the binary graph by using a horizontal density clustering algorithm to obtain an edge feature chain set further comprises:
setting a long strip-shaped search neighborhood with a long axis distributed along the horizontal direction and a longitudinal axis distributed along the vertical direction;
and setting neighborhood parameters of the long strip-shaped search neighborhood, and carrying out horizontal density clustering on the refined response edge characteristics to obtain the edge characteristic chain set.
According to another aspect of the invention, the system for automatically identifying the sea bottom line of the shallow stratum profile sonar image is further disclosed, and is characterized by comprising a storage module and an execution module, wherein the storage module is used for storing the method for automatically identifying the sea bottom line of the shallow stratum profile sonar image as described in any one of the previous aspects, and the execution module is used for calling the method in the storage module and executing specific steps of the method.
The method comprehensively considers the edge characteristic attribute and the spatial distribution attribute of the sea bottom line on the shallow stratum profile sonar image, utilizes a specific edge characteristic extraction operator to highlight the edge characteristic on the shallow stratum profile image, adopts a maximum suppression method with horizontal direction inclination to refine the edge characteristic, and realizes accurate and automatic identification and tracking of the sea bottom line of the shallow stratum profile image based on a density clustering algorithm and a specific search strategy. The problem that a traditional threshold value method and a time sequence filtering algorithm are low in noise immunity and difficult to process persistent water body interference is solved, and the submarine line characteristics of the shallow stratum profile image can be automatically identified and tracked in an aviation, high-efficiency and high-reliability mode.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description, serve to explain the principles of the invention.
FIG. 1 is a schematic diagram of an edge response feature convolution template in an embodiment of the present invention.
FIG. 2 is a schematic diagram of a non-maxima suppression step of an edge response feature in an embodiment of the invention.
FIG. 3 is a schematic diagram of horizontal-like density clustering of the edge features of the sea bottom line in the embodiment of the present invention.
FIG. 4 is a flowchart illustrating automatic sea-bottom line identification and tracking of a shallow profile image according to an embodiment of the present invention.
FIG. 5 is an exemplary diagram of automatic sea-bottom line identification and tracking of a shallow stratigraphic section in an embodiment of the present invention.
Detailed Description
Various exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.
Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments and the accompanying drawings.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.
The implementation process of one embodiment is described in detail below with reference to the flow chart of fig. 4, which is applied to the automatic identification and tracking of the seabed line of the shallow stratum section image.
Step one, determining a seabed line reference range:
since the acquisition of the shallow profile image is a flight survey process, the image is generally a time-series image gradually formed from left to right, and therefore, the tracking of the seabed line should also be a time-series tracking process. Considering that the vertical range of the shallow stratum profile image is large, the approximate range of the seabed line in a certain time sequence can be preliminarily determined according to the tracked position of the seabed line, and then the target image tracked by the seabed line can be determined, as shown in fig. 5 a.
Step two, calculating the edge response characteristics:
according to the empirical sea bottom line position, the target image in the pre-tracking sea bottom line time sequence is intercepted in the vertical range, the Gaussian function fuzzy scale value is 3 or 5, namely sigma is 3 or 5 pixels, the template is convoluted according to the edge gradient direction given by the edge response characteristic function, and the gradient Rx and R of each position of the target image are respectively solvedyAnd a response characteristic R, as in fig. 5 b.
Specifically, the sea bottom line demarcates the boundary between the water body and the bed surface, the weak echo in the water body is above the sea bottom line, the strong echo of the bed surface is below the sea bottom line, and the characteristics are represented as edge characteristics on the shallow stratum section image. In order to highlight the edge feature, a special edge response feature operator is designed.
Firstly, recording a shallow stratum section image as I (x, y), wherein x and y are respectively in the horizontal and vertical directions, and the edge characteristics corresponding to the seabed line are represented as step signals distributed approximately in the direction parallel to the x direction, at the moment, the image function I has a maximum value of a first derivative in the gradient direction at the seabed line, and the maximum value is a pre-searched edge response value.
Specifically, the partial derivative formula can be obtained by
Constructing a vector of first partial derivatives (R) of the image function in the x and y directionsx,Ry) The modulus is the edge response value. In addition, considering the image noise characteristics, the noise influence can be weakened through a Gaussian function, and then an edge response characteristic function R can be determined:
wherein R isxIs a gradient in the x direction, RyThe gradient in the y direction is shown, and R is an edge response characteristic;
where sigma is the fuzzy scale of the gaussian function,
and according to the commutative law of the convolution function, converting the edge response characteristic function in the formula (1) into the convolution of the partial derivative of the Gaussian function G (x, Y, sigma) in the Y direction and the original image, wherein the scale parameter sigma can be adjusted according to the requirement of the resolution of the image, and the reference range is 3-7. An edge response feature convolution template for the shallow stratigraphic section image is thus constructed, as shown in FIG. 1, where each of the gray grids in the edge response feature convolution template represents an echo sample or pixel.
Step three, refining edge response characteristics:
r calculated from the above stepx、RyRespectively calculating the edge gradient directions of all positions of the target image, carrying out non-maximum suppression on the edge response characteristic R of all positions of the target image, and thinning to form edgesBinary image of edge features, fig. 5 c.
Specifically, through the convolution template processing constructed by the formula (1), the gradient R along the x and y directions of all parts of the shallow stratum section image can be obtainedx、RyAnd response characteristics R, wherein the gradient response presents Gaussian distribution along the direction vertical to the seabed line due to the edge ambiguity of the convolution kernel, and in order to refine the seabed line characteristics rapidly, non-maximum suppression is carried out on the edge response characteristics, so that the refinement of the edge response characteristics is realized. Gradient R in x and y directions at various places in the formula (1)xAnd RyAnd solving the gradient direction angle theta of the echo as follows:
θ＝arctan(Ry/Rx) (3)
taking each pixel as a center, dividing the pixel into 8 directions according to the neighborhood of the pixel by 45 degrees, and comparing the gradient direction angle obtained by the formula (3) with the gradient of the corresponding neighborhood to determine whether the position of each pixel is maximum. As shown in fig. 2, the gradient direction θ of the C pixel in the graph is 65 °, and the gradient directions thereof are distributed between 22.5 ° to 67.5 ° and 202.5 ° to 247.5 °, and when the maximum suppression processing is performed, the edge response value at the C position needs to be compared with the edge response values in the neighborhood of nos.:
if the edge response at C R ≧ max (R)②，R⑥) Let TC be 1;
if the edge response at C is R<max(R②，R⑥) Let TC be 0;
in the above operation, R②And R⑥The edge response values of adjacent domains of # two and # C, respectively, and TC is the non-maximum suppression result at C. The above operation is performed for each pixel position, and finally, a refined and binarized edge response characteristic image T can be formed.
Fourthly, clustering the horizontal density of the edge features:
the density clustering method is adopted to realize the preliminary screening of the seabed lines, and the neighborhood long axis epsilon is reasonably sethNeighborhood longitudinal axis εvSearching neighborhood parameters such as a seabed feature number threshold value TN and the like, and performing horizontal density clustering on the refined edge feature binary image to form a series of edge feature chain sets, such asFig. 5 d.
Specifically, through the processing of the foregoing steps, various kinds of edge information on the shallow stratum profile image are highlighted, but not all the edge information is the edge corresponding to the seabed line. Considering that the characteristics of the seabed lines are approximately horizontally distributed along the transverse direction, namely the seabed lines are densely distributed along the horizontal direction, the preliminary screening of the seabed lines can be realized by adopting a density clustering method.
In order to highlight the characteristics of the submarine lines in the horizontal direction, a strip-shaped search neighborhood is arranged to realize density clustering, the shape of the neighborhood is shown in figure 3, and the long axis epsilon of the neighborhoodhDistributed in the horizontal direction,. epsilonvThe feature points are vertically distributed, when the number of the seabed features searched in the neighborhood is greater than a threshold value TN each time the neighborhood is searched, all the feature points searched in the neighborhood are grouped into one type, and due to the overlapping property between the neighborhoods, the same linear feature points in the adjacent neighborhoods are combined into one type, so that the clustering and chaining effect of the discrete features can be realized.
In the clustering process, the related parameters should meet the following requirements: epsilonvThe reference value is 1-5 pixels; and epsilonh>εv，arctan(εv/εh) The allowable degree of the vertical variation of the sea bottom line in the clustering process; TN value range of epsilonh0.5 to 0.8 times of the amount of the active ingredient.
Step five, chain search forms a seabed line tracking result:
based on a long chain principle, selecting the longest chain from the edge characteristic chain set, and taking the longest chain as a part of the seabed line; meanwhile, short chains which are overlapped with the currently selected long chains in a time sequence are removed from the rest characteristic chain set. Repeating the selection process until the selected long chain basically covers the width of the target image in time sequence or the edge characteristic chain set is empty. Finally, the default part between the screened long chains is repaired according to the trend principle, and then the final bottom line result is formed, as shown in fig. 5 e.
In some embodiments, due to the complexity of the operating conditions and the seafloor topography, the actual seafloor line features do not change uniformly, the seafloor line may be subject to breaking, missing situations, and other near horizontally distributed edge features may also be identified as seafloor line features. In order to select a final seabed line from chain features formed by clustering, a seabed line search strategy needs to be established, the characteristics of the seabed line are comprehensively considered, and the following two chain search strategies are specifically provided:
long chain principle: when the shallow stratum profile is operated, the general working condition is stable, the submarine topography changes slowly, and the edge characteristics corresponding to the submarine line change more and more gently. Therefore, during clustering chain forming, the characteristic of the seabed line is mostly long chain, in the same time sequence section, the longest chain is selected as the seabed line, the short chain with time overlapping with the longest chain can be eliminated, and the selecting step is repeated until the preliminary seabed line meeting the condition is obtained.
Based on the trend principle, when the long chains are disconnected, short-time seabed line deletion repair can be carried out according to the terminal trend between the adjacent chains, so that the finally desired target seabed line is obtained.
The method considers the clustering characteristic of the sea bottom line in the horizontal direction, fully considers the spatial distribution trend of the sea bottom line characteristic, overcomes the problems of low noise immunity and difficulty in processing continuous water body interference of the traditional threshold value method and the time sequence filtering algorithm, and can automatically identify and track the sea bottom line characteristic of the shallow stratum profile image in an aviation, high-efficiency and high-reliability manner.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and scope of the present invention should be included in the present invention.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
Claims (10)
1. A method for automatically identifying a shallow stratum section sonar image sea bottom line is characterized by comprising the following steps:
acquiring a sonar image of a shallow stratum section;
representing the sonar image of the shallow section as an image function:
I(x,y)
wherein x is the horizontal direction, y is the vertical direction, and I represents an image function;
acquiring the edge response characteristic of the sonar image according to the image function;
denoising the edge response characteristic to obtain a denoising edge response characteristic;
determining an edge response characteristic function according to the denoising edge response characteristic, wherein the edge response characteristic function is as follows:
wherein R isxIs a gradient in the x direction, RyThe gradient in the y direction is shown, and R is an edge response characteristic;
converting the edge response characteristic function based on a convolution function exchange law, and constructing an edge response characteristic convolution template of the sonar image;
carrying out non-maximum suppression on the edge response characteristics by using the edge response characteristic convolution template to obtain a binary image of the edge response characteristics;
clustering the binary image by adopting a horizontal density clustering algorithm to obtain an edge feature chain set;
repeatedly screening longest chains from the characteristic chain set in the same time sequence section, and removing short chains which have time sequence overlapping with the selected longest chains;
and repairing the default part between the longest chains according to a trend principle to obtain a complete seabed line.
2. The method of claim 1, wherein repairing the default portion between longest chains according to a trend principle further comprises:
and when the clustered long chains are broken, performing short-term undersea line deletion repair according to the end trend between adjacent chains.
3. The method of claim 1, wherein the step of determining an edge response feature function based on the denoised edge response values further comprises:
and denoising the sonar image by a Gaussian function.
4. The method of claim 1, wherein said non-maxima suppression of said edge response feature using said edge response feature convolution template further comprises:
according to the gradient RxAnd RyCalculating the gradient direction angle of the edge response characteristic;
dividing the neighborhood of each pixel center of the sonar image into a preset number of direction areas according to a preset angle;
and comparing the edge response value of the pixel center with the edge response value in the direction area corresponding to the gradient direction angle to determine whether the pixel position is the maximum value.
5. The method of claim 4, wherein said determining whether a pixel location is a maximum further comprises:
and if the edge response value of the pixel center is not less than the maximum edge response value in the direction neighborhood, the result of non-maximum suppression of the pixel center is 1.
6. The method of claim 4, wherein said determining whether a pixel location is a maximum further comprises:
and if the edge response value of the pixel center is smaller than the maximum edge response value in the direction neighborhood, the non-maximum suppression result of the pixel center is 0.
7. The method of claim 1, wherein the edge response value is a first derivative maximum of the image function at a seafloor line along a gradient direction.
8. The method of claim 1, wherein said obtaining the edge response value according to the image function further comprises:
by a first expression:
wherein,
representing the derivation operation of the image function I (x, y) in the x and y directions respectively;
and constructing a first-order partial derivative vector of the image function in the x and y directions, wherein the modulus of the first-order partial derivative vector is an edge response value.
9. The method of claim 1, wherein clustering the binary image using a horizontal density clustering algorithm to obtain a set of edge feature chains further comprises:
setting a long strip-shaped search neighborhood with a long axis distributed along the horizontal direction and a longitudinal axis distributed along the vertical direction;
and setting neighborhood parameters of the long strip-shaped search neighborhood, and carrying out horizontal density clustering on the refined response edge characteristics to obtain the edge characteristic chain set.
10. A system for automatically identifying a shallow stratum section sonar image undersea line is characterized by comprising a storage module and an execution module, wherein the storage module is used for storing a method for automatically identifying a shallow stratum section sonar image undersea line according to any one of claims 1-8, and the execution module is used for calling the method in the storage module and executing specific steps of the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111058860.7A CN113963171B (en) | 2021-09-09 | 2021-09-09 | Automatic identification method and system for seabed line of sonar image of shallow stratum section |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111058860.7A CN113963171B (en) | 2021-09-09 | 2021-09-09 | Automatic identification method and system for seabed line of sonar image of shallow stratum section |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113963171A true CN113963171A (en) | 2022-01-21 |
| CN113963171B CN113963171B (en) | 2024-08-23 |
Family
ID=79461227
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111058860.7A Active CN113963171B (en) | 2021-09-09 | 2021-09-09 | Automatic identification method and system for seabed line of sonar image of shallow stratum section |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113963171B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117994464A (en) * | 2024-04-07 | 2024-05-07 | 中国海洋大学 | Method for constructing three-dimensional micro-geomorphic model by processing profile sonar point cloud data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108872997A (en) * | 2018-05-10 | 2018-11-23 | 国家海洋局第二海洋研究所 | A kind of seabed line detecting method based on side scan sonar data fusion and accurate processing |
| US20210174098A1 (en) * | 2018-08-09 | 2021-06-10 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for lane line identification |
| TWI736446B (en) * | 2020-10-14 | 2021-08-11 | 國立高雄科技大學 | Seabed edge detection method capable of filtering horizontal noise |
-
2021
- 2021-09-09 CN CN202111058860.7A patent/CN113963171B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108872997A (en) * | 2018-05-10 | 2018-11-23 | 国家海洋局第二海洋研究所 | A kind of seabed line detecting method based on side scan sonar data fusion and accurate processing |
| US20210174098A1 (en) * | 2018-08-09 | 2021-06-10 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for lane line identification |
| TWI736446B (en) * | 2020-10-14 | 2021-08-11 | 國立高雄科技大學 | Seabed edge detection method capable of filtering horizontal noise |
Non-Patent Citations (3)
| Title |
|---|
| 张济博;潘国富;丁维凤;: "基于LOG算子的侧扫声呐海底线检测", 海洋通报, no. 03, 15 June 2010 (2010-06-15) * |
| 王爱学: "采用方向自适应密度聚类自动检测侧扫声呐图像海底线", 武汉大学学报(信息科学版), 28 November 2023 (2023-11-28) * |
| 赵建虎;王晓;张红梅;: "侧扫声呐图像海底线自动提取方法研究", 武汉大学学报(信息科学版), no. 12, 31 December 2017 (2017-12-31) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117994464A (en) * | 2024-04-07 | 2024-05-07 | 中国海洋大学 | Method for constructing three-dimensional micro-geomorphic model by processing profile sonar point cloud data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113963171B (en) | 2024-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fonseca et al. | Angular range analysis of acoustic themes from Stanton Banks Ireland: A link between visual interpretation and multibeam echosounder angular signatures | |
| CN109738903A (en) | A kind of side scan sonar real-time two-dimensional imaging method and system | |
| CN115187666A (en) | Deep learning and image processing combined side-scan sonar seabed elevation detection method | |
| CN115248436A (en) | Imaging sonar-based fish resource assessment method | |
| CN113807324A (en) | Sonar image recognition method and device, electronic equipment and storage medium | |
| CN116503268B (en) | Quality improvement method for radar echo image | |
| Beaudoin et al. | Geometric and radiometric correction of multibeam backscatter derived from Reson 8101 systems | |
| CN113963171B (en) | Automatic identification method and system for seabed line of sonar image of shallow stratum section | |
| Li et al. | Automatic detection of pipelines from sub-bottom profiler sonar images | |
| CN112164079A (en) | Sonar image segmentation method | |
| Wang et al. | Sea bottom line tracking in side-scan sonar image through the combination of points density clustering and chains seeking | |
| CN112558077B (en) | Multi-beam filtering method and processing terminal based on adjacent banners | |
| Liu et al. | A method of sidelobe effect suppression for multibeam water column images based on an adaptive soft threshold | |
| CN112529072B (en) | Submerged object identification and positioning method based on sonar image processing | |
| CN113554123B (en) | Automatic sand production ship identification method based on acousto-optic linkage | |
| Wang et al. | Bottom Tracking Method Based on LOG/Canny and the Threshold Method for Side-scan Sonar. | |
| Maroni et al. | Horizon picking on subbottom profiles using multiresolution analysis | |
| CN113705605B (en) | Automatic cleaning method for abnormal values of multi-beam sounding data through partial manual intervention | |
| CN114879180B (en) | Seamless situation awareness method for real-time fusion of unmanned ship-borne multi-element multi-scale radar | |
| AU2002246296A1 (en) | Classifying seafloor roughness with SOM and LVQ | |
| Zhang et al. | A mosaic method based on feature matching for side scan sonar images | |
| CN116930976B (en) | Submarine line detection method of side-scan sonar image based on wavelet mode maximum value | |
| CN113313651A (en) | Method for repairing side-scan sonar image texture distortion area based on peripheral change | |
| Kammerer et al. | Comparative study of shallow water multibeam imagery for cleaning bathymetry sounding errors | |
| Ling et al. | Score-Based Multibeam Point Cloud Denoising |
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 |