CN119206518B - A method for extracting floating ball cage-type sea surface aquaculture farms based on SAR images - Google Patents
A method for extracting floating ball cage-type sea surface aquaculture farms based on SAR images Download PDFInfo
- Publication number
- CN119206518B CN119206518B CN202411720964.3A CN202411720964A CN119206518B CN 119206518 B CN119206518 B CN 119206518B CN 202411720964 A CN202411720964 A CN 202411720964A CN 119206518 B CN119206518 B CN 119206518B
- Authority
- CN
- China
- Prior art keywords
- image
- floating ball
- sea surface
- ball cage
- farm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000009360 aquaculture Methods 0.000 title description 6
- 244000144974 aquaculture Species 0.000 title description 6
- 238000000605 extraction Methods 0.000 claims abstract description 51
- 230000000877 morphologic effect Effects 0.000 claims abstract description 47
- 238000012545 processing Methods 0.000 claims abstract description 35
- 238000012258 culturing Methods 0.000 claims abstract description 30
- 238000005457 optimization Methods 0.000 claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 12
- 230000009466 transformation Effects 0.000 claims abstract description 9
- 230000008030 elimination Effects 0.000 claims description 15
- 238000003379 elimination reaction Methods 0.000 claims description 15
- 238000001914 filtration Methods 0.000 claims description 13
- 238000009499 grossing Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000005260 corrosion Methods 0.000 claims description 3
- 230000007797 corrosion Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 230000001788 irregular Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000009364 mariculture Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000013535 sea water Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/40—Filling a planar surface by adding surface attributes, e.g. colour or texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of image processing, and discloses a floating ball cage type sea surface culturing farm extraction method based on SAR images, which comprises the following steps of reading SAR images; the method comprises the steps of roughly extracting a floating ball cage type sea surface culturing farm, finely extracting the floating ball cage type sea surface culturing farm, and guiding out the extracting result of the floating ball cage type sea surface culturing farm in an image file mode. The invention creatively provides a noise point removing method based on neighborhood high-bright point distribution density calculation and a morphological optimization method based on space scale transformation based on the backward scattering intensity characteristic and the space distribution characteristic of a floating ball cage type sea surface farm in SAR images, and innovatively constructs a two-step extraction method of coarse extraction of the farm and fine extraction of the farm, and realizes accurate and rapid automatic extraction of a floating ball cage type sea surface farm area through combined treatment processes of automatic threshold binarization, discrete noise point removing, morphological expansion, hole filling, morphological optimization and the like.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to a floating ball cage type sea surface farm extraction method based on SAR images.
Background
The offshore farms are used for realizing the cultivation of various aquatic products in specific sea areas by manually arranging fishery facilities and utilizing natural ocean environments, and currently become strategic choices of main ocean countries. The offshore farms in China are mainly distributed in coastal and offshore areas of yellow sea, east sea and south sea, and are one of important marine prop industries in China and the main attack direction of fishery development in China. The method has great significance on the effective supervision of the offshore farms, and the traditional offshore cultivation monitoring has the problems of high difficulty, high cost, low efficiency and the like. The remote sensing monitoring mode of the synthetic aperture radar (SYNTHETIC APERTURE RADAR, SAR) has the advantages of large observation range, low cost, all weather, high monitoring speed, high data accuracy and the like, and is suitable for monitoring a large-scale offshore farm. The information such as the spatial distribution, the area and the like of the offshore farms can be comprehensively and accurately obtained through the SAR remote sensing monitoring means, scientific basis can be provided for planning and management of the aquaculture, the supervision efficiency is improved, the supervision degree on the offshore aquaculture activities is enhanced, the reasonable layout of the aquaculture areas is facilitated, the aquaculture structure is optimized, and the aquaculture benefit is improved. In addition, remote sensing monitoring of the offshore farms is carried out, distribution and change conditions of the offshore farms can be found in time, the mariculture facilities are prevented from becoming barriers for offshore traffic, potential safety hazards are further eliminated, the offshore navigation environment is maintained, and smoothness and safety of ship navigation are guaranteed.
The floating ball cage type sea surface floating raft cultivation is one of main sea cultivation modes in China, the water floating balls are made of white or black rubber materials, the diameter is about 30 cm, the distance between the floating balls is about 75 cm, a discontinuous interval arrangement mode is shown in a layout area, the diameter of the underwater cage is about 20 cm, and the length is 4-6 m. Because the floating ball of the floating ball cage type sea surface farm exposed on the sea surface is smaller and is easily influenced by sea conditions of sea water, sea impurities and SAR imaging, the background noise of the sea surface in the SAR image is larger, the bright point targets of the floating ball are more fuzzy, and how to accurately extract the cultivation area based on the locally and intensively distributed bright point targets of the floating ball in the SAR image has higher technical difficulty, and no disclosed and effective extraction method exists at present.
Disclosure of Invention
The invention aims to provide a floating ball cage type sea surface culturing field extraction method based on SAR images, which is based on the fact that the floating ball cage type sea surface culturing field presents backward scattering intensity characteristics (backward scattering intensity of a sea surface floating ball target is stronger than that of a sea surface background area) and spatial distribution characteristics (a single culturing field is integrally distributed in a polygonal block shape, floating ball targets on water are arranged at regular intervals), and a noise point removing method based on neighborhood high-brightness point distribution density calculation and a morphological optimization method based on spatial scale transformation are innovatively provided, and a two-step extracting method of 'culturing field coarse extraction and culturing field fine extraction' is innovatively constructed, and accurate and rapid automatic extraction of a floating ball cage type sea surface culturing field area is realized through combined treatment processes such as automatic threshold binarization, discrete noise point removing, morphological expansion, hole filling and morphological optimization.
The technical scheme adopted by the invention is a floating ball cage type sea surface farm extraction method based on SAR images, which comprises the following steps:
S1, reading SAR images;
S2, a floating ball cage type sea surface culturing farm is extracted in a rough mode, and the method specifically comprises the following steps:
S21, performing automatic threshold binarization on the SAR image read in the S1;
s22, noise elimination is carried out on the image after binarization processing;
S23, performing expansion processing on the image after noise elimination;
s24, filling holes in a floating ball cage type sea surface farm area in the expanded image;
s3, a floating ball cage type sea surface farm is extracted precisely, and the specific steps comprise:
s31, performing morphological optimization on the image after the floating ball cage type sea surface culturing farm is roughly extracted in the S2;
S32, performing morphological open operation processing on the image with the optimized morphology;
S33, performing Gaussian filtering processing on the image subjected to morphological opening operation processing;
S34, removing small areas of the non-floating ball cage type sea surface farms from the Gaussian filtered images, and finally obtaining floating ball cage type sea surface farms;
s4, the extraction result of the floating ball cage type sea surface farm is exported in the form of an image file.
Further, in S21, the specific step of performing automatic thresholding binarization on the SAR image read in S1 includes:
S211, aiming at the SAR image read in S1 Normalizing all pixel values to be within a range of [0,255 ];
s212, adaptively calculating an optimal segmentation threshold value by a maximum inter-class variance method, and obtaining a read SAR image Binarization processing is carried out to obtain a highlight pixel part, wherein the highlight pixel is marked as 1, and other pixels are marked as 0 to obtain an automatic threshold binary image。
Further, in S22, a noise point removing method based on the neighborhood high-brightness point distribution density calculation is used to remove noise from the binarized image, and the specific steps include:
s221, regarding binary image Any of the pixels in (a)The pixel is in a binary imageThe value of (a) isCalculating the pixel positionDistribution density of highlights in neighborhoodThe calculation formula is as follows:
;
wherein n is the radius of the neighborhood range, x is the pixel in the binary image The row number of the pixel is yColumn number in i is pixel in binary imageThe line number index in j is the pixel in binary imageColumn number index in (a);
S222, initializing and binary image Binary image of the same sizeSetting a threshold valueThe range is [0,1], which aims at the binary imageAny of the pixels in (a)If:
And is also provided with ,
Then in the binary imageThe corresponding mark of the pixel is 1, otherwise the mark of the pixel is 0;
s223, traversing the binary image Each pixel in the two-dimensional image is calculated and judged and corresponds to the binary imageAnd (5) assigning values.
Further, in S23, the specific steps of performing expansion processing on the image after noise removal are selecting a circular structural element with a radius of l, and performing morphological expansion processing on the binary image obtained in S22Performing expansion treatment to obtain a binary image。
Further, in the step S24, the hole filling is performed on the floating ball cage type sea surface farm area in the expanded image, specifically, the method comprises the steps of based on the binary image obtained in the step S23Filling isolated hole areas existing in the connected block areas with pixel value of 1 by adopting a flooding filling algorithm to obtain a binary image。
Further, in the step S31, the morphological optimization is performed on the image after the floating ball cage type sea surface culturing farm is extracted in the S2 in a rough way by a morphological optimization method based on space scale transformation, and the specific steps are that a sampling coefficient S is set, and binary images are sequentially subjected to the steps ofS times downsampling and S times upsampling are carried out to obtain a binary imageEdge smoothing is completed.
Further, in the step S32, the specific step of performing morphological open operation on the image after morphological optimization is thatRectangular structural element of (a) is sequentially mapped to binary imagePerforming morphological corrosion and morphological expansion operation to obtain a binary image;
Where u is the length of the rectangular structural element and v is the width of the rectangular structural element.
Further, in the step S33, the method for performing Gaussian filtering processing on the image after morphological open operation processing comprises the following specific steps ofSize filter kernel, gaussian kernel standard deviation in both X and Y directionsFor binary imagePerforming Gaussian filtering to obtain a binary image;
Where k is the length/width of the filter kernel.
Further, in the step S34, the specific step of eliminating the small region of the non-floating ball cage type sea surface farm for the image after Gaussian filtering treatment is that the pixel number threshold value of a single floating ball cage type sea surface farm object is obtained by combining the conversion of the actual resolution of the image, and a binary image is countedThe number of the pixels contained in the independent pixel blocks with the pixel values of 1 in all the space connected are compared with the pixel number threshold one by one, the pixel blocks with the pixel number smaller than the pixel number threshold are integrally assigned to be 0, and the extraction result is removedThe non-floating ball cage type sea surface culturing farm target comprises noise plaque, small fishing boat and other offshore facilities to obtain a binary image。
The invention has the beneficial effects that:
(1) The floating ball cage type sea surface culturing farm extraction method based on the SAR image can automatically finish the extraction of the floating ball cage type sea surface culturing farm;
(2) The noise point removing method based on the neighborhood high-bright point distribution density calculation is innovatively provided, and discrete noise in the extraction result of the floating ball cage type sea surface culturing farm can be effectively removed;
(3) The morphological optimization method based on space scale transformation is innovatively provided, so that the optimization of irregular morphological structure of the extraction result of the floating ball cage type sea surface farm can be effectively realized;
(4) The method has good accuracy, can effectively extract the polygonal distribution range of the floating ball cage type sea surface farm, and is basically consistent with the actual distribution condition of the floating ball cage type sea surface farm;
(5) The method has higher processing efficiency, and the processing time consumption is lower than 3 seconds (8-core processor, 1.60GHz and running memory of 12 GB) for SAR images with the size of 480 multiplied by 512 pixels.
Drawings
FIG. 1 is a schematic overall flow chart of the present invention.
FIG. 2 is a schematic diagram of SAR image comprising floating ball cage type sea surface farm targets in embodiment 1 of the present invention。
FIG. 3 is a binary image after automatic thresholding by maximum inter-class variance in embodiment 1 of the present invention。
FIG. 4 is a binary image of the embodiment 1 of the present invention after noise elimination by the noise elimination method based on the neighborhood high-brightness point distribution density calculation。
FIG. 5 is a binary image of the expansion process according to example 1 of the present invention。
FIG. 6 is a binary image of a floating ball cage type sea surface farm area subjected to hole filling by a flooding filling algorithm in accordance with embodiment 1 of the present invention。
FIG. 7 is a binary image after optimization of the spatial scale transformation morphology in embodiment 1 of the present invention。
FIG. 8 is a binary image after morphological open operation in example 1 of the present invention。
FIG. 9 is a binary image obtained by Gaussian filtering in embodiment 1 of the invention。
FIG. 10 is a binary image of a non-floating ball cage type sea surface farm after small region culling in example 1 of the present invention。
Fig. 11 is a schematic diagram of final extraction results and original SAR image superposition of a floating ball cage type sea surface farm in embodiment 1 of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more clear and apparent, the technical solutions of the present invention will be described in further detail with reference to the accompanying drawings and examples.
A floating ball cage type sea surface farm extraction method based on SAR images is shown in figure 1, and comprises the following steps:
s1, reading SAR images.
Reading SAR image containing floating ball cage type sea surface farm target。
S2, a floating ball cage type sea surface culturing farm is extracted in a rough mode.
The process is based on the read SAR imageThe method comprises the following specific steps of performing discrete noise elimination by sequentially adopting a maximum inter-class variance method automatic threshold image binarization method, performing discrete noise elimination by a noise point elimination method based on neighborhood high-point distribution density calculation, performing expansion treatment by a morphological expansion method, performing hole filling by a flooding filling algorithm to obtain a rough extraction result of each floating ball cage type sea surface farm object, and taking the rough extraction result as input of the precise extraction of the next floating ball cage type sea surface farm, wherein the specific steps comprise:
s21, automatic threshold binarization is carried out on the SAR image read in the S1, and the specific steps comprise:
S211, aiming at the SAR image read in S1 Normalizing all pixel values to be within a range of [0,255 ];
S212, adaptively calculating an optimal segmentation threshold by a maximum inter-class variance method (Otsu), and performing SAR image reading Binarization processing is carried out to obtain a highlight pixel part, wherein the highlight pixel is marked as 1, and other pixels are marked as 0 to obtain an automatic threshold binary image。
S22, noise elimination is carried out on the image after the binarization processing.
Due to the read SAR imageThe noise in the floating ball cage type sea surface farm is obvious, highlight noise points are also identified as highlight points after binarization processing, but the highlight points are scattered and randomly distributed on the area, and the highlight pixels in the floating ball cage type sea surface farm area are locally and intensively distributed, so the invention creatively provides a noise point removing method based on neighborhood highlight point distribution density calculation to remove space discrete highlight noise pixel points, and the method comprises the following specific steps:
s221, regarding binary image Any of the pixels in (a)The pixel is in a binary imageThe value of (a) isCalculating the pixel positionDistribution density of highlights in neighborhoodThe calculation formula is as follows:
;
wherein n is the radius of the neighborhood range, x is the pixel in the binary image The row number of the pixel is yColumn number in i is pixel in binary imageThe line number index in j is the pixel in binary imageColumn number index in (a);
S222, initializing and binary image Binary image of the same sizeSetting a threshold valueThe range is [0,1], which aims at the binary imageAny of the pixels in (a)If:
And is also provided with ,
Then in the binary imageThe corresponding mark of the pixel is 1, otherwise the mark of the pixel is 0;
s223, traversing the binary image Each pixel in the two-dimensional image is calculated and judged and corresponds to the binary imageAnd (5) assigning values.
S23, performing expansion processing on the image after noise elimination.
In order to connect the highlight points and block targets of the local floating ball cage type sea surface farm which are not communicated, a single floating ball cage type sea surface farm object which is actually corresponding to the local floating ball cage type sea surface farm object is constructed, a circular structural element with the radius of l is selected, and a morphological expansion method is adopted to obtain a binary image of S22Performing expansion treatment, converting the locally concentrated and distributed highlight pixel set into a single floating ball cage type sea surface plant object, and obtaining a binary image after treatment。
S24, filling holes in the floating ball cage type sea surface farm area in the expanded image.
Because the single floating ball cage type sea surface farm object is usually in a polygonal block structure with continuous local space, the binary image obtained based on S23And filling an isolated hole area (corresponding to a local connected pixel set with a pixel value of 0) in a connected block area (corresponding to a local connected pixel set with a pixel value of 1) with a pixel value of1 by adopting a flood filling (flood filling) algorithm. I.e. binary imageThe whole value of the 0-value hole area pixels in the area block with the pixel values of 1 is assigned to 1 to obtain a binary image。
S3, precisely extracting the floating ball cage type sea surface farm.
On the basis of a floating ball cage type sea surface farm coarse extraction result, the method integrates morphological optimization, morphological opening operation, gaussian filtering and non-floating ball cage type sea surface farm small region rejection based on spatial scale transformation, carries out further optimization treatment on the floating ball cage type sea surface farm extraction result, and finally obtains a floating ball cage type sea surface farm refined extraction result conforming to a real distribution form, and comprises the following specific steps:
s31, performing morphological optimization on the image after the floating ball cage type sea surface culturing farm is roughly extracted in the step S2.
Because the actual floating ball cage type sea surface farm object usually presents a regular polygonal block space characteristic, in order to further obtain the macroscopic structural characteristic of the floating ball cage type sea surface farm extraction result, the invention creatively provides a form optimization method based on space scale transformation, which optimizes the floating ball cage type sea surface farm extraction result in an irregular form structure, namely, setting a sampling coefficient S and sequentially optimizing a binary imageS-times downsampling (Downsampling) and S-times upsampling (Subsampling) are performed to obtain a binary imageEdge smoothing is completed.
S32, performing morphological open operation processing on the image with the optimized morphology.
In order to smooth the outline of the extraction result of the floating ball cage type sea surface culturing farm, a narrow neck is broken and a tiny protruding area is eliminated, and the floating ball cage type sea surface culturing farm is selectedIs arranged in the shape of a rectangular structural element, for binary imagePerforming morphological open operation, i.e. selectingRectangular structural element of (a) is sequentially mapped to binary imagePerforming morphological corrosion and morphological expansion operation to obtain a binary image;
Where u is the length of the rectangular structural element and v is the width of the rectangular structural element.
S33, performing Gaussian filtering processing on the image subjected to morphological opening operation processing.
In order to further remove discrete noise spots in the extraction result of the floating ball cage type sea surface farm, smoothing the profile of the extraction result of the floating ball cage type sea surface farm, adoptingSize filter kernel, gaussian kernel standard deviation in both X (horizontal lateral direction) and Y (horizontal vertical direction)For binary imagePerforming Gaussian filtering to obtain a binary image;
Where k is the length/width of the filter kernel.
And S34, removing the small region of the non-floating ball cage type sea surface farm from the Gaussian filtered image, and finally obtaining the floating ball cage type sea surface farm extraction result.
According to actual remote sensing image data statistics, the area of a floating ball cage type sea surface farm object is generally larger than 100 square meters, according to the characteristic, the pixel number threshold value of a single floating ball cage type sea surface farm object is obtained through image actual resolution conversion, and a binary image is countedThe number of the pixels contained in the independent pixel blocks with the pixel values of 1 in all the space connected are compared with the pixel number threshold one by one, the pixel blocks with the pixel number smaller than the pixel number threshold are integrally assigned to be 0, and the extraction result is removedThe non-floating ball cage type sea surface culturing farm target includes noise plaque, small fishing boat, other marine facilities, etc. to obtain binary imageThe final extraction result of the floating ball cage type sea surface farm is obtained.
S4, the extraction result of the floating ball cage type sea surface farm is exported in the form of an image file.
Extracting result (binary image) of floating ball cage type sea surface culturing farm) Derived in the form of an image file.
Example 1
In this embodiment, based on a Ka band (K-above band) SAR image (including a floating ball cage type sea surface farm target in a region) covering a certain sea area in China, the method for extracting the floating ball cage type sea surface farm based on the SAR image according to the present invention realizes the extraction of the floating ball cage type sea surface farm in the image, as shown in fig. 1, and specifically includes the following steps:
s1, reading SAR images.
In MATLAB software, reading SAR image files containing floating ball cage type sea surface farm targets through encoding to obtain SAR imagesAs shown in fig. 2.
S2, a floating ball cage type sea surface culturing farm is extracted in a rough mode, and the method specifically comprises the following steps:
S21, automatic threshold binarization is carried out on the SAR image read in the S1.
By MATLAB coding, aiming at read SAR imageNormalizing all pixel values to be within the range of [0,255], adaptively calculating an optimal segmentation threshold value by a maximum inter-class variance method, and obtaining a read SAR imageBinarization processing is carried out to obtain a highlight pixel part, wherein the highlight pixel is marked as 1, and other pixels are marked as 0 to obtain an automatic threshold binary imageAs shown in fig. 3.
S22, noise elimination is carried out on the image after the binarization processing.
The noise point eliminating method based on neighborhood high-brightness point distribution density calculation aims at a binary image through MATLAB codingCalculating the distribution density of the highlight points in the 5 multiplied by 5 neighborhood of any pixel in the image, setting the density threshold value to be 0.4, and initializing and binary-mappingBinary image of the same sizeMapping binary valuesPixels with distribution density of highlight points in the middle neighborhood being more than 0.4 and pixel value being 1 are arranged in a binary image(As in fig. 4) the corresponding pel is assigned a value of 1 and the remaining pels are assigned a value of 0.
S23, performing expansion processing on the image after noise elimination.
Selecting a circular structural element with radius of 3 by MATLAB coding, and adopting a morphological expansion method to obtain a binary image in S22Performing expansion treatment to obtain a binary imageAs shown in fig. 5.
S24, filling holes in the floating ball cage type sea surface farm area in the expanded image.
Binary image obtained based on S23 through MATLAB codingAdopting flooding filling algorithm to make binary imageThe whole pixel of the 0-value hole area in the connected block area with the pixel value of 1 is assigned to be 1 to obtain a binary imageAs shown in fig. 6.
S3, a floating ball cage type sea surface farm is extracted precisely, and the specific steps comprise:
s31, performing morphological optimization on the image after the floating ball cage type sea surface culturing farm is roughly extracted in the S2, and finishing edge smoothing.
By MATLAB coding and the morphological optimization method based on space scale transformation, the sampling coefficient 4 is set, and binary images are sequentially subjected to4 Times of downsampling and 4 times of upsampling are carried out to obtain a binary image(As in fig. 7), edge smoothing is completed.
S32, performing morphological open operation processing on the image with the optimized morphology.
By MATLAB coding, 5×5 rectangular structural elements are selected for binary imagePerforming morphological open operation to obtain a binary imageAs shown in fig. 8.
S33, performing Gaussian filtering processing on the image subjected to morphological opening operation processing.
By MATLAB coding, filtering kernel with 3×3 size, gaussian kernel standard deviation 3 in X (horizontal transverse direction) and Y (horizontal vertical direction), and mapping binary imagePerforming Gaussian filtering to obtain a binary imageAs shown in fig. 9.
And S34, removing the small region of the non-floating ball cage type sea surface farm from the Gaussian filtered image, and finally obtaining the floating ball cage type sea surface farm extraction result.
Setting the pixel number threshold value to be 1000 through MATLAB coding, and calculating a statistical binary imageThe number of the pixels contained in the independent pixel blocks with the pixel values of 1 in all the space connected are compared with the pixel number threshold one by one, the pixel blocks with the pixel number smaller than the pixel number threshold are integrally assigned to be 0, and a binary image is obtainedAs shown in fig. 10, a binary imageThe final extraction result of the floating ball cage type sea surface farm is shown in fig. 11, which is a schematic diagram of the superposition of the original SAR image and the extraction result of the floating ball cage type sea surface farm.
S4, the extraction result of the floating ball cage type sea surface farm is exported in the form of an image file.
In MATLAB software, the floating ball cage type sea surface farm extraction result (binary image) is obtained through encoding) Derived in the form of an image file.
In this embodiment, the total time required for the extraction process of the floating ball cage type sea surface farm target for the SAR image (480×512 pel size) is 2.87 seconds (8-core processor, 1.60GHz, running memory 12 GB).
Claims (8)
1. The floating ball cage type sea surface culturing farm extraction method based on SAR image is characterized by comprising the following steps:
S1, reading SAR images;
S2, a floating ball cage type sea surface culturing farm is extracted in a rough mode, and the method specifically comprises the following steps:
S21, performing automatic threshold binarization on the SAR image read in the S1;
s22, noise elimination is carried out on the image after binarization processing by using a noise point elimination method based on neighborhood high-brightness point distribution density calculation, and the specific steps comprise:
s221, regarding binary image Any of the pixels in (a)The pixel is in a binary imageThe value of (a) isCalculating the pixel positionDistribution density of highlights in neighborhoodThe calculation formula is as follows:
;
wherein n is the radius of the neighborhood range, x is the pixel in the binary image The row number of the pixel is yColumn number in i is pixel in binary imageThe line number index in j is the pixel in binary imageColumn number index in (a);
S222, initializing and binary image Binary image of the same sizeSetting a threshold valueThe range is [0,1], which aims at the binary imageAny of the pixels in (a)If:
And is also provided with ,
Then in the binary imageThe corresponding mark of the pixel is 1, otherwise the mark of the pixel is 0;
s223, traversing the binary image Each pixel in the two-dimensional image is calculated and judged and corresponds to the binary imagePerforming a middle assignment;
S23, performing expansion processing on the image after noise elimination;
s24, filling holes in a floating ball cage type sea surface farm area in the expanded image;
s3, a floating ball cage type sea surface farm is extracted precisely, and the specific steps comprise:
s31, performing morphological optimization on the image after the floating ball cage type sea surface culturing farm is roughly extracted in the S2;
S32, performing morphological open operation processing on the image with the optimized morphology;
S33, performing Gaussian filtering processing on the image subjected to morphological opening operation processing;
S34, removing small areas of the non-floating ball cage type sea surface farms from the Gaussian filtered images, and finally obtaining floating ball cage type sea surface farms;
s4, the extraction result of the floating ball cage type sea surface farm is exported in the form of an image file.
2. The method for extracting the floating ball cage type sea surface farm based on the SAR image according to claim 1, wherein in the step S21, the specific step of performing automatic threshold binarization on the SAR image read in the step S1 comprises the following steps:
S211, aiming at the SAR image read in S1 Normalizing all pixel values to be within a range of [0,255 ];
s212, adaptively calculating an optimal segmentation threshold value by a maximum inter-class variance method, and obtaining a read SAR image Binarization processing is carried out to obtain a highlight pixel part, wherein the highlight pixel is marked as 1, and other pixels are marked as 0 to obtain an automatic threshold binary image。
3. The SAR image-based floating ball cage sea surface farm extraction method as set forth in claim 1, wherein in S23, the expansion processing of the noise-removed image comprises selecting circular structural element with radius of l, and adopting morphological expansion method to obtain binary image of S22Performing expansion treatment to obtain a binary image。
4. The SAR image-based floating ball cage type sea surface farm extraction method as set forth in claim 3, wherein in S24, the hole filling is performed on the floating ball cage type sea surface farm region in the expanded image, specifically comprising the steps of based on the binary image obtained in S23Filling isolated hole areas existing in the connected block areas with pixel value of 1 by adopting a flooding filling algorithm to obtain a binary image。
5. The method for extracting the floating ball cage type sea surface farm based on the SAR image as set forth in claim 4, wherein in the step S31, the image after the floating ball cage type sea surface farm is roughly extracted in the step S2 is subjected to morphological optimization by a morphological optimization method based on spatial scale transformation, and the specific steps are that a sampling coefficient S is set, and binary images are sequentially subjected to the steps ofS times downsampling and S times upsampling are carried out to obtain a binary imageEdge smoothing is completed.
6. The SAR image-based floating ball cage sea surface farm extraction method as set forth in claim 5, wherein in S32, the morphological open operation of the morphological optimized image is performed by selectingRectangular structural element of (a) is sequentially mapped to binary imagePerforming morphological corrosion and morphological expansion operation to obtain a binary image;
Where u is the length of the rectangular structural element and v is the width of the rectangular structural element.
7. The SAR image-based floating ball cage sea surface farm extraction method as set forth in claim 6, wherein in S33, the specific step of performing Gaussian filter processing on the morphological open operation processed image comprises the steps ofSize filter kernel, gaussian kernel standard deviation in both X and Y directionsFor binary imagePerforming Gaussian filtering to obtain a binary image;
Where k is the length/width of the filter kernel.
8. The SAR image-based floating ball cage type sea surface farm extraction method as set forth in claim 7, wherein in the step S34, the specific step of performing non-floating ball cage type sea surface farm small area elimination on the Gaussian filtered image is to obtain the pixel number threshold value of the single floating ball cage type sea surface farm object by combining the actual resolution conversion of the image, and the binary image is calculatedThe number of the pixels contained in the independent pixel blocks with the pixel values of 1 in all the space connected are compared with the pixel number threshold one by one, the pixel blocks with the pixel number smaller than the pixel number threshold are integrally assigned to be 0, and the extraction result is removedThe non-floating ball cage type sea surface culturing farm target comprises noise plaque, small fishing boat and other offshore facilities to obtain a binary image。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411720964.3A CN119206518B (en) | 2024-11-28 | 2024-11-28 | A method for extracting floating ball cage-type sea surface aquaculture farms based on SAR images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411720964.3A CN119206518B (en) | 2024-11-28 | 2024-11-28 | A method for extracting floating ball cage-type sea surface aquaculture farms based on SAR images |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119206518A CN119206518A (en) | 2024-12-27 |
| CN119206518B true CN119206518B (en) | 2025-03-14 |
Family
ID=94057097
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411720964.3A Active CN119206518B (en) | 2024-11-28 | 2024-11-28 | A method for extracting floating ball cage-type sea surface aquaculture farms based on SAR images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119206518B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101714252A (en) * | 2009-11-26 | 2010-05-26 | 上海电机学院 | Method for extracting road in SAR image |
| CN111353371A (en) * | 2019-11-22 | 2020-06-30 | 中国电子科技集团公司电子科学研究院 | Shoreline extraction method based on spaceborne SAR images |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100500100C (en) * | 2007-11-20 | 2009-06-17 | 深圳市迈科龙影像技术有限公司 | Ultrasonic uterine birth control ring image intelligent recognition method and equipment |
| JP5753791B2 (en) * | 2008-12-25 | 2015-07-22 | メディック ビジョン イメージング ソルーション リミティッド | Method for providing denominated predetermined resolution medical image, system for providing predetermined denoised predetermined resolution medical image |
| CN102938071B (en) * | 2012-09-18 | 2015-06-03 | 西安电子科技大学 | Fuzzy clustering analysis method for detecting synthetic aperture radar (SAR) image changes based on non-local means |
| KR101590483B1 (en) * | 2013-11-25 | 2016-02-01 | 조선대학교산학협력단 | Method for processing image segmentation using Morphological operation |
| CN106327442A (en) * | 2016-08-22 | 2017-01-11 | 上海奥通激光技术有限公司 | Multispectral micro-imaging field depth extension method and system |
| CN107909589B (en) * | 2017-11-01 | 2020-10-09 | 浙江工业大学 | Tooth image segmentation method combining C-V level set and GrabCont algorithm |
| CN117152607A (en) * | 2023-08-23 | 2023-12-01 | 福建省经纬数字科技有限公司 | Remote sensing image water body extraction method, system and equipment based on unetMForier network |
-
2024
- 2024-11-28 CN CN202411720964.3A patent/CN119206518B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101714252A (en) * | 2009-11-26 | 2010-05-26 | 上海电机学院 | Method for extracting road in SAR image |
| CN111353371A (en) * | 2019-11-22 | 2020-06-30 | 中国电子科技集团公司电子科学研究院 | Shoreline extraction method based on spaceborne SAR images |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119206518A (en) | 2024-12-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110751075A (en) | Remote sensing image culture pond detection method based on example segmentation | |
| CN109117802A (en) | Ship Detection towards large scene high score remote sensing image | |
| CN111582218A (en) | Remote sensing image island extraction method based on deep learning | |
| CN115035293B (en) | An unsupervised deep learning method for intelligent extraction of marine aquaculture from SAR images | |
| CN111310681B (en) | A remote sensing extraction method of mangrove distribution incorporating geoscience knowledge | |
| CN103020975A (en) | Wharf and ship segmentation method combining multi-source remote sensing image characteristics | |
| CN109064479B (en) | Sea-sky-line detection method based on gray dynamic features of adjacent video frames | |
| CN110956183A (en) | Asphalt pavement crack morphology extraction method | |
| CN115049834B (en) | Urban built-up area extraction method based on night light data and high-resolution image | |
| CN113920300A (en) | Visible light remote sensing image ship target identification method and system | |
| CN113408547B (en) | Multi-temporal multi-polarization SAR landslide extraction method | |
| CN110097549A (en) | Based on morphologic land, water and air boundary line detecting method, system, medium and equipment | |
| CN117408534A (en) | A short-term evaluation method for the salvage effect of Enteromorpha green tide based on satellite remote sensing | |
| CN105913425B (en) | A multi-pig contour extraction method based on adaptive ellipse block and wavelet transform | |
| CN107578414B (en) | A kind of processing method of pavement crack image | |
| CN117292272A (en) | Method for identifying large-range cultivation measures by combining high-resolution remote sensing image and open source DEM | |
| CN114627367B (en) | Sea bottom line detection method for side-scan sonar image | |
| CN119206518B (en) | A method for extracting floating ball cage-type sea surface aquaculture farms based on SAR images | |
| CN114241054B (en) | Image processing method, device, system and storage medium based on mesh light spot | |
| CN115620164A (en) | Extraction method, device, equipment and storage medium of a culture pond in the coastal zone | |
| CN119314049A (en) | Method for extracting temporal changes and detecting mutations in reservoir water area in plain farmland areas | |
| CN118521901A (en) | Remote sensing interpretation method for sea elements for cultivation | |
| CN118570475A (en) | Sea level segmentation method based on deep learning | |
| CN112613464A (en) | Plateau terrace extraction method based on unmanned aerial vehicle image matching point cloud | |
| CN101894373A (en) | Adopt the Weather fax image forward line extracting method of boundary rectangle |
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 |