CN109490301B - Method, system and storage medium for monitoring detection of attachments on floating platform - Google Patents
Method, system and storage medium for monitoring detection of attachments on floating platform Download PDFInfo
- Publication number
- CN109490301B CN109490301B CN201811243761.4A CN201811243761A CN109490301B CN 109490301 B CN109490301 B CN 109490301B CN 201811243761 A CN201811243761 A CN 201811243761A CN 109490301 B CN109490301 B CN 109490301B
- Authority
- CN
- China
- Prior art keywords
- floating platform
- image
- cylinder
- detecting
- arc
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N2021/8466—Investigation of vegetal material, e.g. leaves, plants, fruits
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a method for monitoring detection of attachments on a floating platform, which comprises the following steps: step 1: acquiring a floating platform picture through a camera and transmitting the floating platform picture to a computer unit of a floating platform; step 2: detecting the attachment of the floating platform by an intelligent image analysis algorithm; and carrying out intelligent analysis on the image transmitted to the floating platform calculating unit, and detecting whether attachments exist on the surface of the floating platform. The invention has the beneficial effects that: 1. the cost of the adopted camera is very low, and the manufacturing cost of the unmanned monitoring floating platform cannot be obviously increased; 2. the energy consumption of the camera is very low, and only images need to be shot at regular intervals, so that the energy is saved; 3. meanwhile, the three cameras are used for shooting the underwater part of the unmanned floating platform, so that dead angles are avoided, and monitoring can be performed in every region; 4. the images are automatically processed in the intelligent unit of the floating platform, and the growth condition of attachments is sensed, so that image analysis is not needed to be carried out in the background, and the information transmission quantity is small.
Description
Technical Field
The present invention relates to marine informatization, computer vision, and artificial intelligence, and more particularly, to a method, system, and storage medium for monitoring detection of attachments on a floating platform.
Background
The oceans occupy a large portion of the surface area of the earth. Human activities are not in the ocean. In recent years, with the increasing importance of the ocean interests of the countries, ocean informatization has become more important. By constructing the marine informatization system, various industries such as fishery, sea police, national defense, ocean voyage, meteorology, ecological environment monitoring, environmental protection, seawater planting industry and the like can obtain great benefits.
The small unmanned floating platform is used for detecting, monitoring and collecting relevant information on sea surface and sea water, and is a new monitoring means in recent years. Compared with other monitoring means, the unmanned floating platform has the following advantages:
1) manpower is not needed, and the cost is low.
2) The small unmanned floating platform has the advantages that the size of a visible part on the water surface is small, the visible part is not easy to find, and the concealment is high.
3) The service time is long. The product can be used for years after being thrown once. The energy consumption is low.
4) The collected information is rich in variety.
5) High safety and no influence on various facilities at the coast.
6) The monitoring range is large, the depth can be changed in the sea surface and the seawater, and various information can be collected.
7) Environmental protection and no fuel pollution to seawater.
In view of the above advantages, the investment of unmanned floating platforms is increased in all countries around the world, and it is expected to take a sufficient advantage in future marine competition.
The biggest problem of the current unmanned floating platforms is that whether attachments exist on the surfaces of the unmanned floating platforms or not can not be monitored, and the growth degree of the unmanned floating platforms is not monitored. This drawback, however, has the following serious consequences:
1) corroding the surface of the floating platform, affecting its life
2) The key parts of some sensors are possibly shielded, and the data acquisition precision is influenced.
3) The interfered data enters a background information system, so that the decision of a user is deviated, and a serious accident is caused.
4) The weight of the floating platform is increased, and the working state of the floating platform is influenced.
5) It is possible to carry the living things in the current sea area to other sea areas, causing problems such as invasion of living things.
6) The interference degree of the data cannot be known, so that whether the data of the current floating platform is discarded or not or whether the data is processed to a certain degree cannot be judged.
7) The inability to detect which of the floating platforms is a problem results in all floating platform data becoming untrustworthy, resulting in a breakdown of the entire system.
However, unmanned floating platforms still suffer from a large problem, surface attachment. Surface attachment is a common phenomenon in ocean-going shipping. It has serious influence on the service life and the navigation speed of the naval vessel. The problem is not solved, and a large amount of resources are required to be invested every year to clean the surface attachments.
Unmanned floating platforms are also deeply plagued by surface adhesion. Since the pontoons have no active power system, they are floating in the sea and move with the current, so that they are stationary for long periods of time relative to the sea. In a stationary situation, the growth rate of the attachments is significantly higher than that of a surface vessel which is often operated.
When the surface attachment grows to a certain extent, the working efficiency of the floating platform may be affected and even the deviation of the result may be caused. Therefore, in order to solve the problem, relevant personnel can know the growth state of the current attachment of each floating platform, and whether the data is credible or not and how credible the data is can be judged, so that the influence on the data credibility of the whole information system is prevented. For a floating platform with serious attachment growth, special processing is needed to carry out data. At the same time, it is necessary to recover and clean these floating platforms.
Disclosure of Invention
The invention provides a method for monitoring detection of attachments on a floating platform, which comprises the following steps:
step 1: acquiring a floating platform picture through a camera and transmitting the floating platform picture to a computer unit of a floating platform;
step 2: detecting the attachment of the floating platform by an intelligent image analysis algorithm; the image transmitted to the floating platform calculation unit is intelligently analyzed, and whether attachments exist on the surface of the floating platform is detected; the step 2 comprises the following steps of:
s1: storing an initial state image;
storing photos in advance, storing the photos in a template library for later intelligent analysis;
s2: collecting a floating platform image;
acquiring a required image through a camera on the floating platform;
s3: detecting floating platform attachments;
the acquired image is transmitted to a computing unit of the floating platform, and intelligent analysis is completed in the computing unit, so that whether the surface of the floating platform has attachments or not is detected.
As a further improvement of the present invention, the following information is extracted in step 2 for future analysis, and specifically includes:
the precise position of the side boundary of the column part of the floating platform and the precise position of the bottom arc; because the side boundary of the cylinder is a straight line, a straight line detection method is adopted for analysis to obtain boundary coordinates; then, the length of the column body is determined by detecting the arc at the bottom of the column body of the floating platform;
recording the size of the floating platform cylinder; calculating and recording the width and height dimensions of the floating platform cylinder;
color information of the floating platform cylinder part in seawater; according to the straight lines on the two sides of the floating platform, combining the circular arc at the bottom to obtain the accurate position of the column part of the floating platform, extracting the average color of all pixels in the area, and reserving the average color information;
brightness information of the surface of the floating platform cylinder in seawater;
surface texture information of the cylinder part of the floating platform; and describing the texture by using the LBP characteristics, recording a histogram of the LBP characteristics of the cylinder surface under the floating platform, and storing an original image of the LBP characteristics for subsequent analysis.
As a further improvement of the present invention, the method for detecting straight lines includes the following steps:
b1: graying the color image;
b2: detecting a canny edge in the gray level image;
b3: performing straight line fitting by using a Hough straight line detection algorithm;
and observing whether the straight lines on the two sides of the floating platform cylinder are parallel or are equal in length, if so, successfully detecting, and otherwise, acquiring the image again for detection.
As a further improvement of the invention, the detection of the bottom arc of the floating platform column comprises the following steps of:
c1: determining the approximate range of the bottom by utilizing straight lines on two sides of a floating platform cylinder;
c2: in the canny diagram, arc fitting is performed;
and observing whether the circular arc can be connected with the straight lines on the two sides, if so, detecting successfully, and otherwise, acquiring the image again for detection.
As a further improvement of the present invention, the step of S3 includes the following steps performed in sequence:
d1: differential comparison of the current map of the floating platform with the template map;
the first step of detecting the attachments is to directly make a difference between the current image acquired by the camera and the corresponding template image and watch the result of the difference image so as to detect whether the difference image has larger change;
d2: detecting and comparing the outer edge of the floating platform cylinder;
detecting a side straight line and a bottom arc in a current image;
d3: analyzing and comparing color and brightness;
comparing the average color of the current image with the color of the template image, if the difference value of any one of the three RGB channels is larger than a predetermined value, indicating that attachment is present, otherwise, entering the step D4;
if the difference between the average brightness of the current graph and the average brightness of the template graph is larger than the set value, indicating that the attachment exists, otherwise, entering the step D4;
d4: detecting and comparing the holes on the surface of the floating platform column;
detecting whether the floating platform is polluted or not by counting the number of the holes;
d5: detecting and comparing straight lines and circular arc lines on the surface of the floating platform cylinder;
detecting whether attached organisms exist or not by analyzing lines in the current image;
d6: detecting and comparing structures of sharp corners at two sides of the floating platform cylinder;
d7: analyzing and comparing textures of the cylinder regions of the floating platform;
calculating texture by using LBP characteristics in a floating platform cylinder area, calculating a histogram of the LBP characteristics on the surface of the floating platform cylinder, and then calculating the correlation degree of the histogram and the texture histogram of the original template image, wherein if the correlation degree is lower than a set value, the existence of attached organisms is indicated, and early warning is performed;
the specific detection method of step D5 includes the following steps:
g1: extracting the edge of canny in the column area;
g2: detecting straight line segments in the canny edge by adopting a hough straight line fitting method;
g3: filtering straight segments with shorter length;
g4: counting the number and average length of straight line segments in the image;
g5: detecting and fitting the arc segment in the canny edge by adopting a hough arc fitting method; the specific detection method of step D6 includes the following steps:
h1: separating the cylinder of the floating platform from the background image;
h2: extracting an external contour curve of the cylinder;
h3: traversing an external contour curve, and calculating a contour included angle of each point position;
h4: the included angle degree is smaller than the set degree, and the structure is regarded as a sharp corner structure;
h5: and counting the total number of sharp corners at two sides, wherein the total number exceeds a set number, and early warning.
As a further improvement of the present invention, the step D1 includes the following steps performed in sequence:
e1: respectively calculating absolute values of difference values of pixels at corresponding positions of the template image and the current image according to the RGB three color components to obtain a new color difference image;
e2: extracting a differential result of the column region;
e3: counting the sum of the difference values of all pixel positions and three channels in the cylinder area;
e4: a sum of the differential values, an average value being calculated with respect to the area of the cylinder region; if the average value is greater than the set threshold, it indicates that attachment is present, otherwise it is not processed.
As a further improvement of the present invention, the step D2 includes the following steps performed in sequence:
f1: calculating the width of a floating platform column body by using the position of a straight line on the side surface of the floating platform;
f2: comparing the new width value obtained in the above step with the width value of the floating platform column in the template, if the difference is larger than the set pixel, indicating that there is an attachment, otherwise, not processing;
f3: and comparing the distance between the bottom arc position and the arc position in the template drawing by using the bottom arc position, if the distance deviation is larger than the set pixel, indicating that the attachment exists, and otherwise, not processing.
As a further improvement of the invention, the specific detection method in the step D4 is as follows:
in grayscale images, filtering is performed using a DOG operator (difference gaussian operator), which is mathematically described as follows:
The invention also discloses a system for monitoring the detection of attachments on the floating platform, which comprises: memory, a processor and a computer program stored on the memory, the computer program being configured to carry out the steps of the method as claimed when invoked by the processor.
The invention also discloses a computer-readable storage medium, in which a computer program is stored which, when being invoked by a processor, is configured to carry out the steps of the method as claimed in the claims.
The invention has the beneficial effects that: 1. the cost of the adopted camera is very low, and the manufacturing cost of the unmanned monitoring floating platform cannot be obviously increased; 2. the energy consumption of the camera is low, and only images need to be shot at regular intervals (such as 3 days), so that energy is saved; 3. meanwhile, the three cameras are used for shooting the underwater part of the unmanned floating platform, so that dead angles are avoided, and monitoring can be performed in every region; 4. the images are automatically processed in the intelligent unit of the floating platform, and the growth condition of attachments is sensed, so that image analysis is not needed to be carried out in the background, and the information transmission quantity is small.
Drawings
FIG. 1 is a flow chart of an intelligent analysis algorithm for images of the present invention;
FIG. 2 is a schematic diagram of the floating platform attachment detection and comparison process of the present invention.
Detailed Description
As shown in fig. 1-2, the present invention discloses a method for monitoring detection of attachments on a floating platform, comprising the steps of:
step 1: acquiring a floating platform picture through a camera and transmitting the floating platform picture to a computer unit of a floating platform;
mounting position and angle of camera:
for the convenience of observation, a camera is arranged at the top position of the floating platform. The camera is angled down looking down on the entire floating platform portion. The three cameras are arranged in a circular ring shape and are equidistant, so that the visual field can be ensured to cover all the surfaces of the floating platforms. The camera has the functions of water drainage and corrosion prevention, and can normally work in seawater.
Considering that the illumination is insufficient at a deep position in the sea, the camera generally collects images when the floating platform floats to the position near the sea surface. At the moment, the image definition is higher, and the color identification degree is also higher.
The depth of the floating platform in the sea water can be sensed by a sensor carried by the floating platform. The camera can thus be informed to take a picture. In each shooting, the floating platform is ensured to be positioned at the same depth position as much as possible, so that the interference of seawater light transmission can be reduced.
The shooting time also needs to be kept consistent, and interference caused by inconsistent illumination intensity is prevented.
And shooting by utilizing a network communication system of the floating platform and combining weather information. When the illumination is insufficient in rainy days, the shooting is not performed. And the image shooting and detection are only carried out when the sunshine is sufficient on sunny days.
Step 2: detecting the attachment of the floating platform by an intelligent image analysis algorithm; the image shot by the camera can be transmitted to the computing unit of the floating platform, and then the image transmitted to the computing unit of the floating platform is intelligently analyzed to detect whether attachments exist on the surface of the floating platform;
as shown in fig. 2, step 2 includes performing the following steps in sequence:
s1: storing an initial state image;
storing photos in advance, storing the photos in a template library for later intelligent analysis;
when unmanned floating platform comes into use for the first time, perhaps just when accomplishing the clearance, store the photo in advance for the intelligent analysis in later stage, this picture needs to be used in the later stage and contrasts, and the initial image of three camera all needs to be stored to when just beginning to put in the use, can store some images more, store among the template storehouse.
S2: collecting a floating platform image;
acquiring a required image through a camera on the floating platform;
s3: detecting floating platform attachments;
the acquired image is transmitted to a computing unit of the floating platform, and intelligent analysis is completed in the computing unit, so that whether the surface of the floating platform has attachments or not is detected.
The following information is extracted in the step 2 for future analysis, and the following information is specifically extracted:
the precise position of the side boundary of the column part of the floating platform and the precise position of the bottom arc; because the side boundary of the cylinder is a straight line, a straight line detection method is adopted for analysis to obtain boundary coordinates; then, the length of the column body is determined by detecting the arc at the bottom of the column body of the floating platform;
recording the size of the floating platform cylinder; calculating and recording the width and height dimensions of the floating platform cylinder; the size of the column includes two aspects-the width of the column and the height of the column. The width of the cylinder can be calculated by using parallel straight lines on two sides. The height of the column body can be recorded only by recording the accurate position of the bottom arc.
Color information of the floating platform cylinder part in seawater; according to the straight lines on the two sides of the floating platform, combining the circular arc at the bottom to obtain the accurate position of the column part of the floating platform, extracting the average color of all pixels in the area, and reserving the average color information; in order to improve the accuracy, multiple times of shooting and detection can be performed at the same position. And the depth positions can be unified in different sea areas, and multiple times of shooting and detection can be realized.
Brightness information of the surface of the floating platform cylinder in seawater; in addition to the color information, the corresponding brightness information of the calculator is also required.
Surface texture information of the cylinder part of the floating platform; and describing the texture by using the LBP characteristics, recording a histogram of the LBP characteristics of the cylinder surface under the floating platform, and storing an original image of the LBP characteristics for subsequent analysis.
The method for detecting the straight line comprises the following steps of:
b1: graying the color image;
b2: detecting a canny edge in the gray level image; canny edge detection is a conventional method and is not described in detail.
B3: performing straight line fitting by using a Hough straight line detection algorithm; (not described in detail)
And observing whether the straight lines on the two sides of the floating platform cylinder are parallel or are equal in length, if so, successfully detecting, and otherwise, acquiring the image again for detection.
Under the overlooking view angle of the camera, the bottom boundary of the cylinder presents a circular arc shape. The detection of the circular arc at the bottom of the floating platform cylinder comprises the following steps of:
c1: determining the approximate range of the bottom by utilizing straight lines on two sides of a floating platform cylinder;
c2: in the canny diagram, arc fitting is performed; (Hough circle detection algorithm, not repeated) whether the arc can be connected with the straight lines on the two sides is observed, if so, the detection is successful, otherwise, the image is collected again for detection. By connected, it is meant that the end points of the arc are less than a threshold (e.g., 5 pixels) from the line on either side.
After the floating platform works for a long time, images are shot at regular intervals and are compared and analyzed with the originally stored template map. As shown in fig. 1, the step of S3 includes the following steps performed in sequence:
d1: differential comparison of the current map of the floating platform with the template map;
since the position of the camera is fixed, the position of the cylinder in the image does not change. The first step of detecting the attachments is to directly make a difference between the current image acquired by the camera and the corresponding template image and watch the result of the difference image so as to detect whether the difference image has larger change;
d2: detecting and comparing the outer edge of the floating platform cylinder;
detecting a side straight line and a bottom arc in a current image;
d3: analyzing and comparing color and brightness;
similarly calculating the color and brightness information of the cylinder area in the current image, comparing the average color of the current image with the color of the template image, if the difference value of any one of the three channels of RGB is greater than a set value of 20, indicating that an attachment is present, otherwise, entering the step D4;
if the difference between the average brightness of the current map and the average brightness of the template map is greater than the set value of 20, indicating that the attachment is present, the process proceeds to step D4;
d4: detecting and comparing the holes on the surface of the floating platform column;
holes are easy to appear on the attachments. And no holes exist on the smooth and clean cylinder. Detecting whether the floating platform is polluted or not by counting the number of the holes;
d5: detecting and comparing straight lines and circular arc lines on the surface of the floating platform cylinder;
many surface aids have smooth line structures, such as shells, barnacles, etc., and clearly border curves can be seen. These lines are absent on the surface of the float where there is no attachment. Detecting whether attached organisms exist or not by analyzing lines in the current image;
d6: detecting and comparing structures of sharp corners at two sides of the floating platform cylinder;
when the attached organisms appear on the floating platform, the sharp-angled structures are easily observed at the boundaries of the two sides of the cylinder image. They are caused by the external curvature of organisms such as shells, barnacles, etc.
D7: analyzing and comparing textures of the cylinder regions of the floating platform;
calculating texture by using LBP (local binary pattern) characteristics in a floating platform cylinder area, calculating a histogram of the LBP characteristics on the surface of the floating platform cylinder, and then calculating the correlation degree of the histogram and the texture histogram of the original template image, wherein if the correlation degree is lower than a set value of 0.7, the existence of attached organisms is indicated, and early warning is performed; the calculation method of the correlation degree is known knowledge and is not described in detail.
The specific detection method of step D5 includes the following steps:
g1: extracting the edge of canny in the column area;
g2: detecting straight line segments in the canny edge by adopting a hough straight line fitting method; also, hough line fitting was used.
G3: filtering straight segments with shorter length; straight line segments less than 50 pixels in length are filtered out.
G4: counting the number and average length of straight line segments in the image; if the number of the filter screen exceeds 3, and the average length exceeds 80 pixels, early warning is needed.
G5: detecting and fitting the arc segment in the canny edge by adopting a hough arc fitting method; the method used was hough arc fitting. The arc segments with a length of less than 50 pixels are filtered. And counting the number and the average length of the arcs in the image. If the number of the filter screen exceeds 3, and the average length exceeds 80 pixels, early warning is needed.
The specific detection method of step D6 includes the following steps:
h1: separating the cylinder of the floating platform from the background image; the steps are simple, and because the areas outside the columns are all seawater, the color and the texture are simple.
H2: extracting an external contour curve of the cylinder;
h3: traversing an external contour curve, and calculating a contour included angle of each point position; the included angle is calculated by taking two points in front of and behind the contour sequence at intervals of 20 pixel positions. The two points form an included angle with the current contour point, and the degrees of the two points are calculated.
H4: the included angle degree is smaller than the set degree of 135 degrees, and the structure is regarded as a sharp corner structure;
h5: and counting the total number of sharp corners at two sides, wherein the total number exceeds a set number of 4, and early warning.
The step D1 includes the following steps:
e1: respectively calculating absolute values of difference values of pixels at corresponding positions of the template image and the current image according to the RGB three color components to obtain a new color difference image;
e2: extracting a differential result of the column region;
e3: counting the sum of the difference values of all pixel positions and three channels in the cylinder area;
e4: a sum of the differential values, an average value being calculated with respect to the area of the cylinder region; if the average value is greater than a set threshold (default 20), it indicates that attachment is present, otherwise it is not processed.
The step D2 includes the following steps:
f1: calculating the width of a floating platform column body by using the position of a straight line on the side surface of the floating platform;
f2: comparing the new width value obtained in the above step with the width value of the floating platform column in the template, if the difference is greater than the set pixel, namely 10 pixels, indicating that there is an attachment, otherwise, not processing; f3: and comparing the distance with the arc position in the template drawing by using the bottom arc position, and if the distance deviation is larger than the set pixel, namely 5 pixels, indicating that the attachment exists, otherwise, not processing.
The specific detection method in the step D4 is as follows:
in grayscale images, filtering is performed using a DOG operator (difference gaussian operator), which is mathematically described as follows:
The invention also discloses a system for monitoring the detection of attachments on the floating platform, which comprises: memory, a processor and a computer program stored on the memory, the computer program being configured to carry out the steps of the method as claimed when invoked by the processor.
The invention also discloses a computer-readable storage medium, in which a computer program is stored which, when being invoked by a processor, is configured to carry out the steps of the method as claimed in the claims.
The invention provides a strategy for monitoring the underwater part of a floating platform by using an underwater camera. And (4) sensing the growth degree of the attachments on the upper surface of the current floating platform by utilizing automatic analysis of the floating platform image.
Brief introduction of the shape of the floating platform: for ease of understanding, an Argo buoy is used as an example for description.
The external shape of the unmanned surveillance floating platform is an elongated cylinder. In the working state, the cylinder is vertically floated in the seawater, and a part of the cylinder is exposed out of the water surface or is completely submerged in the seawater.
The invention has the beneficial effects that: 1. the cost of the adopted camera is very low, and the manufacturing cost of the unmanned monitoring floating platform cannot be obviously increased; 2. the energy consumption of the camera is low, and only images need to be shot at regular intervals (such as 3 days), so that energy is saved; 3. meanwhile, the three cameras are used for shooting the underwater part of the unmanned floating platform, so that dead angles are avoided, and monitoring can be performed in every region; 4. the images are automatically processed in the intelligent unit of the floating platform, and the growth condition of attachments is sensed, so that image analysis is not needed to be carried out in the background, and the information transmission quantity is small.
The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.
Claims (9)
1. A method for monitoring detection of fouling on a floating platform, the method comprising the steps of:
step 1: acquiring a floating platform picture through a camera and transmitting the floating platform picture to a computer unit of a floating platform;
step 2: detecting the attachment of the floating platform by an intelligent image analysis algorithm; the image transmitted to the floating platform calculation unit is intelligently analyzed, and whether attachments exist on the surface of the floating platform is detected;
the step 2 comprises the following steps of:
s1: storing an initial state image;
storing photos in advance, storing the photos in a template library for later intelligent analysis;
s2: collecting a floating platform image;
acquiring a required image through a camera on the floating platform;
s3: detecting floating platform attachments;
transmitting the acquired image to a computing unit of the floating platform, and completing intelligent analysis in the computing unit so as to detect whether attachments exist on the surface of the floating platform;
the step of S3 includes the following steps performed in sequence:
d1: differential comparison of the current map of the floating platform with the template map;
the first step of detecting the attachments is to directly make a difference between the current image acquired by the camera and the corresponding template image and watch the result of the difference image so as to detect whether the difference image has larger change;
d2: detecting and comparing the outer edge of the floating platform cylinder;
detecting a side straight line and a bottom arc in a current image;
d3: analyzing and comparing color and brightness;
comparing the average color of the current image with the color of the template image, if the difference value of any one of the three RGB channels is larger than a predetermined value, indicating that attachment is present, otherwise, entering the step D4;
if the difference between the average brightness of the current graph and the average brightness of the template graph is larger than the set value, indicating that the attachment exists, otherwise, entering the step D4;
d4: detecting and comparing the holes on the surface of the floating platform column;
detecting whether the floating platform is polluted or not by counting the number of the holes;
d5: detecting and comparing straight lines and circular arc lines on the surface of the floating platform cylinder;
detecting whether attached organisms exist or not by analyzing lines in the current image;
d6: detecting and comparing structures of sharp corners at two sides of the floating platform cylinder;
d7: analyzing and comparing textures of the cylinder regions of the floating platform;
calculating texture by using LBP characteristics in a floating platform cylinder area, calculating a histogram of the LBP characteristics on the surface of the floating platform cylinder, and then calculating the correlation degree of the histogram and the texture histogram of the original template image, wherein if the correlation degree is lower than a set value, the existence of attached organisms is indicated, and early warning is performed;
the specific detection method of step D5 includes the following steps:
g1: extracting the edge of canny in the column area;
g2: detecting straight line segments in the canny edge by adopting a hough straight line fitting method;
g3: filtering straight segments with shorter length;
g4: counting the number and average length of straight line segments in the image;
g5: detecting and fitting the arc segment in the canny edge by adopting a hough arc fitting method;
the specific detection method of step D6 includes the following steps:
h1: separating the cylinder of the floating platform from the background image;
h2: extracting an external contour curve of the cylinder;
h3: traversing an external contour curve, and calculating a contour included angle of each point position;
h4: the included angle degree is smaller than the set degree, and the structure is regarded as a sharp corner structure;
h5: and counting the total number of sharp corners at two sides, wherein the total number exceeds a set number, and early warning.
2. The method according to claim 1, wherein the following information is extracted in step 2 for later analysis, and specifically comprises:
the precise position of the side boundary of the column part of the floating platform and the precise position of the bottom arc; because the side boundary of the cylinder is a straight line, a straight line detection method is adopted for analysis to obtain boundary coordinates; then, the length of the column body is determined by detecting the arc at the bottom of the column body of the floating platform;
recording the size of the floating platform cylinder; calculating and recording the width and height dimensions of the floating platform cylinder;
color information of the floating platform cylinder part in seawater; according to the straight lines on the two sides of the floating platform, combining the circular arc at the bottom to obtain the accurate position of the column part of the floating platform, extracting the average color of all pixels in the area, and reserving the average color information;
brightness information of the surface of the floating platform cylinder in seawater;
surface texture information of the cylinder part of the floating platform; and describing the texture by using the LBP characteristics, recording a histogram of the LBP characteristics of the cylinder surface under the floating platform, and storing an original image of the LBP characteristics for subsequent analysis.
3. The method according to claim 2, wherein the method of line detection comprises the following steps performed in sequence:
b1: graying the color image;
b2: detecting a canny edge in the gray level image;
b3: performing straight line fitting by using a Hough straight line detection algorithm;
and observing whether the straight lines on the two sides of the floating platform cylinder are parallel or are equal in length, if so, successfully detecting, and otherwise, acquiring the image again for detection.
4. The method of claim 2, wherein the detection of the circular arc of the bottom of the floating platform column comprises the following steps performed in sequence:
c1: determining the approximate range of the bottom by utilizing straight lines on two sides of a floating platform cylinder;
c2: in the canny diagram, arc fitting is performed;
and observing whether the circular arc can be connected with the straight lines on the two sides, if so, detecting successfully, and otherwise, acquiring the image again for detection.
5. The method of claim 1, wherein the step D1 comprises the steps of:
e1: respectively calculating absolute values of difference values of pixels at corresponding positions of the template image and the current image according to the RGB three color components to obtain a new color difference image;
e2: extracting a differential result of the column region;
e3: counting the sum of the difference values of all pixel positions and three channels in the cylinder area;
e4: a sum of the differential values, an average value being calculated with respect to the area of the cylinder region; if the average value is greater than the set threshold, it indicates that attachment is present, otherwise it is not processed.
6. The method of claim 1, wherein the step D2 comprises the steps of:
f1: calculating the width of a floating platform column body by using the position of a straight line on the side surface of the floating platform;
f2: comparing the new width value obtained in the above step with the width value of the floating platform column in the template, if the difference is larger than the set pixel, indicating that there is an attachment, otherwise, not processing;
f3: and comparing the distance between the bottom arc position and the arc position in the template drawing by using the bottom arc position, if the distance deviation is larger than the set pixel, indicating that the attachment exists, and otherwise, not processing.
7. The method according to claim 1, wherein the specific detection method in the step D4 is as follows:
in grayscale images, filtering is performed using a DOG operator (difference gaussian operator), which is mathematically described as follows:
8. A system for monitoring detection of fouling on a floating platform, comprising: memory, a processor and a computer program stored on the memory, the computer program being configured to carry out the steps of the method of any one of claims 1-7 when invoked by the processor.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program configured to, when invoked by a processor, implement the steps of the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811243761.4A CN109490301B (en) | 2018-10-24 | 2018-10-24 | Method, system and storage medium for monitoring detection of attachments on floating platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811243761.4A CN109490301B (en) | 2018-10-24 | 2018-10-24 | Method, system and storage medium for monitoring detection of attachments on floating platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109490301A CN109490301A (en) | 2019-03-19 |
| CN109490301B true CN109490301B (en) | 2021-06-11 |
Family
ID=65690864
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811243761.4A Active CN109490301B (en) | 2018-10-24 | 2018-10-24 | Method, system and storage medium for monitoring detection of attachments on floating platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109490301B (en) |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002335803A (en) * | 2001-05-16 | 2002-11-26 | Suga Marin Mechanic:Kk | Monitaring device for artificial fish reef installed at or deeper than 60 m water depth |
| CN1293836C (en) * | 2005-01-06 | 2007-01-10 | 重庆大学 | Lines feature on-line identifying method for detecting foreign matter in food |
| CN1806501B (en) * | 2005-01-17 | 2011-08-24 | 厦门市汇阳科技有限公司 | Marine phytoplankton automatic distinguishing method and apparatus |
| CN102243323B (en) * | 2010-05-10 | 2013-05-15 | 中南大学 | Rock mass slope stability analyzing method based on video detection |
| CN103147903B (en) * | 2013-02-04 | 2015-04-29 | 郑贵林 | Sea observation platform for unmanned and automatic navigation |
| CN103412345B (en) * | 2013-08-16 | 2016-01-06 | 中国舰船研究设计中心 | Automatic aircraft carrier flight deck foreign matter detection recognition system |
| CN105225251B (en) * | 2015-09-16 | 2017-11-24 | 三峡大学 | Over the horizon movement overseas target based on machine vision quickly identifies and positioner and method |
| CN106845346A (en) * | 2016-12-16 | 2017-06-13 | 北京无线电计量测试研究所 | A kind of image detecting method for airfield runway foreign bodies detection |
| CN107481233A (en) * | 2017-08-22 | 2017-12-15 | 广州辰创科技发展有限公司 | A kind of image-recognizing method being applied in FOD foreign bodies detection radars |
| CN108022453A (en) * | 2017-12-07 | 2018-05-11 | 深圳云天励飞技术有限公司 | Barrier intrusion detection method, system, terminal and storage medium |
-
2018
- 2018-10-24 CN CN201811243761.4A patent/CN109490301B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109490301A (en) | 2019-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11591762B2 (en) | Method for tracking, monitoring and evaluating ecological impact of channel project based on long-term time series satellite remote sensing data | |
| CN109815807B (en) | Method for detecting ship in shore based on edge line analysis and aggregation channel characteristics | |
| CN104217215B (en) | A kind of classifying identification method of water surface foggy image and picture rich in detail | |
| CN106770087B (en) | Greasy dirt remote sensing module, system and method | |
| CN104933680B (en) | A kind of intelligent quick sea fog minimizing technology of unmanned boat vision system video | |
| CN109472200B (en) | Intelligent sea surface rubbish detection method, system and storage medium | |
| CN109409283A (en) | A kind of method, system and the storage medium of surface vessel tracking and monitoring | |
| Bonin-Font et al. | Visual discrimination and large area mapping of posidonia oceanica using a lightweight auv | |
| CN108229342A (en) | A kind of surface vessel target automatic testing method | |
| CN110084792B (en) | Anomaly detection method for net of marine aquaculture net cage | |
| CN111027446A (en) | Coastline automatic extraction method of high-resolution image | |
| CN111292371A (en) | Intelligent water level detection equipment, intelligent water level monitoring management device and detection method | |
| CN108985274A (en) | Water surface method for recognizing impurities | |
| CN103971370A (en) | Intelligent ocean oil spill detection method for remote sensing large image | |
| CN111597930A (en) | Coastline extraction method based on remote sensing cloud platform | |
| CN108288268B (en) | Method for detecting ship landing in high-resolution remote sensing image based on projection analysis | |
| Isiacik Colak et al. | Coastline zone extraction using Landsat-8 OLI imagery, case study: Bodrum Peninsula, Turkey | |
| CN114359282A (en) | Multi-view-angle-fused power transmission line bird nest defect identification method and device | |
| CN109490301B (en) | Method, system and storage medium for monitoring detection of attachments on floating platform | |
| CN114140484A (en) | High-robustness sea-sky-line extraction method based on photoelectric sensor | |
| CN117911885A (en) | Red tide detection method, system, medium, computer equipment and terminal | |
| CN108447045B (en) | SAR remote sensing image water area detection method based on SAT integral graph | |
| Feng et al. | Real-time riverbank line detection for USV system | |
| CN104820818A (en) | Fast recognition method for moving object | |
| CN113052117B (en) | AIS data-assisted Rayleigh CFAR detection method for SAR image |
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 | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A method, system, and storage medium for detecting attachments on a floating platform Effective date of registration: 20230906 Granted publication date: 20210611 Pledgee: Shanghai Pudong Development Bank Co.,Ltd. Shenzhen Branch Pledgor: SHENZHEN JINRUN DEFENSE TECHNOLOGY Co.,Ltd. Registration number: Y2023980055371 |