CN113436162A - Method and device for identifying weld defects on surface of hydraulic oil pipeline of underwater robot - Google Patents

Method and device for identifying weld defects on surface of hydraulic oil pipeline of underwater robot Download PDF

Info

Publication number
CN113436162A
CN113436162A CN202110697224.2A CN202110697224A CN113436162A CN 113436162 A CN113436162 A CN 113436162A CN 202110697224 A CN202110697224 A CN 202110697224A CN 113436162 A CN113436162 A CN 113436162A
Authority
CN
China
Prior art keywords
image
hydraulic oil
oil pipeline
gray level
sub
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110697224.2A
Other languages
Chinese (zh)
Other versions
CN113436162B (en
Inventor
周天
肖志伟
吕冰冰
杨睿
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan Guotian Electronic Technology Co ltd
Original Assignee
Hunan Guotian Electronic Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hunan Guotian Electronic Technology Co ltd filed Critical Hunan Guotian Electronic Technology Co ltd
Priority to CN202110697224.2A priority Critical patent/CN113436162B/en
Publication of CN113436162A publication Critical patent/CN113436162A/en
Application granted granted Critical
Publication of CN113436162B publication Critical patent/CN113436162B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/40—Image enhancement or restoration using histogram techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G06T7/41—Analysis of texture based on statistical description of texture
    • G06T7/45—Analysis of texture based on statistical description of texture using co-occurrence matrix computation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Image Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Molecular Biology (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Quality & Reliability (AREA)
  • Probability & Statistics with Applications (AREA)
  • Biophysics (AREA)

Abstract

The invention relates to the technical field of defect identification, and discloses a method for identifying weld defects on the surface of a hydraulic oil pipeline of an underwater robot, which comprises the following steps: acquiring a hydraulic oil pipeline image, and performing image graying and grayscale stretching pretreatment on the hydraulic oil pipeline image to obtain a hydraulic oil pipeline grayscale image; carrying out image enhancement processing on the gray level image of the hydraulic oil pipeline by using an image enhancement strategy; segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images; performing characteristic parameter extraction processing on the subimages by utilizing a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix to obtain the image characteristics of the subimages; and (3) taking the image characteristics of the sub-images as the input of a convolutional neural network, and identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network. The invention also provides a device for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot. The invention realizes the defect identification of the hydraulic oil pipeline.

Description

Method and device for identifying weld defects on surface of hydraulic oil pipeline of underwater robot
Technical Field
The invention relates to the technical field of defect identification, in particular to a method and a device for identifying weld defects on the surface of a hydraulic oil pipeline of an underwater robot.
Background
In order to ensure safe operation and operation of the robot body after submergence, detection of weld defect identification on the surface of the hydraulic oil pipeline needs to be carried out. Common detection methods include visual methods and pressure testing methods. The visual method has low detection efficiency, is greatly influenced by subjective factors of detection personnel, and is easy to have missed detection or wrong detection; the pressure test method has long preparation period of test conditions, consumes time and labor and has high detection cost. If unstable factors exist in the weld joint on the surface of the hydraulic oil pipeline, serious accidents such as shutdown and sinking of the underwater robot body can be caused. The performance of the hydraulic pipeline is related to navigation and control of the underwater robot after launching.
In view of this, how to automatically extract characteristic parameters of weld surface defect representation and accurately identify the weld surface defects of the hydraulic oil pipeline becomes a problem to be solved by technical personnel in the field.
Disclosure of Invention
The invention provides a method for identifying weld defects on the surface of a hydraulic oil pipeline of an underwater robot, which comprises the steps of carrying out image enhancement processing on a collected hydraulic oil pipeline image by using an image enhancement strategy, and segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images; and performing characteristic parameter extraction processing on the sub-images by using a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix, processing the extracted characteristic parameters by using a convolutional neural network, and identifying and confirming the positions of the defect points.
In order to achieve the purpose, the invention provides a method for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot, which comprises the following steps:
acquiring a hydraulic oil pipeline image, and performing image graying and grayscale stretching pretreatment on the hydraulic oil pipeline image to obtain a hydraulic oil pipeline grayscale image;
carrying out image enhancement processing on the gray level image of the hydraulic oil pipeline by using an image enhancement strategy;
segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images;
performing characteristic parameter extraction processing on the subimages by utilizing a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix to obtain the image characteristics of the subimages;
and (3) taking the image characteristics of the sub-images as the input of a convolutional neural network, and identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network.
Optionally, the preprocessing of performing image graying and grayscale stretching on the hydraulic oil pipeline image includes:
solving the maximum value of three components of each pixel in the hydraulic oil pipeline image, setting the maximum value as the gray value of the pixel point, and obtaining the gray map of the hydraulic oil pipeline image, wherein the formula of the graying treatment is as follows:
G(i,j)＝max{R(i,j),G(i,j),B(i,j)}
wherein:
(i, j) is a pixel point in the hydraulic oil pipeline image;
r (i, j), G (i, j) and B (i, j) are respectively the values of the pixel point (i, j) in R, G, B three color channels;
g (i, j) is the gray value of the pixel point (i, j);
for the gray level image of the hydraulic oil pipeline, stretching the gray level of the image in a piecewise linear transformation mode, wherein the formula of the gray level stretching is as follows:
Figure BDA0003128335950000011
wherein:
f (x, y) is a gray scale map;
MAXf(x,y),MINf(x,y)respectively the maximum and minimum grey values of the grey map.
Optionally, the image enhancement policy flow is:
1) constructing a Gaussian filter kernel function matrix, and performing convolution operation on the Gaussian filter kernel function matrix and the hydraulic oil pipeline gray level image to obtain the hydraulic oil pipeline gray level image after Gaussian filtering; in one embodiment of the present invention, the constructed gaussian filter kernel function matrix is:
Figure BDA0003128335950000026
2) the histogram equalization processing is carried out on the gray level image of the hydraulic oil pipeline, and the method comprises the following steps:
counting the number of pixels corresponding to each gray level of the gray level image of the hydraulic oil pipeline to obtain a histogram of the image:
Figure BDA0003128335950000021
wherein:
k is 0,1, …, L-1, representing the gray level of the image;
nkrepresents the number of pixels whose gray level is k;
n represents the total number of pixels of the hydraulic oil pipeline gray level image;
calculating a cumulative histogram of the gray level image of the hydraulic oil pipeline:
Figure BDA0003128335950000022
k＝0，1，…，L-1
mapping the cumulative histogram to a gray scale range L0，Lk]：
S＝L0+(Lk-L0)c(k)
Counting the number of pixels of each gray level in S after mapping transformation to obtain a new image histogram;
3) correcting the gray level image color of the hydraulic oil pipeline equalized by the histogram by adopting a self-adaptive underwater image color correction algorithm, wherein the formula of the self-adaptive underwater image color correction is as follows:
Figure BDA0003128335950000023
wherein:
i (R, G and B) represents the sum of the gray images of the hydraulic oil pipeline in three color channels of R, G and B;
μ represents the minkowski distance mean of the image color channel;
alpha represents the maximum value of the gray level image of the hydraulic oil pipeline in three color channels of R, G and B;
beta is a correction parameter, the closer the value is to 0, the higher the brightness of the corrected image is, and the value is set to 0.2;
4) carrying out image brightness enhancement processing on the gray level image of the hydraulic oil pipeline after color correction by using an underwater image brightness enhancement algorithm based on image gradient, wherein the formula of the image brightness enhancement is as follows:
Figure BDA0003128335950000024
Figure BDA0003128335950000025
Figure BDA0003128335950000027
wherein:
[Tmin,Tmax]representing a range of image brightness enhancement;
s (x, y) represents a luminance value of the image pixel (x, y);
e (x, y) represents the luminance value of the enhanced image pixel (x, y);
i represents a gradient direction;
Figure BDA0003128335950000028
representing the partial derivatives in different gradient directions;
qirepresenting the target gradient in different gradient directions;
wi(x, y) represents the influence weight of the gradient residual in different gradient directions, a represents the sensitivity of the gradient residual, and its value range is [ -0.8,2]In one embodiment of the invention, the value is 0.6;
g (x, y) is a luminance constraint function, limiting the enhanced luminance within the target range.
Optionally, the segmenting the enhanced image by using the image segmentation network includes:
the image segmentation network is MASK R-CNN, the neural network adopts an FPN pyramid structure, and Resnet-101 is used as a convolution network to output sub-images with different sizes;
the objective function of the image segmentation network is as follows:
Figure BDA0003128335950000031
wherein:
t' is the predicted image segmentation boundary;
t is a binarization result of the image segmentation boundary;
m is an input image;
r is a segmentation image;
d (t) represents the distance transformation of the segmentation frame, i.e. the distance map between different segmentation images;
and inputting the enhanced image into an image segmentation network, and performing image segmentation according to the segmentation boundary t' obtained by prediction to obtain a plurality of sub-images.
Optionally, the performing, by using a feature parameter extraction algorithm based on a gray level co-occurrence matrix, feature parameter extraction processing on the sub-image includes:
1) for a sub-image f (x, y) of size M × N, there are k gray levels, where the distance between any two pixels i and pixel j is
Figure BDA0003128335950000032
And the included angle formed by the two pixel connecting lines and the same coordinate axis is theta, the gray level co-occurrence matrix value of the image pixel i and the image pixel j is Pij(d, θ), the gray level co-occurrence matrix of the sub-images is:
Figure BDA0003128335950000033
2) extracting angular second moment features in the gray level co-occurrence matrix:
Figure BDA0003128335950000034
the angle second moment feature represents the square sum of each element in the arrangement combination of texture elements in the co-occurrence matrix; when the ASM value is large, the texture presented by the image is rough and the energy is large; on the contrary, the image has fine texture and small energy;
3) extracting entropy characteristics in the gray level co-occurrence matrix:
Figure BDA0003128335950000035
the entropy characteristics are used for expressing the information quantity expressed by the two-dimensional gray level image and representing the complexity of the vein texture in the image; when Ent approaches 0, it represents that there is almost no texture information in the grayscale image; if Ent is larger, the presented image context tends to be more complex;
4) extracting contrast characteristics in the gray level co-occurrence matrix:
Figure BDA0003128335950000036
the contrast characteristic represents the definition of texture venation of the gray level image and the depth degree of the groove; when the contrast is high, the image is clearer, and the grooves are deeper; otherwise, the image is fuzzy;
5) extracting clustering shadow features in the gray level co-occurrence matrix:
Figure BDA0003128335950000041
Figure BDA0003128335950000042
Figure BDA0003128335950000043
and for each sub-image, taking the features extracted from the gray level co-occurrence matrix as the features of the sub-image.
Optionally, the identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network includes:
taking the image characteristics of the sub-images as the input of a convolutional neural network; the convolutional neural network model consists of an input layer, a convolutional layer, a pooling layer, a full-link layer and an activation function layer; the convolutional layer is used for feature extraction, and the formula is as follows:
Figure BDA0003128335950000044
wherein:
Figure BDA0003128335950000045
an ith feature map representing an nth layer;
f () represents an activation function, which in one embodiment of the present invention is a ReLU activation function;
m represents a set of input sub-images;
Figure BDA0003128335950000046
j features representing the n-1 st layer;
Figure BDA0003128335950000047
representing a convolution kernel between the ith feature map of the nth layer and the jth feature map connection of the n-1 layer;
"+" represents the convolution operation;
Figure BDA0003128335950000048
a bias representing an ith characteristic of the nth layer;
the pooling layer performs feature compression on the input feature graph, so that the network complexity is simplified; and the full connection layer maps the acquired features to the original sample mark space and sends the output value to the classifier.
In the transmission process, input information is processed layer by layer from an input layer through a hidden unit layer and is transmitted to an output layer, and the output of each layer of neurons only influences the input of the next layer of neurons; if the output layer can not obtain the expected output, the reverse propagation is carried out, and the weight and the threshold of the network are learned and corrected through an error reverse propagation algorithm;
and when the mean square error of the network is smaller than a given value, entering NMS algorithm operation, and determining the position of the defect point, namely finishing the surface defect identification of the hydraulic oil pipeline.
In addition, in order to achieve the above object, the present invention also provides an underwater robot hydraulic oil pipeline surface weld defect recognition apparatus, including:
the image acquisition device is used for acquiring an image of the hydraulic oil pipeline;
the data processor is used for preprocessing the image graying and the gray stretching of the hydraulic oil pipeline image and carrying out image enhancement processing on the hydraulic oil pipeline gray image by utilizing an image enhancement strategy; segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images;
the defect identification device is used for extracting the characteristic parameters of the sub-images by utilizing a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix to obtain the image characteristics of the sub-images; and (3) taking the image characteristics of the sub-images as the input of a convolutional neural network, and identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network.
In addition, to achieve the above object, the present invention also provides a computer readable storage medium having stored thereon defect identification program instructions, which are executable by one or more processors to implement the steps of the method for implementing surface weld defect identification of a hydraulic oil pipeline of an underwater robot as described above.
Compared with the prior art, the invention provides a method for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot, which has the following advantages:
firstly, the invention provides an image enhancement strategy, aiming at the problem that the red tone of an underwater image is weakened and the blue-green tone is the main tone, therefore, the invention adopts a self-adaptive underwater image color correction algorithm to carry out color correction on the gray level image of a hydraulic oil pipeline, and the formula of the self-adaptive underwater image color correction is as follows:
Figure BDA0003128335950000051
wherein: i (R, G and B) represents the sum of the gray images of the hydraulic oil pipeline in three color channels of R, G and B; mu represents the minkowski distance mean of the image color channel, since the distance between the red channel and the other color channels is large, so for underwater images mu > 1,
Figure BDA0003128335950000052
alpha represents the maximum value of the gray level image of the hydraulic oil pipeline in three color channels of R, G and B; beta is a correction parameter, the closer the value is to 0, the higher the brightness of the corrected image is, the set value is 0.2, and compared with the traditional algorithm, the algorithm disclosed by the invention has the advantages that the images are in three R, G and BThe sum of the color channels is used for color correction, the color values of other color channels are reduced under the condition that the distance between the red color channel and other color channels is large, and the correction parameters are used for carrying out integral color brightness enhancement on the image. Meanwhile, the invention utilizes an underwater image brightness enhancement algorithm based on image gradient to carry out image brightness enhancement processing on the gray level image of the hydraulic oil pipeline after color correction, and the formula of the image brightness enhancement is as follows:
Figure BDA0003128335950000053
Figure BDA0003128335950000054
Figure BDA0003128335950000059
wherein: [ T ]min,Tmax]Representing a range of image brightness enhancement; s (x, y) represents a luminance value of the image pixel (x, y); e (x, y) represents the luminance value of the enhanced image pixel (x, y); i represents a gradient direction;
Figure BDA00031283359500000510
representing the partial derivatives in different gradient directions; q. q.siRepresenting the target gradient in different gradient directions; w is ai(x, y) represents the influence weight of the gradient residual in different gradient directions, a represents the sensitivity of the gradient residual, and its value range is [ -0.8,2]To (c) to (d); g (x, y) is a luminance constraint function, limiting the enhanced luminance within the target range.
Meanwhile, the invention utilizes a characteristic parameter extraction algorithm based on a gray level co-occurrence matrix to extract the characteristic parameters of the sub-images, and as for the sub-images f (x, y), the size of the sub-images is MxN, k gray levels exist, wherein the distance between any two pixels i and pixel j is
Figure BDA0003128335950000055
And the included angle formed by the two pixel connecting lines and the same coordinate axis is theta, the gray level co-occurrence matrix value of the image pixel i and the image pixel j is Pij(d, θ), the gray level co-occurrence matrix of the sub-images is:
Figure BDA0003128335950000056
extracting angular second moment features in the gray level co-occurrence matrix:
Figure BDA0003128335950000057
the angle second moment feature represents the square sum of each element in the arrangement combination of texture elements in the co-occurrence matrix; when the ASM value is large, the texture presented by the image is rough and the energy is large; on the contrary, the image has fine texture and small energy; extracting entropy characteristics in the gray level co-occurrence matrix:
Figure BDA0003128335950000058
the entropy characteristics are used for expressing the information quantity expressed by the two-dimensional gray level image and representing the complexity of the vein texture in the image; when Ent approaches 0, it represents that there is almost no texture information in the grayscale image; if Ent is larger, the presented image context tends to be more complex; extracting contrast characteristics in the gray level co-occurrence matrix:
Figure BDA0003128335950000061
the contrast characteristic represents the definition of texture venation of the gray level image and the depth degree of the groove; when the contrast is high, the image is clearer, and the grooves are deeper; otherwise, the image is fuzzy; extracting clustering shadow features in the gray level co-occurrence matrix:
Figure BDA0003128335950000062
Figure BDA0003128335950000063
Figure BDA0003128335950000064
for each sub-image, taking the features extracted from the gray level co-occurrence matrix as the features of the sub-image; and according to the extracted characteristics representing the uniformity of the gray level distribution of the image and the thickness degree of the texture, utilizing a convolutional neural network to identify the defects of the weld joint on the surface of the hydraulic oil pipeline.
Drawings
Fig. 1 is a schematic flow chart of a method for identifying a weld defect on a surface of a hydraulic oil pipeline of an underwater robot according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a device for identifying a weld defect on a surface of a hydraulic oil pipeline of an underwater robot according to an embodiment of the present invention;
fig. 3 is a schematic device diagram of a device for identifying weld defects on the surface of a hydraulic oil pipeline of an underwater robot according to an embodiment of the present invention;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Carrying out image enhancement processing on the acquired hydraulic oil pipeline image by using an image enhancement strategy, and segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images; and performing characteristic parameter extraction processing on the sub-images by using a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix, processing the extracted characteristic parameters by using a convolutional neural network, and identifying and confirming the positions of the defect points. Referring to fig. 1, a schematic diagram of a method for identifying a weld defect on a surface of a hydraulic oil pipeline of an underwater robot according to an embodiment of the present invention is shown.
In this embodiment, the method for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot comprises the following steps:
and S1, acquiring the hydraulic oil pipeline image, and performing image graying and grayscale stretching pretreatment on the hydraulic oil pipeline image to obtain a hydraulic oil pipeline grayscale image.
Firstly, the invention uses a visual detection device of an underwater robot to obtain a hydraulic oil pipeline image, and carries out preprocessing of image graying and gray level stretching on the hydraulic oil pipeline image, wherein the flow of the image graying and gray level stretching is as follows:
solving the maximum value of three components of each pixel in the hydraulic oil pipeline image, setting the maximum value as the gray value of the pixel point, and obtaining the gray map of the hydraulic oil pipeline image, wherein the formula of the graying treatment is as follows:
G(i,j)＝max{R(i,j),G(i,j),B(i,j)}
wherein:
(i, j) is a pixel point in the hydraulic oil pipeline image;
r (i, j), G (i, j) and B (i, j) are respectively the values of the pixel point (i, j) in R, G, B three color channels;
g (i, j) is the gray value of the pixel point (i, j);
for the gray level image of the hydraulic oil pipeline, stretching the gray level of the image in a piecewise linear transformation mode, wherein the formula of the gray level stretching is as follows:
Figure BDA0003128335950000071
wherein:
f (x, y) is a gray scale map;
MAXf(x,y),MINf(x,y)respectively the maximum and minimum grey values of the grey map.
And S2, performing image enhancement processing on the hydraulic oil pipeline gray level image by using an image enhancement strategy.
Further, the invention utilizes an image enhancement strategy to carry out image enhancement processing on the gray level image of the hydraulic oil pipeline, and the image enhancement strategy flow comprises the following steps:
1) constructing a Gaussian filter kernel function matrix, and performing convolution operation on the Gaussian filter kernel function matrix and the hydraulic oil pipeline gray level image to obtain the hydraulic oil pipeline gray level image after Gaussian filtering; in one embodiment of the present invention, the constructed gaussian filter kernel function matrix is:
Figure BDA0003128335950000077
2) the histogram equalization processing is carried out on the gray level image of the hydraulic oil pipeline, and the method comprises the following steps:
counting the number of pixels corresponding to each gray level of the gray level image of the hydraulic oil pipeline to obtain a histogram of the image:
Figure BDA0003128335950000072
wherein:
k is 0,1, …, L-1, representing the gray level of the image;
nkrepresents the number of pixels whose gray level is k;
n represents the total number of pixels of the hydraulic oil pipeline gray level image;
calculating a cumulative histogram of the gray level image of the hydraulic oil pipeline:
Figure BDA0003128335950000073
mapping the cumulative histogram to a gray scale range L0，Lk]：
S＝L0+(Lk-L0)c(k)
Counting the number of pixels of each gray level in S after mapping transformation to obtain a new image histogram;
3) carrying out color correction on the gray level image of the hydraulic oil pipeline with the histogram equalization by adopting a self-adaptive underwater image color correction algorithm, wherein the formula of the self-adaptive underwater image color correction is as follows:
Figure BDA0003128335950000074
wherein:
i (R, G and B) represents the sum of the gray images of the hydraulic oil pipeline in three color channels of R, G and B;
μ represents the minkowski distance mean of the image color channel;
alpha represents the maximum value of the gray level image of the hydraulic oil pipeline in three color channels of R, G and B;
beta is a correction parameter, the closer the value is to 0, the higher the brightness of the corrected image is, and the value is set to 0.2;
4) carrying out image brightness enhancement processing on the gray level image of the hydraulic oil pipeline after color correction by using an underwater image brightness enhancement algorithm based on image gradient, wherein the formula of the image brightness enhancement is as follows:
Figure BDA0003128335950000075
Figure BDA0003128335950000076
Figure BDA0003128335950000078
wherein:
[Tmin，Tmax]representing a range of image brightness enhancement;
s (x, y) represents a luminance value of the image pixel (x, y);
e (x, y) represents the luminance value of the enhanced image pixel (x, y);
i represents a gradient direction;
Figure BDA0003128335950000086
representing the partial derivatives in different gradient directions;
qirepresenting the target gradient in different gradient directions;
wi(x, y) represents the influence weight of the gradient residual in different gradient directions, a represents the sensitivity of the gradient residual, and its value range is [ -0.8,2]In one embodiment of the invention, the value is 0.6;
g (x, y) is a luminance constraint function, limiting the enhanced luminance within the target range.
And S3, segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images.
Further, the enhanced image is input into an image segmentation network, the enhanced image is segmented by the image segmentation network, the image segmentation network is MASK R-CNN, the neural network adopts an FPN pyramid structure, Resnet-101 is used as a convolution network, and sub-images with different sizes are output;
the objective function of the image segmentation network is as follows:
Figure BDA0003128335950000081
wherein:
t' is the predicted image segmentation boundary;
t is a binarization result of the image segmentation boundary;
m is an input image;
r is a segmentation image;
d (t) represents the distance transformation of the segmentation frame, i.e. the distance map between different segmentation images;
and inputting the enhanced image into an image segmentation network, and performing image segmentation according to the segmentation boundary t' obtained by prediction to obtain a plurality of sub-images.
And S4, performing characteristic parameter extraction processing on the sub-images by utilizing a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix to obtain the image characteristics of the sub-images.
Further, the invention utilizes a characteristic parameter extraction algorithm based on a gray level co-occurrence matrix to extract and process the characteristic parameters of the sub-images, and the characteristic parameter extraction process comprises the following steps:
1) for a sub-image f (x, y) of size M × N, there are k gray levels, where the distance between any two pixels i and pixel j is
Figure BDA0003128335950000082
And the included angle formed by the two pixel connecting lines and the same coordinate axis is theta, the gray level co-occurrence matrix value of the image pixel i and the image pixel j is Pij(d, θ), the gray level co-occurrence matrix of the sub-images is:
Figure BDA0003128335950000083
2) extracting angular second moment features in the gray level co-occurrence matrix:
Figure BDA0003128335950000084
the angle second moment feature represents the square sum of each element in the arrangement combination of texture elements in the co-occurrence matrix; when the ASM value is large, the texture presented by the image is rough and the energy is large; on the contrary, the image has fine texture and small energy;
3) extracting entropy characteristics in the gray level co-occurrence matrix:
Figure BDA0003128335950000085
the entropy characteristics are used for expressing the information quantity expressed by the two-dimensional gray level image and representing the complexity of the vein texture in the image; when Ent approaches 0, it represents that there is almost no texture information in the grayscale image; if Ent is larger, the presented image context tends to be more complex;
4) extracting contrast characteristics in the gray level co-occurrence matrix:
Figure BDA0003128335950000091
the contrast characteristic represents the definition of texture venation of the gray level image and the depth degree of the groove; when the contrast is high, the image is clearer, and the grooves are deeper; otherwise, the image is fuzzy;
5) extracting clustering shadow features in the gray level co-occurrence matrix:
Figure BDA0003128335950000092
Figure BDA0003128335950000093
Figure BDA0003128335950000094
and for each sub-image, taking the features extracted from the gray level co-occurrence matrix as the features of the sub-image.
And S5, taking the image characteristics of the sub-images as the input of a convolutional neural network, and identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network.
Further, the invention takes the image characteristics of the sub-images as the input of the convolutional neural network; the convolutional neural network model consists of an input layer, a convolutional layer, a pooling layer, a full-link layer and an activation function layer; the convolutional layer is used for feature extraction, and the formula is as follows:
Figure BDA0003128335950000095
wherein:
Figure BDA0003128335950000096
an ith feature map representing an nth layer;
f () represents an activation function, which in one embodiment of the present invention is a ReLU activation function;
m represents a set of input sub-images;
Figure BDA0003128335950000097
j features representing the n-1 st layer;
Figure BDA0003128335950000098
representing a convolution kernel between the ith feature map of the nth layer and the jth feature map connection of the n-1 layer;
"+" represents the convolution operation;
Figure BDA0003128335950000099
a bias representing an ith characteristic of the nth layer;
the pooling layer performs feature compression on the input feature graph, so that the network complexity is simplified; and the full connection layer maps the acquired features to the original sample mark space and sends the output value to the classifier.
In the transmission process, input information is processed layer by layer from an input layer through a hidden unit layer and is transmitted to an output layer, and the output of each layer of neurons only influences the input of the next layer of neurons; if the output layer can not obtain the expected output, the reverse propagation is carried out, and the weight and the threshold of the network are learned and corrected through an error reverse propagation algorithm;
and when the mean square error of the network is smaller than a given value, entering NMS algorithm operation, and determining the position of the defect point, namely finishing the surface defect identification of the hydraulic oil pipeline.
The following describes embodiments of the present invention through an algorithmic experiment and tests of the inventive treatment method. The hardware test environment of the algorithm of the invention is as follows: inter (R) core (TM) i7-6700K CPU with software Matlab2018 b; the comparison method is a hydraulic oil pipeline surface weld defect identification method based on LSTM and a hydraulic oil pipeline surface weld defect identification method based on random forest.
In the algorithm experiment, the data set is 10G of hydraulic oil pipeline images. In the experiment, the image data of the hydraulic oil pipeline is input into the algorithm model, and the accuracy of defect identification is used as an evaluation index of algorithm feasibility, wherein the higher the accuracy of defect identification is, the higher the effectiveness and the feasibility of the algorithm are.
According to experimental results, the defect identification accuracy of the LSTM-based hydraulic oil pipeline surface weld defect identification method is 81.31%, the defect identification accuracy of the random forest-based hydraulic oil pipeline surface weld defect identification method is 83.22%, the defect identification accuracy of the method is 86.95%, and compared with a comparison algorithm, the method for identifying the underwater robot hydraulic oil pipeline surface weld defect can achieve higher defect identification accuracy.
The invention further provides a device for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot. Referring to fig. 2, a schematic diagram of an internal structure of a device for identifying a weld defect on a surface of a hydraulic oil pipeline of an underwater robot according to an embodiment of the present invention is shown; referring to fig. 3, a schematic device diagram of a device for identifying a weld defect on a surface of a hydraulic oil pipeline of an underwater robot according to an embodiment of the present invention is shown;
in the embodiment, the underwater robot hydraulic oil pipeline surface weld defect identification device 1 at least comprises an image acquisition device 11, a data processor 12, a defect identification device 13, a communication bus 14 and a network interface 15.
The image acquiring device 11 includes a mobile light source, an industrial endoscope, an industrial camera, and the like.
The data processor 12 includes at least one type of readable storage medium including flash memory, hard disks, multi-media cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, and the like. The data processor 12 may in some embodiments be an internal memory unit of the underwater robotic hydraulic oil pipeline surface weld defect identification apparatus 1, for example a hard disk of the underwater robotic hydraulic oil pipeline surface weld defect identification apparatus 1. The data processor 12 may also be an external storage device of the underwater robot hydraulic oil pipeline surface weld defect identification apparatus 1 in other embodiments, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), or the like provided on the underwater robot hydraulic oil pipeline surface weld defect identification apparatus 1. Further, the data processor 12 may also include both an internal storage unit and an external storage device of the underwater robot hydraulic oil pipeline surface weld defect recognition apparatus 1. The data processor 12 can be used for storing not only application software and various data installed on the surface weld defect recognition device 1 of the hydraulic oil pipeline of the underwater robot, but also temporarily storing data which is output or is to be output.
Defect identification device 13, which in some embodiments may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor or other data Processing chip, includes a monitoring Unit for running program code stored in data processor 12 or Processing data, such as defect identification program instructions 16.
The communication bus 14 is used to enable connection communication between these components.
The network interface 15 may optionally comprise a standard wired interface, a wireless interface (e.g. WI-FI interface), typically used for establishing a communication connection between the apparatus 1 and other electronic devices.
Optionally, the underwater robot hydraulic oil pipeline surface weld defect identification apparatus 1 may further include a user interface, the user interface may include a Display (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface may further include a standard wired interface and a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. Wherein the display, which may also be appropriately referred to as a display screen or a display unit, is used for displaying information processed in the underwater robot hydraulic oil pipeline surface weld defect recognition device 1 and for displaying a visual user interface.
Fig. 2 only shows the underwater robot hydraulic oil pipeline surface weld defect recognition device 1 with the components 11-15, and it can be understood by those skilled in the art that the structure shown in fig. 1 does not constitute a definition of the underwater robot hydraulic oil pipeline surface weld defect recognition device 1, and may include fewer or more components than those shown, or some components in combination, or a different arrangement of components.
In the embodiment of the device 1 for identifying weld defects on the surface of the hydraulic oil pipeline of the underwater robot, shown in fig. 2, the data processor 12 stores therein defect identification program instructions 16; the defect recognition device 13 executes the steps of the defect recognition program instructions 16 stored in the data processor 12, and the implementation method of the method for recognizing the weld defects on the surface of the hydraulic oil pipeline of the underwater robot is the same as the method for recognizing the weld defects on the surface of the hydraulic oil pipeline of the underwater robot, and the method is not described herein.
Furthermore, an embodiment of the present invention also provides a computer-readable storage medium having stored thereon defect identification program instructions executable by one or more processors to implement the following operations:
acquiring a hydraulic oil pipeline image, and performing image graying and grayscale stretching pretreatment on the hydraulic oil pipeline image to obtain a hydraulic oil pipeline grayscale image;
carrying out image enhancement processing on the gray level image of the hydraulic oil pipeline by using an image enhancement strategy;
segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images;
performing characteristic parameter extraction processing on the subimages by utilizing a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix to obtain the image characteristics of the subimages;
and (3) taking the image characteristics of the sub-images as the input of a convolutional neural network, and identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network.
It should be noted that the above-mentioned numbers of the embodiments of the present invention are merely for description, and do not represent the merits of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (8)

1. A method for identifying weld defects on the surface of a hydraulic oil pipeline of an underwater robot is characterized by comprising the following steps:
acquiring a hydraulic oil pipeline image, and performing image graying and grayscale stretching pretreatment on the hydraulic oil pipeline image to obtain a hydraulic oil pipeline grayscale image;
carrying out image enhancement processing on the gray level image of the hydraulic oil pipeline by using an image enhancement strategy;
segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images;
performing characteristic parameter extraction processing on the subimages by utilizing a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix to obtain the image characteristics of the subimages;
and (3) taking the image characteristics of the sub-images as the input of a convolutional neural network, and identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network.
2. The method for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot as claimed in claim 1, wherein the preprocessing of image graying and grayscale stretching on the hydraulic oil pipeline image comprises the following steps:
solving the maximum value of three components of each pixel in the hydraulic oil pipeline image, setting the maximum value as the gray value of the pixel point, and obtaining the gray map of the hydraulic oil pipeline image, wherein the formula of the graying treatment is as follows:
G(i,j)＝max{R(i,j),G(i,j),B(i,j)}
wherein:
(i, j) is a pixel point in the hydraulic oil pipeline image;
r (i, j), G (i, j) and B (i, j) are respectively the values of the pixel point (i, j) in R, G, B three color channels;
g (i, j) is the gray value of the pixel point (i, j);
for the gray level image of the hydraulic oil pipeline, stretching the gray level of the image in a piecewise linear transformation mode, wherein the formula of the gray level stretching is as follows:
Figure FDA0003128335940000011
wherein:
f (x, y) is a gray scale map;
MAXf(x,y),MINf(x,y)respectively the maximum and minimum grey values of the grey map.
3. The method for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot as claimed in claim 2, wherein the image enhancement strategy flow comprises the following steps:
1) constructing a Gaussian filter kernel function matrix, and performing convolution operation on the Gaussian filter kernel function matrix and the hydraulic oil pipeline gray level image to obtain the hydraulic oil pipeline gray level image after Gaussian filtering;
2) the histogram equalization processing is carried out on the gray level image of the hydraulic oil pipeline, and the method comprises the following steps:
counting the number of pixels corresponding to each gray level of the gray level image of the hydraulic oil pipeline to obtain a histogram of the image:
Figure FDA0003128335940000012
wherein:
k is 0,1, …, L-1, representing the gray level of the image;
nkrepresents the number of pixels whose gray level is k;
n represents the total number of pixels of the hydraulic oil pipeline gray level image;
calculating a cumulative histogram of the gray level image of the hydraulic oil pipeline:
Figure FDA0003128335940000013
mapping the cumulative histogram to a gray scale range L0,Lk]：
S＝L0+(Lk-L0)c(k)
Counting the number of pixels of each gray level in S after mapping transformation to obtain a new image histogram;
3) carrying out color correction on the gray level image of the hydraulic oil pipeline with the histogram equalization by adopting a self-adaptive underwater image color correction algorithm, wherein the formula of the self-adaptive underwater image color correction is as follows:
Figure FDA0003128335940000021
wherein:
i (R, G and B) represents the sum of the gray images of the hydraulic oil pipeline in three color channels of R, G and B;
μ represents the minkowski distance mean of the image color channel;
alpha represents the maximum value of the gray level image of the hydraulic oil pipeline in three color channels of R, G and B;
beta is a correction parameter, which is set to 0.2;
4) carrying out image brightness enhancement processing on the gray level image of the hydraulic oil pipeline after color correction by using an underwater image brightness enhancement algorithm based on image gradient, wherein the formula of the image brightness enhancement is as follows:
Figure FDA0003128335940000022
Figure FDA0003128335940000023
Figure FDA0003128335940000024
wherein:
[Tmin,Tmax]representing a range of image brightness enhancement;
s (x, y) represents a luminance value of the image pixel (x, y);
e (x, y) represents the luminance value of the enhanced image pixel (x, y);
i represents a gradient direction;
Figure FDA0003128335940000027
representing the partial derivatives in different gradient directions;
qirepresenting the target gradient in different gradient directions;
wi(x, y) represents the influence weight of the gradient residual in different gradient directions, a represents the sensitivity of the gradient residual, and its value range is [ -0.8,2]To (c) to (d);
g (x, y) is a luminance constraint function, limiting the enhanced luminance within the target range.
4. The method for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot as claimed in claim 3, wherein the segmenting the enhanced image by using the image segmentation network comprises the following steps:
the objective function of the image segmentation network is as follows:
Figure FDA0003128335940000025
wherein:
t' is the predicted image segmentation boundary;
t is a binarization result of the image segmentation boundary;
m is an input image;
r is a segmentation image;
d (t) represents the distance transformation of the segmentation frame, i.e. the distance map between different segmentation images;
and inputting the enhanced image into an image segmentation network, and performing image segmentation according to the segmentation boundary t' obtained by prediction to obtain a plurality of sub-images.
5. The method for identifying the weld defects on the surface of the hydraulic oil pipeline of the underwater robot as claimed in claim 4, wherein the performing the feature parameter extraction processing on the sub-images by using the feature parameter extraction algorithm based on the gray level co-occurrence matrix comprises the following steps:
1) for a sub-image f (x, y) of size M N, there are kGray scale in which the distance between any two pixels i and pixel j is
Figure FDA0003128335940000026
And the included angle formed by the two pixel connecting lines and the same coordinate axis is theta, the gray level co-occurrence matrix value of the image pixel i and the image pixel j is Pij(d, θ), the gray level co-occurrence matrix of the sub-images is:
Figure FDA0003128335940000031
2) extracting angular second moment features in the gray level co-occurrence matrix:
Figure FDA0003128335940000032
3) extracting entropy characteristics in the gray level co-occurrence matrix:
Figure FDA0003128335940000033
4) extracting contrast characteristics in the gray level co-occurrence matrix:
Figure FDA0003128335940000034
5) extracting clustering shadow features in the gray level co-occurrence matrix:
Figure FDA0003128335940000035
Figure FDA0003128335940000036
Figure FDA0003128335940000037
and for each sub-image, taking the features extracted from the gray level co-occurrence matrix as the features of the sub-image.
6. The method for identifying the weld defect on the surface of the hydraulic oil pipeline of the underwater robot as claimed in claim 5, wherein the identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network comprises the following steps:
taking the image characteristics of the sub-images as the input of a convolutional neural network; the convolutional layer in the convolutional neural network is used for feature extraction, and the formula is as follows:
Figure FDA0003128335940000038
wherein:
Figure FDA0003128335940000039
an ith feature map representing an nth layer;
f () represents an activation function;
m represents a set of input sub-images;
Figure FDA00031283359400000310
j features representing the n-1 st layer;
Figure FDA00031283359400000311
representing a convolution kernel between the ith feature map of the nth layer and the jth feature map connection of the n-1 layer;
Figure FDA00031283359400000312
a bias representing an ith characteristic of the nth layer;
the pooling layer performs feature compression on the input feature graph, so that the network complexity is simplified; and the full connection layer maps the acquired features to the original sample mark space and sends the output value to the classifier.
In the transmission process, input information is processed layer by layer from an input layer through a hidden unit layer and is transmitted to an output layer, and the output of each layer of neurons only influences the input of the next layer of neurons; if the output layer can not obtain the expected output, the reverse propagation is carried out, and the weight and the threshold of the network are learned and corrected through an error reverse propagation algorithm;
and when the mean square error of the network is smaller than a given value, entering NMS algorithm operation, and determining the position of the defect point, namely finishing the surface defect identification of the hydraulic oil pipeline.
7. An underwater robot hydraulic oil pipeline surface weld defect recognition device, characterized in that the device includes:
the image acquisition device is used for acquiring an image of the hydraulic oil pipeline;
the data processor is used for preprocessing the image graying and the gray stretching of the hydraulic oil pipeline image and carrying out image enhancement processing on the hydraulic oil pipeline gray image by utilizing an image enhancement strategy; segmenting the enhanced image by using an image segmentation network to obtain a plurality of sub-images;
the defect identification device is used for extracting the characteristic parameters of the sub-images by utilizing a characteristic parameter extraction algorithm based on the gray level co-occurrence matrix to obtain the image characteristics of the sub-images; and (3) taking the image characteristics of the sub-images as the input of a convolutional neural network, and identifying the position of the weld defect point on the surface of the hydraulic oil pipeline by using the convolutional neural network.
8. A computer readable storage medium having stored thereon defect identification program instructions executable by one or more processors to implement the steps of a method of implementing weld defect identification on a surface of a hydraulic oil pipeline of an underwater robot as described above.
CN202110697224.2A 2021-06-23 2021-06-23 Method and device for identifying weld defects on surface of hydraulic oil pipeline of underwater robot Active CN113436162B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110697224.2A CN113436162B (en) 2021-06-23 2021-06-23 Method and device for identifying weld defects on surface of hydraulic oil pipeline of underwater robot

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110697224.2A CN113436162B (en) 2021-06-23 2021-06-23 Method and device for identifying weld defects on surface of hydraulic oil pipeline of underwater robot

Publications (2)

Publication Number Publication Date
CN113436162A true CN113436162A (en) 2021-09-24
CN113436162B CN113436162B (en) 2022-12-09

Family

ID=77753498

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110697224.2A Active CN113436162B (en) 2021-06-23 2021-06-23 Method and device for identifying weld defects on surface of hydraulic oil pipeline of underwater robot

Country Status (1)

Country Link
CN (1) CN113436162B (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113763294A (en) * 2021-09-26 2021-12-07 上海航天精密机械研究所 Weld image rapid preprocessing method and system based on dynamic CLAHE
CN114399432A (en) * 2021-12-13 2022-04-26 广西北投信创科技投资集团有限公司 Target identification method, device, equipment, medium and product
CN115598131A (en) * 2022-09-21 2023-01-13 江苏华电昆山热电有限公司（Cn） Positioning method suitable for inner welding seam of pipeline
CN117576105A (en) * 2024-01-17 2024-02-20 高科建材（咸阳）管道科技有限公司 Pipeline production control method and system based on artificial intelligence
CN117710365A (en) * 2024-02-02 2024-03-15 中国电建集团华东勘测设计研究院有限公司 Processing method and device for defective pipeline image and electronic equipment
CN117830300A (en) * 2024-03-04 2024-04-05 新奥新能源工程技术有限公司 Visual-based gas pipeline appearance quality detection method
CN118279314A (en) * 2024-06-04 2024-07-02 河南安盛技术管理有限公司 Water quality abnormality detection method

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130112673A1 (en) * 2011-11-07 2013-05-09 Lincoln Global, Inc. Use of mobile communications devices as user interface for welding equipment and systems
CN104408724A (en) * 2014-11-27 2015-03-11 中南大学 Depth information method and system for monitoring liquid level and recognizing working condition of foam flotation
CN104985290A (en) * 2015-08-04 2015-10-21 梁彦云 Neural-network-identification-based underwater weld joint tracking method
CN105938563A (en) * 2016-04-14 2016-09-14 北京工业大学 Weld surface defect identification method based on image texture
US10282914B1 (en) * 2015-07-17 2019-05-07 Bao Tran Systems and methods for computer assisted operation
CN110349114A (en) * 2019-05-24 2019-10-18 江西理工大学 Applied to the image enchancing method of AOI equipment, device and road video monitoring equipment
KR102125167B1 (en) * 2019-01-21 2020-06-19 한양대학교 에리카산학협력단 Automatic welding device, working method thereof
CN111360780A (en) * 2020-03-20 2020-07-03 北京工业大学 Garbage picking robot based on visual semantic SLAM
CN111932489A (en) * 2020-06-03 2020-11-13 西安电子科技大学 Weld defect detection method, system, storage medium, computer device and terminal
CN112396564A (en) * 2020-11-19 2021-02-23 汪金玲 Product packaging quality detection method and system based on deep learning
CN112734693A (en) * 2020-12-18 2021-04-30 平安科技（深圳）有限公司 Pipeline weld defect detection method and related device

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130112673A1 (en) * 2011-11-07 2013-05-09 Lincoln Global, Inc. Use of mobile communications devices as user interface for welding equipment and systems
CN104408724A (en) * 2014-11-27 2015-03-11 中南大学 Depth information method and system for monitoring liquid level and recognizing working condition of foam flotation
US10282914B1 (en) * 2015-07-17 2019-05-07 Bao Tran Systems and methods for computer assisted operation
CN104985290A (en) * 2015-08-04 2015-10-21 梁彦云 Neural-network-identification-based underwater weld joint tracking method
CN105938563A (en) * 2016-04-14 2016-09-14 北京工业大学 Weld surface defect identification method based on image texture
KR102125167B1 (en) * 2019-01-21 2020-06-19 한양대학교 에리카산학협력단 Automatic welding device, working method thereof
CN110349114A (en) * 2019-05-24 2019-10-18 江西理工大学 Applied to the image enchancing method of AOI equipment, device and road video monitoring equipment
CN111360780A (en) * 2020-03-20 2020-07-03 北京工业大学 Garbage picking robot based on visual semantic SLAM
CN111932489A (en) * 2020-06-03 2020-11-13 西安电子科技大学 Weld defect detection method, system, storage medium, computer device and terminal
CN112396564A (en) * 2020-11-19 2021-02-23 汪金玲 Product packaging quality detection method and system based on deep learning
CN112734693A (en) * 2020-12-18 2021-04-30 平安科技（深圳）有限公司 Pipeline weld defect detection method and related device

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ZIMU XIAO ET AL.: "Development of a CNN edge detection model of noised X-ray images for enhanced performance of non-destructive testing", 《MEASUREMENT》 *
黄晔: "基于BP神经网络的焊缝缺陷建模及其识别算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113763294A (en) * 2021-09-26 2021-12-07 上海航天精密机械研究所 Weld image rapid preprocessing method and system based on dynamic CLAHE
CN113763294B (en) * 2021-09-26 2023-08-08 上海航天精密机械研究所 Quick preprocessing method and system for weld image based on dynamic CLAHE
CN114399432A (en) * 2021-12-13 2022-04-26 广西北投信创科技投资集团有限公司 Target identification method, device, equipment, medium and product
CN115598131A (en) * 2022-09-21 2023-01-13 江苏华电昆山热电有限公司（Cn） Positioning method suitable for inner welding seam of pipeline
CN117576105A (en) * 2024-01-17 2024-02-20 高科建材（咸阳）管道科技有限公司 Pipeline production control method and system based on artificial intelligence
CN117576105B (en) * 2024-01-17 2024-03-29 高科建材（咸阳）管道科技有限公司 Pipeline production control method and system based on artificial intelligence
CN117710365A (en) * 2024-02-02 2024-03-15 中国电建集团华东勘测设计研究院有限公司 Processing method and device for defective pipeline image and electronic equipment
CN117710365B (en) * 2024-02-02 2024-05-03 中国电建集团华东勘测设计研究院有限公司 Processing method and device for defective pipeline image and electronic equipment
CN117830300A (en) * 2024-03-04 2024-04-05 新奥新能源工程技术有限公司 Visual-based gas pipeline appearance quality detection method
CN117830300B (en) * 2024-03-04 2024-05-14 新奥新能源工程技术有限公司 Visual-based gas pipeline appearance quality detection method
CN118279314A (en) * 2024-06-04 2024-07-02 河南安盛技术管理有限公司 Water quality abnormality detection method

Also Published As

Publication number Publication date
CN113436162B (en) 2022-12-09

Similar Documents

Publication Publication Date Title
CN113436162B (en) Method and device for identifying weld defects on surface of hydraulic oil pipeline of underwater robot
CN110060237B (en) Fault detection method, device, equipment and system
CN110414507B (en) License plate recognition method and device, computer equipment and storage medium
CN106934795B (en) A kind of automatic testing method and prediction technique of glue into concrete beam cracks
CN111257341B (en) Underwater building crack detection method based on multi-scale features and stacked full convolution network
CN111680690B (en) Character recognition method and device
CN110033040B (en) Flame identification method, system, medium and equipment
TW201732651A (en) Word segmentation method and apparatus
EP3176751A1 (en) Information processing device, information processing method, computer-readable recording medium, and inspection system
CN109118473A (en) Angular-point detection method, storage medium and image processing system neural network based
CN110852311A (en) Three-dimensional human hand key point positioning method and device
CN112614136A (en) Infrared small target real-time instance segmentation method and device
CN107563290A (en) A kind of pedestrian detection method and device based on image
JP2010157118A (en) Pattern identification device and learning method for the same and computer program
CN111951283A (en) Medical image identification method and system based on deep learning
CN113435377A (en) Medical palm vein image acquisition monitoring method and system
CN115439804A (en) Monitoring method and device for high-speed rail maintenance
JP2006285956A (en) Red eye detecting method and device, and program
CN116128849A (en) Method, device, equipment and storage medium for detecting underwater cracks of concrete dam
CN117557565B (en) Detection method and device for lithium battery pole piece
CN116934762B (en) System and method for detecting surface defects of lithium battery pole piece
CN114998282B (en) Image detection method, device, electronic equipment and storage medium
CN113554748B (en) Medical instrument three-dimensional modeling method and device
CN110070490A (en) Image split-joint method and device
CN115270841A (en) Bar code detection method and device, storage medium and computer equipment

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