CN113838001A - Ultrasonic full-focus image defect processing method and device based on nuclear density estimation - Google Patents
Ultrasonic full-focus image defect processing method and device based on nuclear density estimation Download PDFInfo
- Publication number
- CN113838001A CN113838001A CN202110979361.5A CN202110979361A CN113838001A CN 113838001 A CN113838001 A CN 113838001A CN 202110979361 A CN202110979361 A CN 202110979361A CN 113838001 A CN113838001 A CN 113838001A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel point
- ultrasonic
- density estimation
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 134
- 238000003672 processing method Methods 0.000 title claims description 4
- 238000012545 processing Methods 0.000 claims abstract description 104
- 239000011159 matrix material Substances 0.000 claims abstract description 44
- 238000007781 pre-processing Methods 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000011002 quantification Methods 0.000 claims description 7
- 230000004807 localization Effects 0.000 claims 1
- 230000006870 function Effects 0.000 description 20
- 238000003384 imaging method Methods 0.000 description 8
- 238000010606 normalization Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 238000002604 ultrasonography Methods 0.000 description 6
- 230000002950 deficient Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Images
Classifications
-
- 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
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a method and a device for processing ultrasonic full-focus image defects based on nuclear density estimation. The method comprises the following steps: acquiring an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, and performing image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image; obtaining an ultrasonic intensity threshold value and a kernel density estimation value of each pixel point based on the preprocessed image, and setting a pixel value corresponding to each pixel point according to a comparison result of the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image; and carrying out binarization processing on the intermediate processing image to obtain a target image, and positioning and quantifying a defect area in the target image. The method can effectively eliminate the artifacts in the ultrasonic full-focus image, position and quantify the defects in the ultrasonic full-focus image, and improve the accuracy and efficiency of defect processing of the ultrasonic full-focus image.
Description
Technical Field
The invention relates to the technical field of ultrasonic imaging image processing, in particular to a method and a device for processing ultrasonic full-focus image defects based on nuclear density estimation.
Background
The ultrasonic full-focusing imaging technology is a new technology in the field of ultrasonic nondestructive detection, and realizes the rapid detection of workpieces by analyzing and imaging detection echo data. The ultrasonic full-focus imaging technology has attracted extensive attention in the industrial fields of aerospace, petroleum pipelines, nuclear power stations and the like due to the advantages of wide coverage, high imaging resolution, high sensitivity for detecting small defects and the like. However, the ultrasonic full-focus imaging technology mainly utilizes scattering information of defects located in a positive reflection region, and a sensor can only receive partial scattering information, so that the scattering information of a transmission region is difficult to acquire, artifacts are easy to appear in an ultrasonic full-focus image, and the defects are difficult to locate and quantify. Therefore, how to effectively eliminate the artifacts in the ultrasonic full-focus image and locate and quantify the defects becomes a big problem to be solved at present.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a method and a device for processing the defects of the ultrasonic full-focus image based on nuclear density estimation, which can effectively eliminate artifacts in the ultrasonic full-focus image, position and quantify the defects in the ultrasonic full-focus image, and improve the accuracy and efficiency of processing the defects of the ultrasonic full-focus image.
In order to solve the above technical problem, in a first aspect, an embodiment of the present invention provides a method for processing a defect in an ultrasonic full-focus image based on nuclear density estimation, including:
acquiring an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, and performing image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image;
obtaining an ultrasonic intensity threshold value and a kernel density estimation value of each pixel point based on the preprocessed image, and setting a pixel value corresponding to each pixel point according to a comparison result of the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image;
and carrying out binarization processing on the intermediate processing image to obtain a target image, and positioning and quantifying a defect area in the target image.
Further, the image preprocessing is performed on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image, specifically:
and denoising the ultrasonic full-focus image by adopting a median filter, and normalizing the denoised ultrasonic full-focus image according to the image matrix to obtain the preprocessed image.
Further, the obtaining of the ultrasonic intensity threshold and the kernel density estimation value of each pixel point based on the preprocessed image specifically includes:
based on the preprocessed image, taking the pixel quantity of which the ultrasonic intensity is greater than a threshold value as an objective function, converging a first derivative of the objective function to zero, and taking the corresponding threshold value as the ultrasonic intensity threshold value;
and respectively inputting each pixel point of the preprocessed image into a pre-constructed kernel density estimation model to obtain a kernel density estimation value of each pixel point.
Further, the step of setting the pixel value corresponding to each pixel point according to the comparison result between the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processing image specifically comprises the following steps:
comparing the kernel density estimation value of the pixel point with the ultrasonic intensity threshold, if the kernel density estimation value of the pixel point is smaller than the ultrasonic intensity threshold, setting the pixel value of the pixel point as a first preset pixel value, otherwise, setting the pixel value of the pixel point as a second preset pixel value, and obtaining the intermediate processing image.
Further, the positioning and quantifying the defect area in the target image specifically comprises:
performing defect identification on the target image to obtain the defect area and the leftmost pixel point, the rightmost pixel point, the uppermost pixel point and the lowermost pixel point of the defect area;
obtaining a first line segment according to the leftmost pixel point and the rightmost pixel point, obtaining a second line segment according to the uppermost pixel point and the bottommost pixel point, and taking the intersection point of the first line segment and the second line segment as the central point of the defect area;
and calculating to obtain the size of the defect region by combining the leftmost pixel point, the rightmost pixel point, the topmost pixel point and the bottommost pixel point.
In a second aspect, an embodiment of the present invention provides an apparatus for processing defects in an ultrasonic full-focus image based on nuclear density estimation, including:
the preprocessing module is used for acquiring an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, and carrying out image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image;
the intermediate processing module is used for obtaining an ultrasonic intensity threshold value and a kernel density estimation value of each pixel point based on the preprocessed image, and setting a pixel value corresponding to each pixel point according to a comparison result of the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image;
and the defect processing module is used for carrying out binarization processing on the intermediate processing image to obtain a target image and positioning and quantifying a defect area in the target image.
Further, the image preprocessing is performed on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image, specifically:
and denoising the ultrasonic full-focus image by adopting a median filter, and normalizing the denoised ultrasonic full-focus image according to the image matrix to obtain the preprocessed image.
Further, the obtaining of the ultrasonic intensity threshold and the kernel density estimation value of each pixel point based on the preprocessed image specifically includes:
based on the preprocessed image, taking the pixel quantity of which the ultrasonic intensity is greater than a threshold value as an objective function, converging a first derivative of the objective function to zero, and taking the corresponding threshold value as the ultrasonic intensity threshold value;
and respectively inputting each pixel point of the preprocessed image into a pre-constructed kernel density estimation model to obtain a kernel density estimation value of each pixel point.
Further, the step of setting the pixel value corresponding to each pixel point according to the comparison result between the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processing image specifically comprises the following steps:
comparing the kernel density estimation value of the pixel point with the ultrasonic intensity threshold, if the kernel density estimation value of the pixel point is smaller than the ultrasonic intensity threshold, setting the pixel value of the pixel point as a first preset pixel value, otherwise, setting the pixel value of the pixel point as a second preset pixel value, and obtaining the intermediate processing image.
Further, the positioning and quantifying the defect area in the target image specifically comprises:
performing defect identification on the target image to obtain the defect area and the leftmost pixel point, the rightmost pixel point, the uppermost pixel point and the lowermost pixel point of the defect area;
obtaining a first line segment according to the leftmost pixel point and the rightmost pixel point, obtaining a second line segment according to the uppermost pixel point and the bottommost pixel point, and taking the intersection point of the first line segment and the second line segment as the central point of the defect area;
and calculating to obtain the size of the defect region by combining the leftmost pixel point, the rightmost pixel point, the topmost pixel point and the bottommost pixel point.
The embodiment of the invention has the following beneficial effects:
the method comprises the steps of obtaining an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, conducting image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image, obtaining an ultrasonic intensity threshold value and a nuclear density estimated value of each pixel point based on the preprocessed image, setting a pixel value of a corresponding pixel point according to a comparison result of the nuclear density estimated value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image, conducting binarization processing on the intermediate processed image to obtain a target image, locating and quantifying a defect area in the target image, and completing defect processing of the ultrasonic full-focus image. Compared with the prior art, the embodiment of the invention obtains the intermediate processing image by comparing the nuclear density estimated value of each pixel point with the ultrasonic intensity threshold value, can effectively eliminate the artifact in the ultrasonic full-focus image, avoids the influence of the artifact on the positioning and quantification of the subsequent defects as much as possible, and can effectively reduce the data volume of the ultrasonic full-focus image by positioning and quantifying the defect area in the binary image obtained by carrying out binarization processing on the target image, namely the intermediate processing image, so as to position and quantify the defects in the ultrasonic full-focus image with lower processing amount, thereby improving the accuracy and efficiency of the defect processing of the ultrasonic full-focus image.
Drawings
FIG. 1 is a schematic flowchart of a defect processing method for an ultrasonic full-focus image based on nuclear density estimation according to a first embodiment of the present invention;
fig. 2 is a schematic structural diagram of an apparatus for processing defect of ultrasonic full-focus image based on nuclear density estimation according to a second embodiment of the present invention.
Detailed Description
The technical solutions in the present invention will be described clearly and completely with reference to the accompanying drawings, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that, the step numbers in the text are only for convenience of explanation of the specific embodiments, and do not serve to limit the execution sequence of the steps.
The first embodiment:
as shown in fig. 1, the first embodiment provides a method for processing defects in an ultrasonic full-focus image based on nuclear density estimation, which includes steps S1 to S3:
s1, acquiring an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, and performing image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image;
s2, obtaining an ultrasonic intensity threshold value and a kernel density estimation value of each pixel point based on the preprocessed image, and setting a pixel value of a corresponding pixel point according to a comparison result of the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image;
and S3, carrying out binarization processing on the intermediate processing image to obtain a target image, and positioning and quantifying a defect area in the target image.
Illustratively, in step S1, a simulation experiment is performed on the defective workpiece by using an ultrasonic full-focus imaging algorithm in the computer software, an ultrasonic full-focus image containing the defective region and an image matrix thereof are obtained, and the image preprocessing is performed on the ultrasonic full-focus image according to the image matrix, so as to obtain a preprocessed image.
In step S2, an objective function is constructed based on the preprocessed image and solved to obtain an ultrasound intensity threshold, a kernel density estimation value of each pixel point of the preprocessed image is obtained through a kernel density estimation model, the kernel density estimation value of each pixel point of the preprocessed image is compared with the ultrasound intensity threshold, and a pixel value of a corresponding pixel point is set according to a comparison result to obtain an intermediate processed image.
In step S3, the intermediate processing image is binarized, that is, the intermediate processing image is converted into a binary image to obtain a target image, the target image is defect-identified to obtain a defect region and edge pixel points of the defect region, and the defect region is located and quantified according to the edge pixel points of the defect region, thereby completing defect processing of the ultrasonic full-focus image.
In the embodiment, the kernel density estimated value of each pixel point is compared with the ultrasonic intensity threshold value to make a decision to obtain the intermediate processing image, so that the artifact in the ultrasonic full-focus image can be effectively eliminated, the influence of the artifact on the positioning and quantification of the subsequent defects can be avoided as much as possible, and the data volume of the ultrasonic full-focus image can be effectively reduced by positioning and quantifying the target image, namely the defect area in the binary image obtained by performing binarization processing on the intermediate processing image, so that the defects in the ultrasonic full-focus image can be positioned and quantified with lower processing amount, and the accuracy and efficiency of the defect processing of the ultrasonic full-focus image can be improved.
In a preferred embodiment, the performing image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image specifically includes: and denoising the ultrasonic full-focus image by adopting a median filter, and normalizing the denoised ultrasonic full-focus image according to the image matrix to obtain a preprocessed image.
Exemplarily, a median filter is adopted to perform denoising processing on the ultrasonic full-focus image to obtain a denoised ultrasonic full-focus image, and normalization processing is performed on the denoised ultrasonic full-focus image according to an image matrix to obtain a preprocessed image, which specifically includes:
in formula (1), max (x) is the maximum pixel value in the image matrix; min (x) is the minimum pixel value in the image matrix; x is the number ofiThe pixel value of the ith pixel point of the denoised ultrasonic full focusing image is obtained; x'iThe pixel value of the ith pixel point of the preprocessed image is obtained.
In the embodiment, the median filter is adopted to denoise the ultrasonic full-focus image, so that the noise interference smooth image can be effectively eliminated, and the denoised ultrasonic full-focus image can be effectively ensured to be a standard image by normalizing the denoised ultrasonic full-focus image according to the image matrix.
In a preferred embodiment, the obtaining of the ultrasonic intensity threshold and the kernel density estimation value of each pixel point based on the preprocessed image specifically includes: based on the preprocessed image, taking the pixel quantity of which the ultrasonic intensity is greater than the threshold value as a target function, enabling a first derivative of the target function to be converged to zero, and taking the corresponding threshold value as an ultrasonic intensity threshold value; and respectively inputting each pixel point of the preprocessed image into a pre-constructed kernel density estimation model to obtain a kernel density estimation value of each pixel point.
Illustratively, based on the preprocessed image, let the image matrix after the image normalization process be a, the threshold value δ be 0.1, 0.2, …, 1, and take the pixel amount of the ultrasound intensity greater than the threshold value δ as the objective function S (δ), then the objective function is expressed as S (δ) sum (sum (a > δ)), then normalize the obtained S (δ), and when the first derivative S '(δ) of S (δ) after the normalization process converges to zero, solve to obtain an optimal threshold value δ, while the change of S (δ) is insignificant, make the first derivative S' (δ) of the objective function converge to zero, and take the corresponding threshold value δ as the ultrasound intensity threshold value.
A kernel density estimation model is constructed in advance, a Gaussian kernel is selected by the model to serve as a kernel function of kernel density estimation, and then the kernel density estimation function at any pixel point is as follows:
in the formula (2), n is the number of pixel points; h is the optimal bandwidth of the kernel density estimation; x is the pixel value of any pixel point; x is the number ofjIn practical application, the pixel value of the jth pixel point of the ultrasonic full-focus image after normalization processing is used as a kernel density estimation functionConsidered as a normal distribution, the optimal bandwidth isThe standard deviation of the pixel values of all the pixel points.
Respectively inputting each pixel point of the preprocessed image into a kernel density estimation model to obtain a kernel density estimation value of each pixel point
In an embodiment of the present invention, the step of setting the pixel value of the corresponding pixel point according to the comparison result between the kernel density estimation value of each pixel point and the ultrasonic intensity threshold to obtain the intermediate processing image specifically includes: and comparing the kernel density estimated value of the pixel point with the ultrasonic intensity threshold, if the kernel density estimated value of the pixel point is smaller than the ultrasonic intensity threshold, setting the pixel value of the pixel point as a first preset pixel value, and otherwise, setting the pixel value of the pixel point as a second preset pixel value to obtain an intermediate processing image.
In a preferred implementation manner of this embodiment, the first predetermined pixel value is 255, and the second predetermined pixel value is other values except 255.
Illustratively, for each pixel point of the intermediate processing image, comparing the kernel density estimation value of the pixel point with the ultrasonic intensity threshold, if the kernel density estimation value of the pixel point is smaller than the ultrasonic intensity threshold, setting the pixel value of the pixel point to be a first preset pixel value 255, that is, setting the pixel point to be white, otherwise, setting the pixel value of the pixel point to be a second preset pixel value, and setting the pixel point to be other colors.
In the embodiment, the nuclear density estimated value of each pixel point is compared with the ultrasonic intensity threshold to make a decision, so that the intermediate processing image is obtained, the artifact in the ultrasonic full-focus image can be effectively eliminated, the influence of the artifact on the positioning and quantification of the subsequent defects is avoided as much as possible, and the accuracy of the defect processing of the ultrasonic full-focus image is improved.
In a preferred embodiment, the locating and quantifying the defect region in the target image specifically includes: performing defect identification on the target image to obtain a defect area and leftmost pixel points, rightmost pixel points, uppermost pixel points and lowermost pixel points of the defect area; obtaining a first line segment according to the leftmost pixel point and the rightmost pixel point, obtaining a second line segment according to the uppermost pixel point and the bottommost pixel point, and taking the intersection point of the first line segment and the second line segment as the central point of the defect area; and calculating to obtain the size of the defect region by combining the leftmost pixel point, the rightmost pixel point, the topmost pixel point and the bottommost pixel point.
Illustratively, defect identification is carried out on the target image to obtain a defect area and a leftmost pixel point A (x) of the defect areaA,yA) Rightmost pixel point B (x)B,yB) The top pixel C (x)C,yC) The bottom pixel D (x)D,yD)。
According to the leftmost pixel point A (x)A,yA) And the rightmost pixel point B (x)B,yB) A first segment AB is obtained, namely:
in the formula (3), the coordinates of the point on the first line segment AB are (x)1,y1);
According to the uppermost pixel point C (x)C,yC) And the lowest pixel D (x)D,yD) A second line segment CD is obtained, namely:
in the formula (4), the coordinates of the point on the second line segment CD are (x)2,y2);
Let y1=y2And solving to obtain the intersection point of the first line segment AB and the second line segment CD, and using the intersection point of the first line segment AB and the second line segment CD as the central point of the defect area to realize the positioning of the defect.
Combining the leftmost pixel point A (x)A,yA) Rightmost pixel point B (x)B,yB) The top pixel C (x)C,yC) The bottom pixel D (x)D,yD) Performing coordinate calculation to obtain the size of the defect region, such as the length l ═ x of the defect regionA-xB| and Width w ═ yC-yDAnd quantifying the defects.
In the embodiment, the data volume of the ultrasonic full-focus image can be effectively reduced by positioning and quantifying the defect area in the target image, namely the binary image obtained by performing binarization processing on the intermediate processing image, and the defects in the ultrasonic full-focus image are positioned and quantified with lower processing amount, so that the efficiency of defect processing of the ultrasonic full-focus image is improved.
Second embodiment:
as shown in fig. 2, a second embodiment provides an ultrasonic full-focus image defect processing apparatus based on nuclear density estimation, including: the preprocessing module 21 is configured to obtain an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, and perform image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image; the intermediate processing module 22 is configured to obtain an ultrasonic intensity threshold value and a kernel density estimation value of each pixel point based on the preprocessed image, and set a pixel value of a corresponding pixel point according to a comparison result between the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image; and the defect processing module 23 is configured to perform binarization processing on the intermediate processing image to obtain a target image, and position and quantify a defect area in the target image.
Illustratively, by the preprocessing module 21, a simulation experiment is performed on the defective workpiece by using an ultrasonic full-focus imaging algorithm in computer software, an ultrasonic full-focus image containing a defective region and an image matrix thereof are obtained, and the ultrasonic full-focus image is subjected to image preprocessing according to the image matrix to obtain a preprocessed image.
Through the intermediate processing module 22, an objective function is constructed based on the preprocessed image and is solved to obtain an ultrasonic intensity threshold, a kernel density estimation value of each pixel point of the preprocessed image is obtained through a kernel density estimation model, the kernel density estimation value of each pixel point of the preprocessed image is compared with the ultrasonic intensity threshold, the pixel value of the corresponding pixel point is set according to a comparison result, and the intermediate processed image is obtained.
The defect processing module 23 is used to perform binarization processing on the intermediate processing image, that is, the intermediate processing image is converted into a binary image to obtain a target image, perform defect identification on the target image to obtain a defect area and edge pixel points of the defect area, and position and quantify the defect area according to the edge pixel points of the defect area to complete defect processing of the ultrasonic full-focus image.
In the embodiment, the kernel density estimated value of each pixel point is compared with the ultrasonic intensity threshold value to make a decision to obtain the intermediate processing image, so that the artifact in the ultrasonic full-focus image can be effectively eliminated, the influence of the artifact on the positioning and quantification of the subsequent defects can be avoided as much as possible, and the data volume of the ultrasonic full-focus image can be effectively reduced by positioning and quantifying the target image, namely the defect area in the binary image obtained by performing binarization processing on the intermediate processing image, so that the defects in the ultrasonic full-focus image can be positioned and quantified with lower processing amount, and the accuracy and efficiency of the defect processing of the ultrasonic full-focus image can be improved.
In a preferred embodiment, the performing image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image specifically includes: and denoising the ultrasonic full-focus image by adopting a median filter, and normalizing the denoised ultrasonic full-focus image according to the image matrix to obtain a preprocessed image.
Exemplarily, a median filter is adopted to perform denoising processing on the ultrasonic full-focus image to obtain a denoised ultrasonic full-focus image, and normalization processing is performed on the denoised ultrasonic full-focus image according to an image matrix to obtain a preprocessed image, which specifically includes:
in equation (5), max (x) is the maximum pixel value in the image matrix; min (x) is the minimum pixel value in the image matrix; x is the number ofiThe pixel value of the ith pixel point of the denoised ultrasonic full focusing image is obtained; x'iThe pixel value of the ith pixel point of the preprocessed image is obtained.
In the embodiment, the median filter is adopted to denoise the ultrasonic full-focus image, so that the noise interference smooth image can be effectively eliminated, and the denoised ultrasonic full-focus image can be effectively ensured to be a standard image by normalizing the denoised ultrasonic full-focus image according to the image matrix.
In a preferred embodiment, the obtaining of the ultrasonic intensity threshold and the kernel density estimation value of each pixel point based on the preprocessed image specifically includes: based on the preprocessed image, taking the pixel quantity of which the ultrasonic intensity is greater than the threshold value as a target function, enabling a first derivative of the target function to be converged to zero, and taking the corresponding threshold value as an ultrasonic intensity threshold value; and respectively inputting each pixel point of the preprocessed image into a pre-constructed kernel density estimation model to obtain a kernel density estimation value of each pixel point.
Illustratively, based on the preprocessed image, let the image matrix after the image normalization process be a, the threshold value δ be 0.1, 0.2, …, 1, and take the pixel amount of the ultrasound intensity greater than the threshold value δ as the objective function S (δ), then the objective function is expressed as S (δ) sum (sum (a > δ)), then normalize the obtained S (δ), and when the first derivative S '(δ) of S (δ) after the normalization process converges to zero, solve to obtain an optimal threshold value δ, while the change of S (δ) is insignificant, make the first derivative S' (δ) of the objective function converge to zero, and take the corresponding threshold value δ as the ultrasound intensity threshold value.
A kernel density estimation model is constructed in advance, a Gaussian kernel is selected by the model to serve as a kernel function of kernel density estimation, and then the kernel density estimation function at any pixel point is as follows:
in the formula (2), n is the number of pixel points; h is the optimal bandwidth of the kernel density estimation; x is the pixel value of any pixel point; x is the number ofjIn practical application, the pixel value of the jth pixel point of the ultrasonic full-focus image after normalization processing is used as a kernel density estimation functionConsidered as a normal distribution, the optimal bandwidth isThe standard deviation of the pixel values of all the pixel points.
Respectively inputting each pixel point of the preprocessed image into a kernel density estimation model to obtain a kernel density estimation value of each pixel point
In an embodiment of the present invention, the step of setting the pixel value of the corresponding pixel point according to the comparison result between the kernel density estimation value of each pixel point and the ultrasonic intensity threshold to obtain the intermediate processing image specifically includes: and comparing the kernel density estimated value of the pixel point with the ultrasonic intensity threshold, if the kernel density estimated value of the pixel point is smaller than the ultrasonic intensity threshold, setting the pixel value of the pixel point as a first preset pixel value, and otherwise, setting the pixel value of the pixel point as a second preset pixel value to obtain an intermediate processing image.
In a preferred implementation manner of this embodiment, the first predetermined pixel value is 255, and the second predetermined pixel value is other values except 255.
Illustratively, for each pixel point of the intermediate processing image, comparing the kernel density estimation value of the pixel point with the ultrasonic intensity threshold, if the kernel density estimation value of the pixel point is smaller than the ultrasonic intensity threshold, setting the pixel value of the pixel point to be a first preset pixel value 255, that is, setting the pixel point to be white, otherwise, setting the pixel value of the pixel point to be a second preset pixel value, and setting the pixel point to be other colors.
In the embodiment, the nuclear density estimated value of each pixel point is compared with the ultrasonic intensity threshold to make a decision, so that the intermediate processing image is obtained, the artifact in the ultrasonic full-focus image can be effectively eliminated, the influence of the artifact on the positioning and quantification of the subsequent defects is avoided as much as possible, and the accuracy of the defect processing of the ultrasonic full-focus image is improved.
In a preferred embodiment, the locating and quantifying the defect region in the target image specifically includes: performing defect identification on the target image to obtain a defect area and leftmost pixel points, rightmost pixel points, uppermost pixel points and lowermost pixel points of the defect area; obtaining a first line segment according to the leftmost pixel point and the rightmost pixel point, obtaining a second line segment according to the uppermost pixel point and the bottommost pixel point, and taking the intersection point of the first line segment and the second line segment as the central point of the defect area; and calculating to obtain the size of the defect region by combining the leftmost pixel point, the rightmost pixel point, the topmost pixel point and the bottommost pixel point.
Illustratively, defect identification is carried out on the target image to obtain a defect area and a leftmost pixel point A (x) of the defect areaA,yA) Rightmost pixel point B (x)B,yB) The top pixel C (x)C,yC) Most preferablyLower pixel D (x)D,yD)。
According to the leftmost pixel point A (x)A,yA) And the rightmost pixel point B (x)B,yB) A first segment AB is obtained, namely:
in the formula (7), the coordinates of the point on the first line segment AB are (x)1,y1);
According to the uppermost pixel point C (x)C,yC) And the lowest pixel D (x)D,yD) A second line segment CD is obtained, namely:
in the formula (8), the coordinates of the point on the second line segment CD are (x)2,y2);
Let y1=y2And solving to obtain the intersection point of the first line segment AB and the second line segment CD, and using the intersection point of the first line segment AB and the second line segment CD as the central point of the defect area to realize the positioning of the defect.
Combining the leftmost pixel point A (x)A,yA) Rightmost pixel point B (x)B,yB) The top pixel C (x)C,yC) The bottom pixel D (x)D,yD) Performing coordinate calculation to obtain the size of the defect region, such as the length l ═ x of the defect regionA-xB| and Width w ═ yC-yDAnd quantifying the defects.
In the embodiment, the data volume of the ultrasonic full-focus image can be effectively reduced by positioning and quantifying the defect area in the target image, namely the binary image obtained by performing binarization processing on the intermediate processing image, and the defects in the ultrasonic full-focus image are positioned and quantified with lower processing amount, so that the efficiency of defect processing of the ultrasonic full-focus image is improved.
In summary, the embodiment of the present invention has the following advantages:
the method comprises the steps of obtaining an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, conducting image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image, obtaining an ultrasonic intensity threshold value and a nuclear density estimated value of each pixel point based on the preprocessed image, setting a pixel value of a corresponding pixel point according to a comparison result of the nuclear density estimated value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image, conducting binarization processing on the intermediate processed image to obtain a target image, locating and quantifying a defect area in the target image, and completing defect processing of the ultrasonic full-focus image. According to the embodiment of the invention, the nuclear density estimated value of each pixel point is compared with the ultrasonic intensity threshold value to make a decision to obtain the intermediate processing image, so that the artifact in the ultrasonic full-focus image can be effectively eliminated, the influence of the artifact on the positioning and quantification of the subsequent defects can be avoided as far as possible, and the data volume of the ultrasonic full-focus image can be effectively reduced by positioning and quantifying the defect area in the binary image obtained by performing binarization processing on the target image, namely the intermediate processing image, so that the defects in the ultrasonic full-focus image can be positioned and quantified with lower processing amount, and the accuracy and efficiency of the defect processing of the ultrasonic full-focus image can be improved.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
It will be understood by those skilled in the art that all or part of the processes of the above embodiments may be implemented by hardware related to instructions of a computer program, and the computer program may be stored in a computer readable storage medium, and when executed, may include the processes of the above embodiments. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
Claims (10)
1. A defect processing method of an ultrasonic full-focus image based on nuclear density estimation is characterized by comprising the following steps:
acquiring an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, and performing image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image;
obtaining an ultrasonic intensity threshold value and a kernel density estimation value of each pixel point based on the preprocessed image, and setting a pixel value corresponding to each pixel point according to a comparison result of the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image;
and carrying out binarization processing on the intermediate processing image to obtain a target image, and positioning and quantifying a defect area in the target image.
2. The method for defect processing of an ultrasonic full-focus image based on kernel density estimation as claimed in claim 1, wherein the image preprocessing is performed on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image, and specifically comprises:
and denoising the ultrasonic full-focus image by adopting a median filter, and normalizing the denoised ultrasonic full-focus image according to the image matrix to obtain the preprocessed image.
3. The method for processing the defect of the ultrasonic full-focus image based on the kernel density estimation as claimed in claim 1, wherein the obtaining of the ultrasonic intensity threshold and the kernel density estimation value of each pixel point based on the preprocessed image specifically comprises:
based on the preprocessed image, taking the pixel quantity of which the ultrasonic intensity is greater than a threshold value as an objective function, converging a first derivative of the objective function to zero, and taking the corresponding threshold value as the ultrasonic intensity threshold value;
and respectively inputting each pixel point of the preprocessed image into a pre-constructed kernel density estimation model to obtain a kernel density estimation value of each pixel point.
4. The method for processing defects in an ultrasonic full-focus image based on kernel density estimation according to claim 1, wherein the pixel values corresponding to the pixel points are set according to the comparison result between the kernel density estimation value of each pixel point and the ultrasonic intensity threshold, so as to obtain an intermediate processed image, specifically:
comparing the kernel density estimation value of the pixel point with the ultrasonic intensity threshold, if the kernel density estimation value of the pixel point is smaller than the ultrasonic intensity threshold, setting the pixel value of the pixel point as a first preset pixel value, otherwise, setting the pixel value of the pixel point as a second preset pixel value, and obtaining the intermediate processing image.
5. The method for processing the defect of the ultrasonic full-focus image based on the nuclear density estimation as claimed in claim 1, wherein the locating and quantifying the defect region in the target image are specifically as follows:
performing defect identification on the target image to obtain the defect area and the leftmost pixel point, the rightmost pixel point, the uppermost pixel point and the lowermost pixel point of the defect area;
obtaining a first line segment according to the leftmost pixel point and the rightmost pixel point, obtaining a second line segment according to the uppermost pixel point and the bottommost pixel point, and taking the intersection point of the first line segment and the second line segment as the central point of the defect area;
and calculating to obtain the size of the defect region by combining the leftmost pixel point, the rightmost pixel point, the topmost pixel point and the bottommost pixel point.
6. An apparatus for defect processing of an ultrasonic full-focus image based on nuclear density estimation, comprising:
the preprocessing module is used for acquiring an ultrasonic full-focus image and an image matrix of the ultrasonic full-focus image, and carrying out image preprocessing on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image;
the intermediate processing module is used for obtaining an ultrasonic intensity threshold value and a kernel density estimation value of each pixel point based on the preprocessed image, and setting a pixel value corresponding to each pixel point according to a comparison result of the kernel density estimation value of each pixel point and the ultrasonic intensity threshold value to obtain an intermediate processed image;
and the defect processing module is used for carrying out binarization processing on the intermediate processing image to obtain a target image and positioning and quantifying a defect area in the target image.
7. The apparatus for defect processing of ultrasonic full-focus image based on nuclear density estimation according to claim 6, wherein the image preprocessing is performed on the ultrasonic full-focus image according to the image matrix to obtain a preprocessed image, and specifically:
and denoising the ultrasonic full-focus image by adopting a median filter, and normalizing the denoised ultrasonic full-focus image according to the image matrix to obtain the preprocessed image.
8. The apparatus for defect processing of ultrasonic full-focus images based on kernel density estimation as claimed in claim 6, wherein the obtaining of the ultrasonic intensity threshold and the kernel density estimation value of each pixel point based on the preprocessed image specifically comprises:
based on the preprocessed image, taking the pixel quantity of which the ultrasonic intensity is greater than a threshold value as an objective function, converging a first derivative of the objective function to zero, and taking the corresponding threshold value as the ultrasonic intensity threshold value;
and respectively inputting each pixel point of the preprocessed image into a pre-constructed kernel density estimation model to obtain a kernel density estimation value of each pixel point.
9. The apparatus for processing defect of ultrasonic full-focus image based on kernel density estimation according to claim 6, wherein the pixel value corresponding to the pixel point is set according to the comparison result between the kernel density estimation value of each pixel point and the ultrasonic intensity threshold, so as to obtain an intermediate processing image, specifically:
comparing the kernel density estimation value of the pixel point with the ultrasonic intensity threshold, if the kernel density estimation value of the pixel point is smaller than the ultrasonic intensity threshold, setting the pixel value of the pixel point as a first preset pixel value, otherwise, setting the pixel value of the pixel point as a second preset pixel value, and obtaining the intermediate processing image.
10. The apparatus for defect processing of ultrasonic full-focus image based on nuclear density estimation as claimed in claim 6, wherein the localization and quantification of the defect region in the target image are specifically:
performing defect identification on the target image to obtain the defect area and the leftmost pixel point, the rightmost pixel point, the uppermost pixel point and the lowermost pixel point of the defect area;
obtaining a first line segment according to the leftmost pixel point and the rightmost pixel point, obtaining a second line segment according to the uppermost pixel point and the bottommost pixel point, and taking the intersection point of the first line segment and the second line segment as the central point of the defect area;
and calculating to obtain the size of the defect region by combining the leftmost pixel point, the rightmost pixel point, the topmost pixel point and the bottommost pixel point.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110979361.5A CN113838001B (en) | 2021-08-24 | 2021-08-24 | Ultrasonic wave full focusing image defect processing method and device based on nuclear density estimation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110979361.5A CN113838001B (en) | 2021-08-24 | 2021-08-24 | Ultrasonic wave full focusing image defect processing method and device based on nuclear density estimation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113838001A true CN113838001A (en) | 2021-12-24 |
CN113838001B CN113838001B (en) | 2024-02-13 |
Family
ID=78961220
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110979361.5A Active CN113838001B (en) | 2021-08-24 | 2021-08-24 | Ultrasonic wave full focusing image defect processing method and device based on nuclear density estimation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113838001B (en) |
Citations (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080051660A1 (en) * | 2004-01-16 | 2008-02-28 | The University Of Houston System | Methods and apparatuses for medical imaging |
CN102737382A (en) * | 2012-06-22 | 2012-10-17 | 刘怡光 | Automatic precise partition method for prostate ultrasonic image |
CN103578083A (en) * | 2013-09-07 | 2014-02-12 | 西安电子科技大学 | Single image defogging method based on joint mean shift |
CN104504720A (en) * | 2015-01-07 | 2015-04-08 | 四川大学 | New prostate ultrasonoscopy segmentation technique |
CN106344068A (en) * | 2016-10-09 | 2017-01-25 | 深圳市普罗惠仁医学科技有限公司 | Water bag device for high-intensity focused ultrasound system |
US20170278235A1 (en) * | 2016-03-24 | 2017-09-28 | Hong Kong Applied Science and Technology Research Institute Company Limited | Fast density estimation method for defect inspection application |
US20190114833A1 (en) * | 2018-12-05 | 2019-04-18 | Intel Corporation | Surface reconstruction for interactive augmented reality |
CN110475505A (en) * | 2017-01-27 | 2019-11-19 | 阿特瑞斯公司 | Utilize the automatic segmentation of full convolutional network |
CN110646512A (en) * | 2019-09-18 | 2020-01-03 | 广东工业大学 | Single-channel bidirectional transmitting-receiving ultrasonic guided wave pipeline monitoring system and imaging method |
CN111007151A (en) * | 2019-12-30 | 2020-04-14 | 华东理工大学 | Ultrasonic phased array rapid full-focusing imaging detection method based on defect pre-positioning |
CN111093513A (en) * | 2017-09-14 | 2020-05-01 | 神经***分析公司 | System and method for registering a headset system |
CN111528921A (en) * | 2020-05-27 | 2020-08-14 | 广东工业大学 | Information acquisition device |
CN112067698A (en) * | 2020-09-14 | 2020-12-11 | 南昌航空大学 | Time-frequency combined rapid full-focusing ultrasonic imaging method |
CN212159692U (en) * | 2020-03-11 | 2020-12-15 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Power station steam turbine T type blade root ultrasonic phased array detects simulation defect test block |
CN213270159U (en) * | 2020-10-27 | 2021-05-25 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Wind turbine blade surface damage detection device |
CN112950634A (en) * | 2021-04-22 | 2021-06-11 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Method, equipment and system for identifying damage of wind turbine blade based on unmanned aerial vehicle routing inspection |
CN115452944A (en) * | 2022-09-05 | 2022-12-09 | 吉林大学 | Plate-shaped material multi-damage positioning method based on L-shaped sensor cluster |
WO2022257759A1 (en) * | 2021-06-08 | 2022-12-15 | 百果园技术(新加坡)有限公司 | Image banding artifact removal method and apparatus, and device and medium |
-
2021
- 2021-08-24 CN CN202110979361.5A patent/CN113838001B/en active Active
Patent Citations (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080051660A1 (en) * | 2004-01-16 | 2008-02-28 | The University Of Houston System | Methods and apparatuses for medical imaging |
CN102737382A (en) * | 2012-06-22 | 2012-10-17 | 刘怡光 | Automatic precise partition method for prostate ultrasonic image |
CN103578083A (en) * | 2013-09-07 | 2014-02-12 | 西安电子科技大学 | Single image defogging method based on joint mean shift |
CN104504720A (en) * | 2015-01-07 | 2015-04-08 | 四川大学 | New prostate ultrasonoscopy segmentation technique |
US20170278235A1 (en) * | 2016-03-24 | 2017-09-28 | Hong Kong Applied Science and Technology Research Institute Company Limited | Fast density estimation method for defect inspection application |
CN106344068A (en) * | 2016-10-09 | 2017-01-25 | 深圳市普罗惠仁医学科技有限公司 | Water bag device for high-intensity focused ultrasound system |
CN110475505A (en) * | 2017-01-27 | 2019-11-19 | 阿特瑞斯公司 | Utilize the automatic segmentation of full convolutional network |
CN111093513A (en) * | 2017-09-14 | 2020-05-01 | 神经***分析公司 | System and method for registering a headset system |
US20190114833A1 (en) * | 2018-12-05 | 2019-04-18 | Intel Corporation | Surface reconstruction for interactive augmented reality |
CN110646512A (en) * | 2019-09-18 | 2020-01-03 | 广东工业大学 | Single-channel bidirectional transmitting-receiving ultrasonic guided wave pipeline monitoring system and imaging method |
CN111007151A (en) * | 2019-12-30 | 2020-04-14 | 华东理工大学 | Ultrasonic phased array rapid full-focusing imaging detection method based on defect pre-positioning |
CN212159692U (en) * | 2020-03-11 | 2020-12-15 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Power station steam turbine T type blade root ultrasonic phased array detects simulation defect test block |
CN111528921A (en) * | 2020-05-27 | 2020-08-14 | 广东工业大学 | Information acquisition device |
CN112067698A (en) * | 2020-09-14 | 2020-12-11 | 南昌航空大学 | Time-frequency combined rapid full-focusing ultrasonic imaging method |
CN213270159U (en) * | 2020-10-27 | 2021-05-25 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Wind turbine blade surface damage detection device |
CN112950634A (en) * | 2021-04-22 | 2021-06-11 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Method, equipment and system for identifying damage of wind turbine blade based on unmanned aerial vehicle routing inspection |
WO2022257759A1 (en) * | 2021-06-08 | 2022-12-15 | 百果园技术(新加坡)有限公司 | Image banding artifact removal method and apparatus, and device and medium |
CN115452944A (en) * | 2022-09-05 | 2022-12-09 | 吉林大学 | Plate-shaped material multi-damage positioning method based on L-shaped sensor cluster |
Non-Patent Citations (5)
Title |
---|
NICOLAS BUDYN等: "Characterisation of small embedded two-dimensional defects using multi-view Total Focusing Method imaging algorithm", 《NDT AND E INTERNATIONAL》, pages 1 - 11 * |
孙正等: "生物光声成像中声反射伪影抑制方法的研究进展", 《计算机科学》, pages 10 - 14 * |
宋寿鹏等: "超声全聚焦成像中等声程线伪影剔除方法", 《应用升学》, pages 527 - 534 * |
李灏天等: "基于机器学习和超声成像的缺陷识别与分析", 《南京大学学报(自然科学)》, pages 670 - 679 * |
邹丹等: "核密度估计的道路匹配方法探讨", 《测绘科学》, pages 104 - 108 * |
Also Published As
Publication number | Publication date |
---|---|
CN113838001B (en) | 2024-02-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5920994B2 (en) | Method and system for identifying well wall boundaries of microplates | |
US8131107B2 (en) | Method and system for identifying defects in NDT image data | |
CN113592845A (en) | Defect detection method and device for battery coating and storage medium | |
CN111292305A (en) | Improved YOLO-V3 metal processing surface defect detection method | |
CN112184693B (en) | Intelligent detection method for welding line defects of ray industrial negative film | |
Tyystjärvi et al. | Automated defect detection in digital radiography of aerospace welds using deep learning | |
CN101901342A (en) | Method and device for extracting image target region | |
Saif et al. | Gradient based image edge detection | |
Wang et al. | Application of gradient-based Hough transform to the detection of corrosion pits in optical images | |
CN111275659B (en) | Weld image processing method and device, terminal equipment and storage medium | |
CN110659675A (en) | Welding seam defect detection method based on AdaBoost algorithm | |
Bellens et al. | Evaluating conventional and deep learning segmentation for fast X-ray CT porosity measurements of polymer laser sintered AM parts | |
CN117557565B (en) | Detection method and device for lithium battery pole piece | |
CN115861226A (en) | Method for intelligently identifying surface defects by using deep neural network based on characteristic value gradient change | |
CN110426395B (en) | Method and device for detecting surface of solar EL battery silicon wafer | |
CN108648199B (en) | Ultrasonic phased array NDT image segmentation method based on watershed and CV models | |
CN116934752B (en) | Glass detection method and system based on artificial intelligence | |
Niu et al. | A generalized well neural network for surface defect segmentation in Optical Communication Devices via Template-Testing comparison | |
CN107478656A (en) | Paper pulp mixing effect method of determination and evaluation based on machine vision, device, system | |
CN113838001B (en) | Ultrasonic wave full focusing image defect processing method and device based on nuclear density estimation | |
Klinder et al. | Lobar fissure detection using line enhancing filters | |
CN113807206B (en) | SAR image target identification method based on denoising task assistance | |
CN111415365A (en) | Image detection method and device | |
Jayasudha et al. | Weld defect segmentation and feature extraction from the acquired phased array scan images | |
Chiraz et al. | Automatic detection of weld defects based on hough transform |
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 |