CN117876341A - Conductive foam quality detection method - Google Patents
Conductive foam quality detection method Download PDFInfo
- Publication number
- CN117876341A CN117876341A CN202410056049.2A CN202410056049A CN117876341A CN 117876341 A CN117876341 A CN 117876341A CN 202410056049 A CN202410056049 A CN 202410056049A CN 117876341 A CN117876341 A CN 117876341A
- Authority
- CN
- China
- Prior art keywords
- growth
- region
- growing
- point
- conductive foam
- 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
- 239000006260 foam Substances 0.000 title claims abstract description 99
- 238000001514 detection method Methods 0.000 title abstract description 36
- 230000012010 growth Effects 0.000 claims abstract description 434
- 238000000034 method Methods 0.000 claims abstract description 114
- 238000009826 distribution Methods 0.000 claims abstract description 45
- 238000012937 correction Methods 0.000 claims description 24
- 238000012797 qualification Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 abstract description 10
- 238000004519 manufacturing process Methods 0.000 description 12
- 238000004458 analytical method Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 239000000463 material Substances 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of industrial visual detection, in particular to a conductive foam quality detection method. Firstly, acquiring a cross-sectional image of conductive foam, further acquiring an initial growth point and performing first region growth; further acquiring a growth threshold value of each region growth, performing region growth, stopping region growth when a region growth termination condition is reached, and obtaining a final growth region of each initial growth point; further analyzing the change of the growth threshold value and the change of the number of the growth pixel points in the growth process of each final growth area, obtaining the density distribution uniformity corresponding to each initial growth point, and detecting the quality of the conductive foam. According to the invention, the distribution characteristics of the sponge density around each wire are analyzed in a self-adaptive area growth mode, the uniformity of the sponge density distribution is detected in a visual detection mode, the quality detection of the conductive foam is rapidly realized, and the efficiency and the accuracy of the quality detection of the conductive foam are improved.
Description
Technical Field
The invention relates to the technical field of industrial visual detection, in particular to a conductive foam quality detection method.
Background
In order to ensure the normal operation of the equipment, the quality detection of the conductive foam is an important link in industrial production. If the density distribution of the sponge is uneven, the conductivity of the sponge may be different in the material, so that the electromagnetic shielding effect of a part of the area is poor, and the performance of the whole product is affected; the non-uniform strength of the conductive foam under mechanical stress may be caused, which may cause local deformation or breakage of the product during use, and uneven distribution may cause uniformity problems during production, which may affect reliability and uniformity of the product, especially in applications requiring high precision, so that uniformity detection of sponge density distribution of the conductive foam is necessary.
The quality detection of the conductive foam is performed in a real-time, automatic, efficient and high-precision manner by a visual detection manner, so that human errors can be reduced and the manufacturing cost can be reduced; the area growth can provide information about the uniformity of the conductive foam through the generated area, so that the quality of the conductive foam can be detected; however, the growth result is affected by the parameter setting of the region growth, the subjectivity of the artificial setting of the growth threshold is too strong, so that the reliability and accuracy of the region growth result are low, and the conductive foam cannot be accurately detected in quality.
Disclosure of Invention
In order to solve the technical problems that the density uniformity analysis of the conductive foam is not accurate enough and the quality detection accuracy is affected in the existing region growth algorithm, the invention aims to provide a conductive foam quality detection method, and the adopted technical scheme is as follows:
acquiring a cross-sectional image of the conductive foam;
acquiring initial growth points based on wire pixel points in the cross-sectional image, and carrying out first region growth according to a preset mode;
the pixel points newly added in each region growing process are called growing pixel points; acquiring a growth threshold according to the difference characteristic of the gray value of the growth pixel point grown in the initial growth point and the first region;
and for any subsequent region growing process, carrying out primary region growing according to the growing threshold of the current region growing, wherein the growing range of the primary region growing is a preset neighborhood of the growing pixel point in the previous region growing process, and the acquiring process of the growing threshold of the current region growing comprises the following steps: acquiring correction parameters corresponding to the current region growth according to the gray level difference between the growth pixel points in the previous region growth process and the pixel points in the preset adjacent region which do not belong to the growth region in the previous region growth process; correcting the growth threshold value in the last region growth by using the correction parameters corresponding to the region growth to obtain the growth threshold value of the region growth; stopping the region growth when the region growth termination condition is reached, and obtaining a final growth region of each initial growth point;
analyzing the change of the growth threshold value and the change of the number of the growth pixels in the growth process of the final growth area of each initial growth point to obtain the density distribution uniformity corresponding to each initial growth point; and detecting the quality of the conductive foam according to the density distribution uniformity corresponding to each initial growth point.
Further, the method for acquiring the correction parameters includes:
and obtaining the absolute value of the difference value of the gray value of the growing pixel point in the last region growing process and the pixel point which does not belong to the growing region in the preset adjacent region, averaging all the absolute values of the difference value, and finally normalizing to obtain the correction parameter.
Further, the method for acquiring the growth threshold value comprises the following steps:
for the first growth: obtaining a difference value between the average gray value of each initial growth point and the minimum gray value of the growth pixel points grown in the first region, and taking the difference value as a growth threshold value of each initial growth point;
for any one of the following region growing processes: and taking the product of the correction parameter corresponding to the current region growth and the growth threshold value in the last region growth as the growth threshold value of the current region growth.
Further, the method for obtaining the density distribution uniformity comprises the following steps:
taking the change rate of the growth threshold value in the whole growth process of the initial growth point as a first parameter;
summing the ratio of the number of the growing pixels in the adjacent two growing processes in the whole growing process of the initial growing point to obtain a second parameter;
and obtaining the product of the first parameter and the second parameter of each initial growth point, and taking the ratio of the product to a preset density experience parameter as a density distribution uniformity parameter of each initial growth point.
Further, the method for acquiring the first parameter includes:
acquiring a first parameter by using a first parameter calculation formula; the first parameter calculation formula includes:
wherein D1 i A first parameter representing the ith initial growth point, Q i,j-1 Representing a corresponding growth threshold value when the j-1 th zone of the i initial growth point grows; q (Q) i,j Representing a growth threshold corresponding to the jth region of the ith initial growth point during growth; j (J) i Indicating the total number of growths at the i-th initial growth point.
Further, the method for detecting the quality of the conductive foam comprises the following steps:
when the density distribution uniformity parameters of all the initial growth points of the conductive foam are larger than a preset quality qualification threshold, the quality of the current conductive foam is qualified, otherwise, the quality of the current conductive foam is considered to be problematic.
Further, the conditions for achieving the regional growth termination are:
and when all the pixels which do not belong to the growth area in the growth range of the current growth area do not accord with the growth threshold value, or when the pixels which do not belong to the growth area do not exist in the growth range of the current growth area, the area growth termination condition is reached.
Further, the method for obtaining the wire pixel points comprises the following steps:
and processing the section image by using an Ojin threshold algorithm to obtain a binarized section image, and taking a pixel point with the value of 1 in the binarized section image as a wire pixel point.
Further, the method for performing the first region growth according to the preset mode comprises the following steps:
and (5) outwards growing the initial growth point by a circle of pixel points to finish the first region growth.
Further, the preset quality pass threshold is 0.75.
The invention has the following beneficial effects:
firstly, acquiring a cross-sectional image of the conductive foam, and providing an analysis basis for the subsequent analysis of specific characteristics of the conductive foam; further acquiring an initial growth point and performing first region growth, avoiding overlarge gray scale difference between the initial growth point and surrounding pixel points, failing to perform region growth, acquiring a growth threshold value of the first region growth, and providing a basis for realizing self-adaptive growth by subsequently adjusting the growth threshold value; further, according to the gray level difference between the growing pixel point in the last region growing process and the pixel points which do not belong to the growing region in the preset adjacent region, the growing threshold value in the last region growing process is adjusted, the growing threshold value in each region growing process is obtained, the gray level characteristic consistency of the growing pixel point in each region growing process and the growing pixel point in the last region growing process is enhanced, the uniform density distribution characteristic of the pixel points is analyzed, and meanwhile, the termination of the region growing process is promoted; further carrying out region growth according to the growth threshold value of the region growth, stopping the region growth when the region growth termination condition is reached, obtaining the final growth region of each initial growth point, obtaining the self-adaptive region growth result of each initial growth point, avoiding the condition that the region growth result is unsuitable due to human factors, and providing an accurate basis for analyzing the density distribution characteristics of the conductive foam; further analyzing the change of the growth threshold value and the change of the number of the growth pixel points in the growth process of each final growth area, obtaining the density distribution uniformity corresponding to each initial growth point, realizing the quality detection of the conductive foam in a visual detection mode, improving the production efficiency, reducing the labor detection cost and the influence of subjective factors, and providing a guarantee for the production of the conductive foam. According to the invention, the distribution characteristics of the sponge density around each wire are analyzed in a self-adaptive area growth mode, the uniformity of the sponge density distribution is detected in a visual detection mode, the quality detection of the conductive foam is rapidly realized, and the efficiency and the accuracy of the quality detection of the conductive foam are improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for detecting quality of conductive foam according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a region growing process according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to specific implementation, structure, characteristics and effects of the conductive foam quality detection method according to the invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the conductive foam quality detection method provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for detecting quality of conductive foam according to an embodiment of the present invention is shown, which specifically includes:
the invention aims at quality detection of conductive foam, so that the embodiment of the invention firstly acquires a cross-sectional image of the conductive foam, acquires initial growth points for self-adaptive region growth, analyzes the change of a growth threshold value and the change characteristics of growth pixel points in the growth process, and acquires the density distribution uniformity corresponding to each initial growth point, thereby carrying out quality evaluation detection on the conductive foam.
Step S1: and acquiring a cross-sectional image of the conductive foam.
The initial cross-section image of the conductive foam is shot and acquired by using a high-definition camera, and other irrelevant areas, such as background areas, in the initial cross-section image are not needed because the density uniformity characteristics in the cross-section of the conductive foam need to be analyzed, and the interference to the subsequent processing process is possibly generated, so that the images of other irrelevant areas are firstly eliminated, and the cross-section image of the conductive foam is acquired. Meanwhile, compared with a color image, the gray level image is considered to have lower calculation complexity and higher calculation efficiency when in visual detection, so that the conductive foam image is analyzed in a gray level mode.
In one embodiment of the invention, the context information in the image can be considered by the semantic neural network, so that not only the information of a single pixel can be considered, but also the relation between objects can be better processed, the wire part is prevented from being removed from the section, the accuracy of image segmentation is improved, and therefore, the pre-trained semantic segmentation neural network is utilized to identify the acquired initial cross-section image of the conductive foam, and the cross-section image of the conductive foam is obtained.
In one embodiment of the invention, a semantic segmentation U-net neural network is adopted to acquire a cross-sectional image, a large number of conductive foam initial cross-sectional images acquired in actual production are collected to serve as a training set, a part of images are artificially marked to acquire a test set, and a cross entropy loss function is adopted for training to acquire a pre-trained semantic segmentation neural network. In other embodiments of the present invention, the practitioner may choose to train other types of neural networks, or use an edge detection method to obtain the cross-sectional outline of the conductive foam, so as to obtain the cross-sectional image of the conductive foam.
Step S2: and acquiring initial growth points based on the wire pixel points in the sectional image, and carrying out first region growth according to a preset mode.
In the embodiment of the invention, in consideration of the obvious distinguishing characteristics of the wire part and the sponge part in the cross section of the conductive foam, the density distribution of the sponge around the wire determines the electromagnetic shielding effect of a partial area, and the product is possibly deformed or damaged locally in the use process to influence the performance of the whole product, so that the pixel points of the wire are determined as seed points for the growth of the area; because the cross section of the wire exists, in order to avoid the repeated region growth of the surrounding region of the same wire, the adjacent wire pixel points are integrated into an initial growth point.
It should be noted that, when an isolated wire pixel occurs, the isolated wire pixel is considered as image noise and is not taken as an initial growth point, because the wire usually has a certain width and a certain cross section and is not usually represented as an isolated pixel in an image; when a large number of isolated wire pixel points appear, the current acquisition environment is larger in noise or the image resolution is insufficient, and timely adjustment is needed.
Preferably, in one embodiment of the present invention, in consideration of the fact that the conductive line material has stronger reflection property than the sponge material and the gray scale difference between the conductive line pixel point and the sponge pixel point is larger, the sectional image is processed by using the oxford thresholding algorithm to obtain a binary sectional image, and the pixel point with the value of 1 in the binary sectional image is taken as the conductive line pixel point. In other embodiments of the present invention, the practitioner may select other image segmentation methods, such as Sobel operator, to separate the lead portion from the sponge portion, and obtain the lead pixel.
In the embodiment of the invention, the fact that the gray value difference between the initial growth point and the neighborhood pixel points around the initial growth point is larger and the gray value difference between the sponges is smaller is also considered, so that all the sponges are easily and directly contained in the growth area by setting the growth threshold for area growth, the effect of analyzing the characteristics of uniform density cannot be achieved, and special treatment is needed for the first growth.
Preferably, in one embodiment of the present invention, in consideration of the fact that after the initial growth point is outwardly expanded by one circle, the sponge pixel points are connected with each other, and then the difference between the sponge pixel points can be analyzed for region growth, the initial growth point is outwardly grown by one circle of pixel points, and the first region growth is completed. In other embodiments of the present invention, the practitioner may set other first-time region growing modes, and perform special treatment on the first-time region growth, for example, the initial growing points are similar to the surrounding sponge pixels caused by the production process, and when the surrounding sponge pixels of the initial growing points have a large difference from other sponge pixels, the first-time region growth can be controlled to grow for several circles of pixels.
It should be noted that, in the embodiment of the present invention, in order to improve the visual inspection efficiency, all the initial growing points start to grow at the same time.
Step S3: the pixel points newly added in each region growing process are called growing pixel points; and acquiring a growth threshold according to the difference characteristic of the gray value of the growth pixel point grown in the initial growth point and the first region.
In the embodiment of the invention, the subjectivity of the growth threshold is considered to be too strong through manual setting, so that the growth threshold is continuously regulated and controlled through the gray change characteristics of the sponge pixel points in the region growth process, the self-adaptive region growth is realized, and the final growth region of each initial growth point is obtained.
The adjustment of the growth threshold value firstly needs to have an adjustment basis, and the first growth process is considered to be capable of obtaining the growth threshold value, so that in the embodiment of the invention, the pixel point newly added in each growth process is called a growth pixel point; and acquiring a growth threshold according to the difference characteristic of the gray value of the initial growth point and the gray value of the growth pixel point grown for the first time.
Preferably, in one embodiment of the present invention, a difference value between the average gray value of each initial growth point and the minimum gray value of the grown pixel point of the first growth is obtained, and the difference value is used as the growth threshold value of the growth of the first region corresponding to each initial growth point.
Step S4: for any subsequent region growing process, performing region growing once according to the growing threshold of the region growing once, wherein the growing range of the region growing once is a preset neighborhood of growing pixel points in the previous region growing process, and the acquiring process of the growing threshold of the region growing once comprises the following steps: acquiring correction parameters corresponding to the current region growth according to the gray difference between the growth pixel points in the previous region growth process and the pixel points in the preset adjacent region which do not belong to the growth region in the previous region growth process; correcting the growth threshold value in the last region growth by using the correction parameters corresponding to the region growth to obtain the growth threshold value of the region growth; and stopping the region growth when the region growth termination condition is reached, and obtaining a final growth region of each initial growth point.
After the growth threshold value of the first region growth is obtained as a basis, specific characteristics of the image can be analyzed, the growth threshold value is adjusted once, the growth threshold value of each region growth is obtained, the self-adaptive region growth is realized, and the final growth region of each initial growth point is obtained.
In the embodiment of the invention, the growth threshold is obtained through the difference between the growth pixel point and the initial growth point in the first growth, so the growth threshold is larger, however, the gray scale difference between sponges is not larger than the difference between the lead and the sponge, and the growth process of the region needs to be stopped, so the growth threshold needs to be continuously reduced; and in addition, the smaller the gray difference between the growing pixel point and the pixel point which does not belong to the growing region in the neighborhood pixel point in each growing process is, the more the growing threshold value needs to be reduced, and the more consistent the density characteristics are ensured in each growing process, so that the correction parameters corresponding to the growing of the region are obtained according to the gray difference between the growing pixel point in the previous region growing process and the pixel point which does not belong to the growing region in the preset neighborhood.
Preferably, in one embodiment of the present invention, the absolute values of the differences between the gray values of the growing pixel points in the previous region growing process and the gray values of the pixel points in the preset adjacent regions not belonging to the growing region are obtained, all the absolute values of the differences are averaged, and finally normalization is performed to obtain the correction parameters, which are expressed as follows:
wherein C is i,j A correction parameter indicating the growth of the jth zone of the ith initial growth point; norm () represents a normalization function; avg () represents an averaging function; m is M i,j-1 Representing the number of corresponding growth pixel points when the j-1 th region of the ith initial growth point grows; k (K) i,j-1,m When the j-1 th region of the ith initial growing point grows, the number of the pixel points which do not belong to the growing region in the preset neighborhood of the mth growing pixel point is represented; h i,j-1,m The gray value of the m-th growth pixel point when the j-1 th region of the i-th initial growth point grows; h ′ i,j-1,m,k And when the j-1 th region of the ith initial growth point grows, the gray value of the pixel point of the ith growth pixel point which does not belong to the growth region is displayed.
In the calculation formula of the correction parameter, when the last region grows, the smaller the absolute value of the difference value of the gray value of the growing pixel point and the pixel point which does not belong to the growing region in the preset adjacent region is, the smaller the average value is, which means that the smaller the integral difference between the growing pixel point grown in the last region and the pixel point which does not belong to the growing region in the adjacent region is, the smaller the growth threshold value is required to be set, so that the smaller the gray difference in the final growing region is, the more obvious the density characteristic is, and the smaller the correction parameter is.
It should be noted that, in one embodiment of the present invention, the preset neighborhood is eight neighborhoods of each pixel point; in other embodiments of the present invention, the practitioner may choose to set other preset neighbors; when the primary region growth is carried out according to the growth threshold value of the current region growth, when the gray scale difference between the growing pixel point in the previous region growth process and the pixel points which do not belong to the growth region in the eight adjacent regions is smaller than the growth threshold value of the current region growth, the pixel points which do not belong to the growth region are considered to be in accordance with the growth condition, and are brought into the growth region to serve as the growing pixel points of the current region growth.
After the correction parameters of each region growth are obtained, the growth threshold value in the last region growth can be corrected by using the correction parameters, the growth threshold value of the current region growth is obtained, and the current region growth is carried out.
Preferably, in one embodiment of the present invention, the product of the correction parameter and the growth threshold at the time of the last region growth is taken as the threshold of the current growth region growth.
Preferably, in one embodiment of the present invention, considering that all initial growing points start growing at the same time, in order to avoid overlapping of different growing areas, it is a condition of reaching the region growth termination when there is no pixel point in the growing area of the current growing area that does not belong to the growing area, i.e. there is no pixel point that does not exist to be grown; meanwhile, when no pixel points meeting the growth conditions exist in the neighborhood of the edge pixel points of the growth area, the pixel points are also a representation of the growth termination condition of the area, so that when any one of the two conditions occurs, the growth of the area is stopped.
It should be noted that two conditions may occur simultaneously in the region growth termination condition, and it is considered that the region growth termination condition is satisfied and the region growth is stopped. When the next region growth of the growth region with different initial growth points can incorporate the same pixel points into the growth region, the pixel points can still be assigned to one region when the growth threshold is smaller, which means that the pixel points are more similar to the region with smaller growth threshold and are more matched with the region with smaller growth threshold, so that the pixel points are defined to belong to the growth region with smaller growth threshold.
In the embodiment of the invention, each growth is selected from the preset adjacent areas of the pixel points at the edge of the current growth area, the state of the growth is presented in a circle-by-circle manner, and when the condition of stopping the growth of the area is reached, the growth of the area is stopped, and the final growth area of each initial growth point is obtained.
The primary region growth is also simply referred to as primary growth.
Referring to fig. 2, which is a schematic diagram illustrating a region growing process according to an embodiment of the present invention, fig. 2 is a part of a cross-sectional image, and shows a first 3 times of growing of a certain initial growing point, wherein each square represents a pixel point, a number in the square represents a number of growing times, a number 0 represents a wire pixel point, 1 represents a first growing, 2 represents a second growing, and 3 represents a third growing; because the preset neighborhood is selected to be eight neighbors, the growing pixel points in the last region growing process become edge pixel points of the growing region, namely, the growing range of the region growing is eight neighbors of the edge pixel points of the current region; meanwhile, considering that the growth threshold is continuously reduced, in order to avoid that one pixel point does not accord with the growth threshold in the n-th growth process, the situation that the pixel point accords with the growth threshold in the n+1th and subsequent growth processes, for example, the pixel point with an X mark in the three parts in the figure 2, may not accord with the growth threshold in the first growth and the second growth, but the difference of the gray value between the pixel point and the pixel point with the mark number 3 is smaller, and the pixel point accords with the growth threshold in the third growth; in order to avoid the occurrence of the situation, the pixels which do not accord with the growth threshold value in the growth range are marked in the growth process, as shown by a mark X in fig. 2, so that the marked pixels lose the possibility of becoming growth pixels, the fluctuation of density uniformity is more highlighted, if the pixels need to be presented to related personnel in the growth process, gaps exist in the growth area, the related personnel can know the positions of the defects of the products more conveniently, and the defect reasons are analyzed.
In fig. 2, 4 wire pixels are adjacent to each other, and then the 4 wire pixels are integrated into an initial growth point, 12 pixels are arranged around the initial growth point, and after the initial growth point is outwards grown for one circle of pixels, the first growth is completed; 12 growing pixel points grow in the first region, searching pixel points which do not belong to the growing region and have gray value difference smaller than the second growing threshold value from the edge pixel points in eight adjacent regions of the edge pixel points of the current region by calculating the growing threshold value and the second corresponding growing threshold value, and incorporating the pixel points into the current growing region to complete the second region growth; 18 growing pixel points grow in the second region, calculating a growing threshold corresponding to the third growth, searching pixel points which do not belong to the growing region and have gray value difference smaller than the third growing threshold from the edge pixel points in the eight adjacent regions of the edge pixel points of the current region, taking the pixel points into the current growing region, completing the growth of the third region, repeating the growing process until the region growth termination condition is reached, and obtaining a final growing region of each initial growing point.
In other embodiments of the present invention, when a pixel does not meet the growth threshold in the n-th growth process and meets the growth threshold in the n+1th and subsequent growth processes, the implementer may also incorporate the pixel into the growth area, mark the corresponding number of growth times, and also visually present the growth process to the relevant personnel for analysis when needed.
Step S5: analyzing the change of the growth threshold value and the change of the number of the growth pixels in the growth process of the final growth area of each initial growth point to obtain the density distribution uniformity corresponding to each initial growth point; and detecting the quality of the conductive foam according to the density distribution uniformity corresponding to each initial growth point.
Along with the continuous growth of the growth area of each initial growth point, the number of the growth pixel points continuously fluctuates, and the growth threshold value also continuously fluctuates; in the growth process, the growth threshold is adjusted from large to small, the faster the change speed of the growth threshold is, the larger the adjustment amplitude of the growth threshold is, and the larger the number of the growth pixels is, the smaller the difference between the growth pixels and the growth area in each growth is reflected, and the stronger the density distribution uniformity is, so that the change of the growth threshold and the change of the number of the growth pixels in the growth process can be analyzed, and the density distribution uniformity corresponding to each initial growth point is obtained.
Preferably, in one embodiment of the present invention, considering that the growing area is continuously enlarged, so that the edge pixels are continuously increased, and the larger the selecting range of the growing pixels is, the larger the ratio of the number of the growing pixels to the number of the growing pixels in the previous growing process is, which means that the more the growing pixels are selected in the current growing process, the smaller the difference between the growing pixels and the growing area is, and the stronger the density distribution uniformity is; meanwhile, the larger the change speed of the growth threshold value in the adjacent growth process is, the smaller the correction parameter is, the smaller the integral difference between the grown pixel point grown in the last region and the pixel point in the adjacent region which does not belong to the growth region is reflected, and the density uniformity is also stronger;
based on the first parameter, the change rate of the growth threshold value in the whole growth process of the initial growth point is taken as a first parameter;
summing the ratio of the number of the growing pixel points in the adjacent two growing processes in the whole growing process of the initial growing point to obtain a second parameter;
obtaining the product of the first parameter and the second parameter of each initial growth point, taking the ratio of the product to the preset density empirical parameter as the density distribution uniformity parameter of each initial growth point, and expressing the density distribution uniformity parameter as follows by a formula:
wherein W is i A density distribution uniformity parameter indicating an i-th initial growth point; D1D 1 i A first parameter representing an i-th initial growth point,Q i,j-1 representing a corresponding growth threshold value when the j-1 th zone of the i initial growth point grows; q (Q) i,j Representing a growth threshold corresponding to the jth region of the ith initial growth point during growth; j (J) i Indicating the total number of growing times of the ith initial growing point; D2D 2 i Second parameter indicating the ith initial growth point,/->M i,j Representing the number of corresponding growth pixel points when the jth region of the ith initial growth point grows; m is M i,j-1 Representing the number of corresponding growth pixel points when the j-1 th region of the ith initial growth point grows; wb represents a preset densityEmpirical parameters.
When the first parameter is calculated, since the correction parameter is normalized, Q is as follows i,j-1 Q is greater than or equal to i,j ，Q i,j-1 -Q i,j The result of (2) is a non-negative number; in other embodiments of the present invention, other basic mathematical operations or function mapping may be used to implement the relevant mapping, which are all technical means well known to those skilled in the art, and are not described herein.
It should be noted that, in one embodiment of the present invention, the preset density experience parameter is analyzed by producing a cross-sectional image of a plurality of quality products in the current type of conductive foam, products of the first parameter and the second parameter of each quality product conductive foam are obtained respectively, and an average value of the products is taken as the density experience parameter, and the density experience parameter is related to a specific production process and a specific product type in the production process, so that the present invention is not limited herein; in another embodiment of the present invention, the empirical value may be manually set or the empirical density parameter may be obtained by analyzing the qualified product, but the quality qualification threshold corresponding to the empirical density parameter obtained in different manners is different, for example, the empirical density parameter obtained according to the good quality product is larger compared with the qualified product, the empirical density parameter obtained according to the qualified product is smaller, the uniform density distribution parameter obtained by the same conductive foam is different, and the corresponding quality qualification threshold needs to be adjusted.
Preferably, in one embodiment of the present invention, considering that there is a problem in density distribution in a final growth area of any initial growth point in a cross-sectional image of the conductive foam, performance of the conductive foam may be affected, a quality qualification threshold is set, and when a density distribution uniformity parameter of all initial growth points of the conductive foam is greater than a preset quality qualification threshold, quality of the current conductive foam is considered to be qualified, otherwise, quality of the current conductive foam is considered to be problematic.
It should be noted that, the quality qualification threshold is related to the product production process and the product performance requirement, and the practitioner needs to set according to the actual production scenario, and in one embodiment of the present invention, the preset quality qualification threshold is 0.75.
In another embodiment of the invention, the product classification can be performed rapidly in consideration of the quality detection of the visual detection of the cross section of the conductive foam, and the product classification is facilitated to be classified into different classes in consideration of the analysis of the density experience parameters of the high-quality product, so that the high-quality product is selected through experimental test, and the density experience parameters are obtained; the larger the density distribution uniform parameter corresponding to the conductive foam product is, the better the quality of the conductive foam product is, so that a classification interval is set to rapidly classify the product, for example, the product is a non-qualified product when all the density distribution uniform parameters corresponding to the conductive foam exist in the interval (0,0.75), the product is a qualified product when all the density distribution uniform parameters corresponding to the conductive foam are greater than 0.75, and part or all of the density distribution uniform parameters exist in the interval (0.75,0.85); when all density distribution uniformity parameters corresponding to the conductive foam are larger than 0.85 and part or all density distribution uniformity parameters belong to a [0.85,0.95 ] interval, the conductive foam is good; when all the density distribution uniform parameters corresponding to the conductive foam are larger than 0.95, the conductive foam is a high-quality product.
In another embodiment of the invention, the corresponding part of the unqualified conductive foam can be visually presented, so that the related personnel can conveniently analyze and search the production problem, for example, the growth process of the growth area with the uniform density distribution parameter lower than the preset quality qualification threshold is displayed in a detection screen, and the related personnel can be helped to analyze.
In summary, in order to solve the technical problems that the density uniformity analysis of the conductive foam is not accurate enough and the quality detection accuracy is affected in the existing region growth algorithm, the invention provides a conductive foam quality detection method. Firstly, acquiring a cross-sectional image of conductive foam, further acquiring an initial growth point and performing first growth; further acquiring a growth threshold value of each region growth, performing region growth, stopping region growth when a region growth termination condition is reached, and obtaining a final growth region of each initial growth point; further analyzing the change of the growth threshold value and the change of the number of the growth pixel points in the growth process of each final growth area, obtaining the density distribution uniformity corresponding to each initial growth point, and detecting the quality of the conductive foam. According to the invention, the distribution characteristics of the sponge density around each wire are analyzed in a self-adaptive area growth mode, the uniformity of the sponge density distribution is detected in a visual detection mode, the quality detection of the conductive foam is rapidly realized, and the efficiency and the accuracy of the quality detection of the conductive foam are improved.
An embodiment of a self-adaptive region growing method suitable for a cross section of conductive foam is provided:
the method for acquiring the related information of the conductive foam through the visual detection is a real-time, automatic, efficient and high-precision mode, so that human errors can be reduced, and the manufacturing cost can be reduced; and the region growth can provide information about the uniformity of the conductive foam through the generated region; however, the parameter setting of the region growth influences the growth result, and the subjectivity of the artificial setting of the growth threshold is too strong, so that the reliability and accuracy of the region growth result are low.
In order to solve the problems that the parameter setting of the existing region growing algorithm is unreasonable and the accuracy of the region growing result is affected, the self-adaptive region growing method suitable for the cross section of the conductive foam is provided, and the method specifically comprises the following steps:
step S1: and acquiring a cross-sectional image of the conductive foam.
Step S2: and acquiring initial growth points based on the wire pixel points in the sectional image, and carrying out first region growth according to a preset mode.
Step S3: the pixel points newly added in each region growing process are called growing pixel points; and acquiring a growth threshold according to the difference characteristic of the gray value of the growth pixel point grown in the initial growth point and the first region.
Step S4: for any subsequent region growing process, performing region growing once according to the growing threshold of the region growing once, wherein the growing range of the region growing once is a preset neighborhood of growing pixel points in the previous region growing process, and the acquiring process of the growing threshold of the region growing once comprises the following steps: acquiring correction parameters corresponding to the current region growth according to the gray difference between the growth pixel points in the previous region growth process and the pixel points in the preset adjacent region which do not belong to the growth region in the previous region growth process; correcting the growth threshold value in the last region growth by using the correction parameters corresponding to the region growth to obtain the growth threshold value of the region growth; and stopping the region growth when the region growth termination condition is reached, and obtaining a final growth region of each initial growth point.
Because the specific implementation process of the self-adaptive region growing method suitable for the cross section of the conductive foam is described in detail in the above-mentioned conductive foam quality detection method, the detailed description is omitted.
In summary, in order to solve the problems that the parameter setting of the existing region growing algorithm is unreasonable and the accuracy of the region growing result is affected, a self-adaptive region growing method suitable for the cross section of the conductive foam is provided. Firstly, acquiring a cross-sectional image of conductive foam, further acquiring an initial growth point and performing first growth; further acquiring a growth threshold value of each region growth, performing region growth, stopping region growth when a region growth termination condition is reached, and obtaining a final growth region of each initial growth point; the invention avoids the condition of unsuitable area growth result caused by human factors, realizes self-adaptive area growth, and provides accurate basis for subsequent analysis of the cross-section image characteristics of the conductive foam.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (10)
1. The method for detecting the quality of the conductive foam is characterized by comprising the following steps:
acquiring a cross-sectional image of the conductive foam;
acquiring initial growth points based on wire pixel points in the cross-sectional image, and carrying out first region growth according to a preset mode;
the pixel points newly added in each region growing process are called growing pixel points; acquiring a growth threshold according to the difference characteristic of the gray value of the growth pixel point grown in the initial growth point and the first region;
and for any subsequent region growing process, carrying out primary region growing according to the growing threshold of the current region growing, wherein the growing range of the primary region growing is a preset neighborhood of the growing pixel point in the previous region growing process, and the acquiring process of the growing threshold of the current region growing comprises the following steps: acquiring correction parameters corresponding to the current region growth according to the gray level difference between the growth pixel points in the previous region growth process and the pixel points in the preset adjacent region which do not belong to the growth region in the previous region growth process; correcting the growth threshold value in the last region growth by using the correction parameters corresponding to the region growth to obtain the growth threshold value of the region growth; stopping the region growth when the region growth termination condition is reached, and obtaining a final growth region of each initial growth point;
analyzing the change of the growth threshold value and the change of the number of the growth pixels in the growth process of the final growth area of each initial growth point to obtain the density distribution uniformity corresponding to each initial growth point; and detecting the quality of the conductive foam according to the density distribution uniformity corresponding to each initial growth point.
2. The method for detecting the quality of the conductive foam according to claim 1, wherein the method for acquiring the correction parameters comprises the following steps:
and obtaining the absolute value of the difference value of the gray value of the growing pixel point in the last region growing process and the pixel point which does not belong to the growing region in the preset adjacent region, averaging all the absolute values of the difference value, and finally normalizing to obtain the correction parameter.
3. The method for detecting the quality of the conductive foam according to claim 1, wherein the method for acquiring the growth threshold comprises the following steps:
for the first growth: obtaining a difference value between the average gray value of each initial growth point and the minimum gray value of the growth pixel points grown in the first region, and taking the difference value as a growth threshold value of each initial growth point;
for any one of the following region growing processes: and taking the product of the correction parameter corresponding to the current region growth and the growth threshold value in the last region growth as the growth threshold value of the current region growth.
4. The method for detecting the quality of the conductive foam according to claim 1, wherein the method for obtaining the uniformity of the density distribution comprises the following steps:
taking the change rate of the growth threshold value in the whole growth process of the initial growth point as a first parameter;
summing the ratio of the number of the growing pixels in the adjacent two growing processes in the whole growing process of the initial growing point to obtain a second parameter;
and obtaining the product of the first parameter and the second parameter of each initial growth point, and taking the ratio of the product to a preset density experience parameter as a density distribution uniformity parameter of each initial growth point.
5. The method for detecting the quality of the conductive foam according to claim 4, wherein the method for acquiring the first parameter comprises:
acquiring a first parameter by using a first parameter calculation formula; the first parameter calculation formula includes:
wherein D1 i A first parameter representing the ith initial growth point, Q i,j-1 Representing a corresponding growth threshold value when the j-1 th zone of the i initial growth point grows; q (Q) i,j Representing a growth threshold corresponding to the jth region of the ith initial growth point during growth; j (J) i Indicating the total number of growths at the i-th initial growth point.
6. The method for detecting the quality of the conductive foam according to claim 1, wherein the method for detecting the quality of the conductive foam comprises the following steps:
when the density distribution uniformity parameters of all the initial growth points of the conductive foam are larger than a preset quality qualification threshold, the quality of the current conductive foam is qualified, otherwise, the quality of the current conductive foam is considered to be problematic.
7. The method for detecting the quality of the conductive foam according to claim 1, wherein the reached region growth termination condition is:
and when all the pixels which do not belong to the growth area in the growth range of the current growth area do not accord with the growth threshold value, or when the pixels which do not belong to the growth area do not exist in the growth range of the current growth area, the area growth termination condition is reached.
8. The method for detecting the quality of the conductive foam according to claim 1, wherein the method for acquiring the pixel points of the conductive wire comprises the following steps:
and processing the section image by using an Ojin threshold algorithm to obtain a binarized section image, and taking a pixel point with the value of 1 in the binarized section image as a wire pixel point.
9. The method for detecting the quality of the conductive foam according to claim 1, wherein the method for performing the first region growth according to the preset mode comprises the following steps:
and (5) outwards growing the initial growth point by a circle of pixel points to finish the first region growth.
10. The method for detecting the quality of conductive foam according to claim 6, wherein the preset quality qualification threshold is 0.75.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410056049.2A CN117876341B (en) | 2024-01-15 | 2024-01-15 | Conductive foam quality detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410056049.2A CN117876341B (en) | 2024-01-15 | 2024-01-15 | Conductive foam quality detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117876341A true CN117876341A (en) | 2024-04-12 |
| CN117876341B CN117876341B (en) | 2024-07-09 |
Family
ID=90591635
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410056049.2A Active CN117876341B (en) | 2024-01-15 | 2024-01-15 | Conductive foam quality detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117876341B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105115547A (en) * | 2015-10-10 | 2015-12-02 | 深圳鸿泽自动化科技有限公司 | Conductive foam testing method |
| CN114842274A (en) * | 2022-06-27 | 2022-08-02 | 深圳市鑫诺诚科技有限公司 | Conductive foam elasticity analysis method, device and equipment based on image analysis |
| CN116152214A (en) * | 2023-03-02 | 2023-05-23 | 山西春福科技有限公司 | Plastic powder weather resistance detection method for metal product coating |
| CN116258719A (en) * | 2023-05-15 | 2023-06-13 | 北京科技大学 | Flotation foam image segmentation method and device based on multi-mode data fusion |
| WO2023196805A2 (en) * | 2022-04-06 | 2023-10-12 | Daniels John James | Wearable electronic for digital healthcare |
-
2024
- 2024-01-15 CN CN202410056049.2A patent/CN117876341B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105115547A (en) * | 2015-10-10 | 2015-12-02 | 深圳鸿泽自动化科技有限公司 | Conductive foam testing method |
| WO2023196805A2 (en) * | 2022-04-06 | 2023-10-12 | Daniels John James | Wearable electronic for digital healthcare |
| CN114842274A (en) * | 2022-06-27 | 2022-08-02 | 深圳市鑫诺诚科技有限公司 | Conductive foam elasticity analysis method, device and equipment based on image analysis |
| CN116152214A (en) * | 2023-03-02 | 2023-05-23 | 山西春福科技有限公司 | Plastic powder weather resistance detection method for metal product coating |
| CN116258719A (en) * | 2023-05-15 | 2023-06-13 | 北京科技大学 | Flotation foam image segmentation method and device based on multi-mode data fusion |
Non-Patent Citations (1)
| Title |
|---|
| 黄士超: "基于海绵复合材料的弹性电极制备及电容性能研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, 15 August 2019 (2019-08-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117876341B (en) | 2024-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101901342B (en) | Method and device for extracting image target region | |
| CN108038838A (en) | A kind of cotton fibriia species automatic testing method and system | |
| CN116071363B (en) | Automatic change shaped steel intelligent production monitoring system | |
| CN115797352B (en) | Tongue picture image processing system for traditional Chinese medicine health-care physique detection | |
| CN111739012A (en) | Camera module white spot detecting system based on turntable | |
| WO2022267270A1 (en) | Crack characteristic representation method and system based on multi-fractal spectrum | |
| CN115294116A (en) | Method, device and system for evaluating dyeing quality of textile material based on artificial intelligence | |
| CN116596905A (en) | Method for detecting surface defects of integrated circuit chip | |
| CN116757972B (en) | Fabric defect detection method capable of resisting influence of shadow noise | |
| CN118225803B (en) | Visual detection method for appearance of blade surface of bulldozer | |
| CN113538424A (en) | Wood board joint classification data identification method based on artificial intelligence | |
| CN117808799B (en) | Chamfering equipment processing quality detection method based on artificial intelligence | |
| CN117372435B (en) | Connector pin detection method based on image characteristics | |
| CN101897592B (en) | Method and device for extracting metallic foreign body from X-ray image | |
| CN117237245B (en) | Industrial material quality monitoring method based on artificial intelligence and Internet of things | |
| CN117078676B (en) | Breaking hammer shell visual detection method based on image processing | |
| CN117876341B (en) | Conductive foam quality detection method | |
| CN114998346B (en) | Waterproof cloth quality data processing and identifying method | |
| CN116563276A (en) | Chemical fiber filament online defect detection method and detection system | |
| CN110838101B (en) | Image segmentation quality evaluation method based on edge characteristics | |
| CN111257422A (en) | Wheel axle defect identification model construction method and defect identification method based on machine vision | |
| CN117949616B (en) | Automatic visual detection method for quality of preserved egg finished product | |
| CN116958134B (en) | Plastic film extrusion quality evaluation method based on image processing | |
| CN115578388B (en) | Data processing method for textile production | |
| CN116721105B (en) | Explosive bead production abnormality online detection method based on artificial intelligence |
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 |