CN115424062B - Method, device, equipment and storage medium for automatically identifying diagonal network - Google Patents
Method, device, equipment and storage medium for automatically identifying diagonal network Download PDFInfo
- Publication number
- CN115424062B CN115424062B CN202211042896.0A CN202211042896A CN115424062B CN 115424062 B CN115424062 B CN 115424062B CN 202211042896 A CN202211042896 A CN 202211042896A CN 115424062 B CN115424062 B CN 115424062B
- Authority
- CN
- China
- Prior art keywords
- image
- detected
- inclined wire
- gray
- black
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 79
- 238000003860 storage Methods 0.000 title claims abstract description 18
- 230000008569 process Effects 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000005520 cutting process Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 13
- 230000003993 interaction Effects 0.000 claims abstract description 7
- 230000000007 visual effect Effects 0.000 claims description 28
- 230000015654 memory Effects 0.000 claims description 20
- 238000012805 post-processing Methods 0.000 claims description 17
- 238000004220 aggregation Methods 0.000 claims description 12
- 230000002776 aggregation Effects 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 10
- 230000009467 reduction Effects 0.000 claims description 10
- 238000003708 edge detection Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 abstract description 11
- 229910052710 silicon Inorganic materials 0.000 abstract description 11
- 239000010703 silicon Substances 0.000 abstract description 11
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- 238000012795 verification Methods 0.000 abstract description 4
- 238000004590 computer program Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 229910003460 diamond Inorganic materials 0.000 description 2
- 239000010432 diamond Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- 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
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06T2207/20081—Training; Learning
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method, a device, equipment and a storage medium for automatically identifying a diagonal network, wherein the method comprises the steps of collecting image information of the diagonal network to be detected; image processing is carried out on the acquired image information, and the number of grooves of each main roller of the inclined wire net is identified; carrying out difference value calculation on the obtained groove number data of each main roller to obtain the actual groove number of the inclined wire network to be detected; comparing the actual number of grooves of the to-be-detected inclined wire net with a preset process standard value; and judging whether the cutting process requirements are met or not according to the comparison result, and outputting the result to the human-computer interaction interface. The invention is used for solving various problems of wire mesh damage, unclear photographing, large error, low efficiency, labor consumption, high product reject ratio and the like existing in the conventional manual verification, and can effectively improve the production yield and the working efficiency of the silicon wafer.
Description
Technical Field
The invention relates to the technical field of photovoltaic equipment, in particular to a method and device for automatically identifying a diagonal network, equipment and a storage medium.
Background
The diamond wire used in the silicon wafer slicing process in the front photovoltaic industry is routed on a main roller of the slicing machine, and the routing positions of a left roller, a right roller and a lower roller of the diamond wire directly influence the slicing effect and mainly influence bad wire marks, thickness and warping of the silicon wafer. The existing control means is that staff photographs the left roller, the right roller and the lower roller oblique line net and compares the net empty groove numbers on the photographs to check whether the photographs meet the cutting process requirements, namely, a camera is manually used for collecting pictures, the oblique line net groove numbers are manually checked, the net groove numbers are recorded, and the main roller groove numbers are compared.
However, the phenomena of damage to the wire net or unclear photographing and the like are easily caused by the fact that staff with limited operation space slightly carelessly operate the machine in the verification process due to complex machine environment. Because the slot channel of the main roller is very tiny (about 0.193-0.221 mm), personnel are very easy to generate wrong numbers and leakage numbers in the slot counting process, at least each oblique line net needs about 15-100 slots, and a large amount of manpower is needed to be input; meanwhile, in the process of manually verifying the groove number of the diagonal net, the diagonal net error is easily caused by human factors, so that the phenomena of line mark, thickness, warping and the like of the silicon wafer are caused.
Accordingly, the prior art has the following problems:
1. the phenomena of damage to the wire net or unclear photographing and the like are easily caused by slight carelessness of staff with limited operation space;
2. the slot channel is very tiny (about 0.193-0.221 mm), and personnel are very easy to generate wrong numbers and leakage numbers in the slot counting process;
3. at least about 15-100 wire slots are needed for each diagonal net, and about 45-300 main roller wire slots which need to be verified by one slicer are needed; a great deal of manpower is required to be input;
4. in the process of manually verifying the groove number of the diagonal net, the diagonal net error is easily caused by human factors, so that the phenomena of line mark, thickness, warping and the like of the silicon wafer are caused.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for automatically identifying a diagonal line network, which are used for solving various problems of line network damage, unclear photographing, large error, low efficiency, labor consumption, high product reject ratio and the like existing in the conventional manual verification.
In a first aspect, the present invention provides a method for automatically identifying a diagonal net, the method comprising:
collecting image information of a to-be-tested inclined wire network;
image processing is carried out on the acquired image information, and the number of grooves of each main roller of the inclined wire net is identified;
carrying out difference value calculation on the obtained groove number data of each main roller to obtain the actual groove number of the inclined wire network to be detected;
comparing the actual number of grooves of the to-be-detected inclined wire net with a preset process standard value;
and judging whether the cutting process requirements are met or not according to the comparison result, and outputting the result to the human-computer interaction interface.
According to the method for automatically identifying the diagonal network, before the image information of the diagonal network is acquired, the following steps are further executed: and placing the industrial camera above the inclined wire net to be detected, and automatically adjusting the focal length according to the distance by the industrial camera to acquire.
According to the method for automatically identifying the diagonal network, before the image information of the diagonal network is acquired, the following steps are further executed: acquiring shooting length and shooting width of a to-be-detected inclined wire net, and acquiring visual field length and visual field width of an industrial camera in the to-be-detected inclined wire net; when shooting, the industrial camera is separated from the plane of the inclined wire network to be detected by a preset distance, the plane of the lens is parallel to the plane of the wire network, and the industrial camera is controlled to shoot images of a plurality of inclined wire networks to be detected according to the acquired shooting length, shooting width, vision length and vision width, so that complete image information of the inclined wire network to be detected is obtained;
wherein, the visual field length includes: starting from a first groove at the head of a main roller of the inclined wire net to be tested to a groove where a first wire of the inclined wire net to be tested is positioned; the field of view width includes: the tangential point position of the inclined wire net to be measured and the main roller is positioned at the position of the width range of the visual field.
According to the method for automatically identifying the diagonal network provided by the invention, the acquired image information is subjected to image processing, and the method comprises the following steps: performing texture extraction on the acquired image information by using an edge detection operator, preprocessing the extracted image, namely removing a noise image, converting the noise image into a gray image, and performing binarization extraction on the gray image to respectively obtain black and white stripe patterns of each main roller;
the edge detection operator adopts one of a Sobel operator and a Canny operator, or uses a convolution kernel which is the same as the local structure of the image repetition to carry out convolution operation with the image.
According to the method for automatically identifying the inclined wire net, provided by the invention, the number of grooves of each main roller of the inclined wire net is identified, and the method comprises the following steps: longitudinally dividing the obtained black-and-white stripe graph into a plurality of columns, respectively detecting the gray average value of each of the plurality of columns, and determining that the column with the gray average value larger than the gray average value of all the adjacent columns is the white stripe column of the black-and-white stripe graph; the column whose gradation average value is smaller than the gradation average value of all the columns adjacent thereto is the black stripe column of the black-and-white stripe pattern, thereby calculating the number of stripes of each black-and-white stripe pattern to determine the number of grooves of each main roller.
According to the method for automatically identifying the wire net of the invention, the number of grooves of each main roller of the wire net to be detected is identified, and the method comprises the following steps: and carrying out groove number identification on grooves of the left roller, the right roller and the lower roller of the wire mesh to be detected according to the image acquisition sequence, sequentially recording the groove number parameters, and calculating the actual numerical value of the wire mesh to be detected through the groove number difference value.
According to the method for automatically identifying the inclined wire net provided by the invention, the obtained groove number data of each main roller is subjected to difference value calculation, and the method comprises the following steps:
left roll groove number-right roll groove number = oblique line net groove number;
and then, comparing the obtained diagonal line net groove number with a preset process standard value, and judging whether the groove number of the lower roller is at the middle value of the groove numbers of the left roller and the right roller.
According to the method for automatically identifying the diagonal network, which is provided by the invention, when the image processing is carried out, the method further comprises the following steps: the method comprises the steps of forming an original image set x by using collected original images of all to-be-detected inclined wire nets, and marking the original images as x= { x1, x2, x3, …, xn };
post-processing the image information, collecting processed images, and forming a post-processed image set X by all the processed images, wherein the post-processed image set X is marked as X= { X1, X2, X3, …, xn }, and Xt is an image of Xt after post-processing;
establishing a context aggregation network model based on full convolution, wherein in the context aggregation network model, an image sequence formed by an original image which is not subjected to post-processing and a corresponding image which is subjected to post-processing is used as input, and parameters in the context aggregation network model are updated by adopting a supervision training mode to obtain a trained network model;
and inputting the image to be processed into the trained network model to obtain the image with enhanced visual effect.
In a second aspect, the present invention also provides an apparatus for automatically identifying a diagonal mesh, comprising:
the image acquisition unit is used for acquiring image information of the inclined wire network to be detected;
the image processing unit is used for carrying out image processing on the acquired image information and carrying out slot number identification on each main roller of the diagonal net;
the calculating unit is used for carrying out difference value calculation on the obtained groove number data of each main roller to obtain the actual groove number of the inclined wire network to be detected;
the judging unit is used for comparing the actual number of grooves of the to-be-detected inclined wire net with a preset process standard value;
and the result output unit is used for judging whether the cutting process requirements are met according to the comparison result and outputting the result to the human-computer interaction interface.
Therefore, the ditch number of the inclined wire net can be automatically identified by photographing through an industrial camera; the automatic focal length adjustment of the industrial camera solves the problems that when the line net is tilted, the line net is easily damaged or the photographing is unclear due to the limited operation space and the careless operation; after photographing is finished, the number of grooves of the inclined line net is automatically identified through a software algorithm, so that the problem that personnel are easy to miss numbers and leak numbers in the process of counting the line grooves can be solved; the invention is applied to the problem caused by the diagonal network in the silicon wafer cutting process, can reduce the abnormity such as line mark, thickness, warping and the like caused by the diagonal network in the silicon wafer cutting process, and effectively improves the silicon wafer production yield.
In a third aspect, the present invention also provides an electronic device, including:
a memory storing computer executable instructions;
a processor configured to execute the computer-executable instructions,
wherein the computer executable instructions, when executed by the processor, perform the steps of any of the methods for automatically identifying a diagonal network described above.
In a fourth aspect, the present invention also provides a computer readable storage medium storing computer executable instructions which when executed by a processor implement the steps of any of the above methods for automatically identifying a diagonal network.
It can be seen that the present invention provides an electronic device and a storage medium for automatically identifying a diagonal network, comprising: one or more memories, one or more processors. The memory is used for storing the program codes, intermediate data generated in the running process of the program, the output result of the model and model parameters; the processor is used for processor resources occupied by code running and a plurality of processor resources occupied when training the model.
The invention is described in further detail below with reference to the drawings and the detailed description.
Drawings
FIG. 1 is a schematic diagram of the number of slots in a prior art wire mesh that manually requires verification of the diagonal wire mesh.
FIG. 2 is a flow chart of an embodiment of a method of automatically identifying a diagonal network in accordance with the present invention.
FIG. 3 is a schematic diagram of a method for automatically identifying a diagonal net according to an embodiment of the present invention with respect to collecting image information of the diagonal net to be tested.
FIG. 4 is a schematic diagram of image processing on system software according to an embodiment of a method of automatically identifying a diagonal net of the present invention.
FIG. 5 is a schematic diagram of the method for automatically identifying a diagonal screen according to an embodiment of the present invention for identifying the number of grooves in a black and white stripe pattern.
FIG. 6 is a schematic diagram of an embodiment of an apparatus for automatically identifying a diagonal web in accordance with the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 2-5, a method for automatically identifying a diagonal web, comprising the steps of:
step S1, collecting image information of a to-be-tested inclined wire network.
And S2, performing image processing on the acquired image information, and recognizing the number of grooves of each main roller of the diagonal web.
And S3, carrying out difference value calculation on the obtained groove number data of each main roller to obtain the actual groove number of the diagonal network to be tested.
And S4, comparing the actual groove number of the to-be-detected inclined wire net with a preset process standard value.
And S5, judging whether the cutting process requirements are met or not according to the comparison result, and outputting the result to the man-machine interaction interface.
In the above step S1, before the acquisition of the image information of the diagonal screen, further execution is performed of: and placing the industrial camera above the inclined wire net to be detected, and automatically adjusting the focal length according to the distance by the industrial camera to acquire.
In the above step S1, before the acquisition of the image information of the diagonal screen, further execution is performed of: acquiring shooting length and shooting width of a to-be-detected inclined wire net, and acquiring visual field length and visual field width of an industrial camera in the to-be-detected inclined wire net; when shooting is carried out, an industrial camera is spaced by a preset distance from the plane of the inclined wire net to be detected, and the plane of the lens is parallel to the plane of the wire net; the whole visual field range is 400 mm-200 mm, and according to the acquired shooting length, shooting width, visual field length and visual field width, the industrial camera is controlled to shoot images of a plurality of inclined wire nets to be detected, complete image information of the inclined wire nets to be detected is obtained, for example, the industrial camera is controlled to shoot images of a left roller, a right roller and a lower roller of the inclined wire nets to be detected in sequence.
The shooting length is the length of a rectangle which can surround the diagonal network to be detected, and the shooting length is the width of the rectangle which can surround the diagonal network to be detected.
Wherein, the visual field length includes: starting from a first groove at the head of a main roller of the inclined wire net to be tested to a groove where a first wire of the inclined wire net to be tested is positioned; the field of view width includes: the tangential point position of the inclined wire net to be measured and the main roller is positioned at the position of the width range of the visual field.
The industrial camera can automatically adjust the focal length according to the definition by manually adjusting the height of the lens within the range of 10cm-12cm through the self-contained function of the industrial camera (similar to the automatic focusing of a mobile phone camera), and the industrial camera can ensure that the picture reaches the identifiable standard (the resolution is 3904 x 2928 or more, ISO: 650-780).
Therefore, in the process of automatically identifying the diagonal network, staff places the industrial camera above the diagonal network to be checked, and the industrial camera automatically adjusts the focal length according to the distance to take a picture, so that the actual situation of the diagonal network can be clearly taken without being limited by space; after the oblique line net photographing is completed, the photographs are transmitted to system software to automatically identify the empty groove and the wire slot on the main roller.
In the step S2, image processing is performed on the acquired image information, including: and performing texture extraction on the acquired image information by using an edge detection operator, preprocessing the extracted image, namely removing a noise image, converting the noise image into a gray image, and performing binarization extraction on the gray image to respectively obtain black and white stripe patterns of each main roller.
The edge detection operator adopts one of a Sobel operator and a Canny operator, or uses a convolution kernel which is the same as the local structure of the image repetition to carry out convolution operation with the image.
Then, performing image morphology operation on the black-and-white stripe graph to obtain an operated black-and-white stripe graph, and then performing contour detection in the image to read all contours in the image; screening the outline according to the information such as the area, the shape and the like; if the contour is rectangular, the minimum bounding rectangle of the contour is taken as the final contour, otherwise, the screened contour is taken as the final contour.
In the step S2, the identifying of the number of grooves is performed on each main roll of the diagonal web, including: longitudinally dividing the obtained black-and-white stripe graph into a plurality of columns, respectively detecting the gray average value of each of the plurality of columns, and determining that the column with the gray average value larger than the gray average value of all the adjacent columns is the white stripe column of the black-and-white stripe graph; the column whose gradation average value is smaller than the gradation average value of all the columns adjacent thereto is the black stripe column of the black-and-white stripe pattern, thereby calculating the number of stripes of each black-and-white stripe pattern to determine the number of grooves of each main roller.
In this embodiment, the calculation of the gray average value specifically includes: first, a column gray value of each of a plurality of columns is detected, and a noise reduction gray value of each column is calculated according to the column gray value, wherein the noise reduction gray value is an average value of column gray values of adjacent columns before and after the column. The purpose of calculating the noise reduction gray value of each row according to the gray value of the row is to eliminate the problem that brightness of some parts is generally higher and brightness of other parts is generally lower caused by uneven brightness of an image of a diagonal line to be measured.
And calculating the average gray value of each column according to the noise reduction gray value, wherein the average gray value is the average value of the noise reduction gray values of each adjacent column before and after the column.
Then, determining a column with a gray average value larger than the gray average value of all columns adjacent to the column as a white stripe column of the black-and-white stripe map; the black stripe column of the black-and-white stripe pattern behaves with a gray average value smaller than the gray average value of all columns adjacent thereto.
Therefore, the influence of uneven brightness on black and white fringe pattern detection can be effectively eliminated by calculating the noise reduction gray value of each column; the middle rows of the black stripes and the white stripes can be effectively extracted by calculating the average gray value of each column, so that the distribution of the black-white stripe patterns is further determined, the calculated amount is simplified, and the black-white stripe recognition accuracy is high. The black-and-white stripe graph comprises a rectangular area where the to-be-detected diagonal line network is located, the gray value of the column is the sum or average value of the gray values of all the pixels in the column in the black-and-white stripe graph, and the sum or average value represents the whole gray condition of all the pixels in the column.
In the step S2, the identifying of the number of grooves of each main roll of the wire web to be tested includes: and carrying out groove number identification on grooves of the left roller, the right roller and the lower roller of the wire mesh to be detected according to the image acquisition sequence, sequentially recording the groove number parameters, and calculating the actual numerical value of the wire mesh to be detected through the groove number difference value.
In the step S3, the difference calculation is performed on the obtained groove number data of each main roller, including:
left roll groove number-right roll groove number = oblique line net groove number;
and then, comparing the obtained diagonal line net groove number with a preset process standard value, and judging whether the groove number of the lower roller is at the middle value of the groove numbers of the left roller and the right roller. If the left roll groove number is 20 grooves, the lower roll groove number is 18 grooves, and the right roll groove number is 15 grooves, the actual bevel line net groove number is 5 grooves. The preset process value is determined according to an actual process standard, if the preset process standard value is 5, the cutting process requirement is met when the number of left roller grooves is detected to be 20 grooves, the number of right roller grooves is detected to be 15 grooves, and the number of lower roller grooves is detected to be 18 grooves in real time.
Therefore, in the embodiment, after the recognition of the bevel line network grooves of the left roller, the right roller and the lower roller is completed, the system automatically compares the empty groove numbers among the three main rollers, automatically judges the bevel line network data and gives a conclusion. The invention solves the problems that the photographing is difficult due to space limitation in the photographing process of the oblique line net by using a camera manually, and the error number and the missing number are easy to occur in the manual counting of the grooves of the oblique line net; meanwhile, the invention realizes automatic identification and judgment of the number of the bevel wires, effectively improves the production efficiency and the product yield.
In the step S2, when performing image processing, the method further includes: the method comprises the steps of forming an original image set x by using collected original images of all to-be-detected inclined wire nets, and marking the original images as x= { x1, x2, x3, …, xn };
post-processing the image information, collecting processed images, and forming a post-processed image set X by all the processed images, wherein the post-processed image set X is marked as X= { X1, X2, X3, …, xn }, and Xt is an image of Xt after post-processing;
establishing a context aggregation network model based on full convolution, wherein in the context aggregation network model, an image sequence formed by an original image which is not subjected to post-processing and a corresponding image which is subjected to post-processing is used as input, and parameters in the context aggregation network model are updated by adopting a supervision training mode to obtain a trained network model;
and inputting the image to be processed into the trained network model to obtain the image with enhanced visual effect.
The post-processing in this embodiment refers to performing color, saturation, and contrast brightness adjustment operations on the image.
Device embodiment for automatically identifying diagonal network
In this embodiment, the present invention further provides an apparatus for automatically identifying a diagonal network, including:
the image acquisition unit 10 is used for acquiring image information of the to-be-detected inclined line network;
an image processing unit 20, configured to perform image processing on the acquired image information, and perform slot number recognition on each main roller of the diagonal web;
a calculating unit 30, configured to perform a difference calculation on the obtained slot number data of each main roller, so as to obtain an actual slot number of the diagonal network to be tested;
a judging unit 40, configured to compare the actual number of slots of the diagonal net to be tested with a preset process standard value;
and the result output unit 50 is used for judging whether the cutting process requirements are met according to the comparison result and outputting the result to the man-machine interaction interface.
Before the acquisition of the image information of the diagonal mesh, further execution: and placing the industrial camera above the inclined wire net to be detected, and automatically adjusting the focal length according to the distance by the industrial camera to acquire.
Before the acquisition of the image information of the diagonal mesh, further execution: acquiring shooting length and shooting width of a to-be-detected inclined wire net, and acquiring visual field length and visual field width of an industrial camera in the to-be-detected inclined wire net; when shooting is carried out, an industrial camera is spaced by a preset distance from the plane of the inclined wire net to be detected, and the plane of the lens is parallel to the plane of the wire net; the whole visual field range is 400 mm-200 mm, and according to the acquired shooting length, shooting width, visual field length and visual field width, the industrial camera is controlled to shoot images of a plurality of inclined wire nets to be detected, complete image information of the inclined wire nets to be detected is obtained, for example, the industrial camera is controlled to shoot images of a left roller, a right roller and a lower roller of the inclined wire nets to be detected in sequence.
In the image processing unit 20, image processing is performed on the acquired image information, including: and performing texture extraction on the acquired image information by using an edge detection operator, preprocessing the extracted image, namely removing a noise image, converting the noise image into a gray image, and performing binarization extraction on the gray image to respectively obtain black and white stripe patterns of each main roller.
In the image processing unit 20, the recognition of the number of grooves is performed for each main roller of the diagonal web, including: longitudinally dividing the obtained black-and-white stripe graph into a plurality of columns, respectively detecting the gray average value of each of the plurality of columns, and determining that the column with the gray average value larger than the gray average value of all the adjacent columns is the white stripe column of the black-and-white stripe graph; the column whose gradation average value is smaller than the gradation average value of all the columns adjacent thereto is the black stripe column of the black-and-white stripe pattern, thereby calculating the number of stripes of each black-and-white stripe pattern to determine the number of grooves of each main roller.
In the image processing unit 20, the identifying of the number of grooves of each main roll of the wire web to be tested includes: and carrying out groove number identification on grooves of the left roller, the right roller and the lower roller of the wire mesh to be detected according to the image acquisition sequence, sequentially recording the groove number parameters, and calculating the actual numerical value of the wire mesh to be detected through the groove number difference value.
In the calculating unit 30, the difference value calculation is performed on the acquired groove number data of each main roller, including:
left roll groove number-right roll groove number = oblique line net groove number;
and then, comparing the obtained diagonal line net groove number with a preset process standard value, and judging whether the groove number of the lower roller is at the middle value of the groove numbers of the left roller and the right roller. If the left roll groove number is 20 grooves, the lower roll groove number is 18 grooves, and the right roll groove number is 15 grooves, the actual bevel line net groove number is 5 grooves. The preset process value is determined according to an actual process standard, if the preset process standard value is 5, the cutting process requirement is met when the number of left roller grooves is detected to be 20 grooves, the number of right roller grooves is detected to be 15 grooves, and the number of lower roller grooves is detected to be 18 grooves in real time.
In performing image processing, the method further comprises: the method comprises the steps of forming an original image set x by using collected original images of all to-be-detected inclined wire nets, and marking the original images as x= { x1, x2, x3, …, xn };
post-processing the image information, collecting processed images, and forming a post-processed image set X by all the processed images, wherein the post-processed image set X is marked as X= { X1, X2, X3, …, xn }, and Xt is an image of Xt after post-processing;
establishing a context aggregation network model based on full convolution, wherein in the context aggregation network model, an image sequence formed by an original image which is not subjected to post-processing and a corresponding image which is subjected to post-processing is used as input, and parameters in the context aggregation network model are updated by adopting a supervision training mode to obtain a trained network model;
and inputting the image to be processed into the trained network model to obtain the image with enhanced visual effect.
Therefore, the ditch number of the inclined wire net can be automatically identified by photographing through an industrial camera; the automatic focal length adjustment of the industrial camera solves the problems that when the line net is tilted, the line net is easily damaged or the photographing is unclear due to the limited operation space and the careless operation; after photographing is finished, the number of grooves of the inclined line net is automatically identified through a software algorithm, so that the problem that personnel are easy to miss numbers and leak numbers in the process of counting the line grooves can be solved; the invention is applied to the problem caused by the diagonal network in the silicon wafer cutting process, can reduce the abnormity such as line mark, thickness, warping and the like caused by the diagonal network in the silicon wafer cutting process, and effectively improves the silicon wafer production yield.
In one embodiment, an electronic device is provided, which may be a server. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic device includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the electronic device is for storing data. The network interface of the electronic device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a method of automatically identifying a diagonal network.
It will be appreciated by those skilled in the art that the electronic device structure shown in this embodiment is merely a partial structure related to the present application and does not constitute a limitation of the electronic device to which the present application is applied, and that a specific electronic device may include more or fewer components than those shown in this embodiment, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method embodiments described above.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like.
Further, the logic instructions in the memory described above may be implemented in the form of software functional units and stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
It can be seen that the present invention provides an electronic device and a storage medium for automatically identifying a diagonal network, comprising: one or more memories, one or more processors. The memory is used for storing the program codes, intermediate data generated in the running process of the program, the output result of the model and model parameters; the processor is used for processor resources occupied by code running and a plurality of processor resources occupied when training the model.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above embodiments are only preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, but any insubstantial changes and substitutions made by those skilled in the art on the basis of the present invention are intended to be within the scope of the present invention as claimed.
Claims (8)
1. A method for automatically identifying a diagonal web, comprising:
collecting image information of a to-be-tested inclined wire network;
image processing is carried out on the acquired image information, and the method comprises the following steps: performing texture extraction on the acquired image information by using an edge detection operator, preprocessing the extracted image, namely removing a noise image, converting the noise image into a gray image, and performing binarization extraction on the gray image to respectively obtain black and white stripe patterns of each main roller; then, performing image morphology operation on the black-and-white stripe graph to obtain an operated black-and-white stripe graph, and then performing contour detection in the image to read all contours in the image; screening the outline according to the information of the area and the shape; if the contour is rectangular, taking the minimum bounding rectangle of the contour as the final contour, otherwise taking the screened contour as the final contour;
the method for identifying the number of the grooves of each main roller of the inclined wire net comprises the following steps: longitudinally dividing the obtained black-and-white stripe graph into a plurality of columns, respectively detecting the gray average value of each of the plurality of columns, and determining that the column with the gray average value larger than the gray average value of all the adjacent columns is the white stripe column of the black-and-white stripe graph; the black stripe columns of the black-and-white stripe patterns are the columns with the gray average value smaller than the gray average value of all the adjacent columns, so that the stripe number of each black-and-white stripe pattern is calculated to determine the groove number of each main roller; the gray average value calculation specifically includes: firstly, respectively detecting a column gray value of each column in a plurality of columns, calculating a noise reduction gray value of each column according to the column gray value, and calculating an average gray value of each column according to the noise reduction gray value;
carrying out difference value calculation on the obtained groove number data of each main roller to obtain the actual groove number of the inclined wire network to be detected;
comparing the actual number of grooves of the to-be-detected inclined wire net with a preset process standard value;
judging whether the cutting process requirements are met according to the comparison result, and outputting the result to a human-computer interaction interface;
wherein, before the acquisition of the image information of the diagonal mesh, further performing: acquiring shooting length and shooting width of a to-be-detected inclined wire net, and acquiring visual field length and visual field width of an industrial camera in the to-be-detected inclined wire net; when shooting, the industrial camera is separated from the plane of the inclined wire network to be detected by a preset distance, the plane of the lens is parallel to the plane of the wire network, and the industrial camera is controlled to shoot images of a plurality of inclined wire networks to be detected according to the acquired shooting length, shooting width, vision length and vision width, so that complete image information of the inclined wire network to be detected is obtained;
wherein, the visual field length includes: starting from a first groove at the head of a main roller of the inclined wire net to be tested to a groove where a first wire of the inclined wire net to be tested is positioned; the field of view width includes: the tangential point position of the inclined wire net to be measured and the main roller is positioned at the position of the width range of the visual field.
2. The method according to claim 1, characterized in that:
before the acquisition of the image information of the diagonal mesh, further execution: and placing the industrial camera above the inclined wire net to be detected, and automatically adjusting the focal length according to the distance by the industrial camera to acquire.
3. The method according to claim 1, characterized in that:
the method for identifying the number of the grooves of each main roller of the wire mesh to be detected comprises the following steps: and carrying out groove number identification on grooves of the left roller, the right roller and the lower roller of the wire mesh to be detected according to the image acquisition sequence, sequentially recording the groove number parameters, and calculating the actual numerical value of the wire mesh to be detected through the groove number difference value.
4. A method according to claim 3, characterized in that:
the calculating the difference value of the obtained groove number data of each main roller comprises the following steps:
left roll groove number-right roll groove number = oblique line net groove number;
and then, comparing the obtained diagonal line net groove number with a preset process standard value, and judging whether the groove number of the lower roller is at the middle value of the groove numbers of the left roller and the right roller.
5. The method according to any one of claims 1 to 4, wherein:
in performing image processing, the method further comprises: the method comprises the steps of forming an original image set x by using collected original images of all to-be-detected inclined wire nets, and marking the original images as x= { x1, x2, x3, …, xn };
post-processing the image information, collecting processed images, and forming a post-processed image set X by all the processed images, wherein the post-processed image set X is marked as X= { X1, X2, X3, …, xn }, and Xt is an image of Xt after post-processing;
establishing a context aggregation network model based on full convolution, wherein in the context aggregation network model, an image sequence formed by an original image which is not subjected to post-processing and a corresponding image which is subjected to post-processing is used as input, and parameters in the context aggregation network model are updated by adopting a supervision training mode to obtain a trained network model;
and inputting the image to be processed into the trained network model to obtain the image with enhanced visual effect.
6. An apparatus for automatically identifying a diagonal web, comprising:
the image acquisition unit is used for acquiring image information of the inclined wire network to be detected;
an image processing unit for performing image processing on the acquired image information, including: performing texture extraction on the acquired image information by using an edge detection operator, preprocessing the extracted image, namely removing a noise image, converting the noise image into a gray image, and performing binarization extraction on the gray image to respectively obtain black and white stripe patterns of each main roller; then, performing image morphology operation on the black-and-white stripe graph to obtain an operated black-and-white stripe graph, and then performing contour detection in the image to read all contours in the image; screening the outline according to the information of the area and the shape; if the contour is rectangular, taking the minimum bounding rectangle of the contour as the final contour, otherwise taking the screened contour as the final contour; the method for identifying the number of the grooves of each main roller of the inclined wire net comprises the following steps: longitudinally dividing the obtained black-and-white stripe graph into a plurality of columns, respectively detecting the gray average value of each of the plurality of columns, and determining that the column with the gray average value larger than the gray average value of all the adjacent columns is the white stripe column of the black-and-white stripe graph; the black stripe columns of the black-and-white stripe patterns are the columns with the gray average value smaller than the gray average value of all the adjacent columns, so that the stripe number of each black-and-white stripe pattern is calculated to determine the groove number of each main roller; the gray average value calculation specifically includes: firstly, respectively detecting a column gray value of each column in a plurality of columns, calculating a noise reduction gray value of each column according to the column gray value, and calculating an average gray value of each column according to the noise reduction gray value;
the calculating unit is used for carrying out difference value calculation on the obtained groove number data of each main roller to obtain the actual groove number of the inclined wire network to be detected;
the judging unit is used for comparing the actual number of grooves of the to-be-detected inclined wire net with a preset process standard value;
the result output unit is used for judging whether the cutting process requirements are met or not according to the comparison result and outputting the result to the human-computer interaction interface;
wherein, before the acquisition of the image information of the diagonal mesh, further performing: acquiring shooting length and shooting width of a to-be-detected inclined wire net, and acquiring visual field length and visual field width of an industrial camera in the to-be-detected inclined wire net; when shooting, the industrial camera is separated from the plane of the inclined wire network to be detected by a preset distance, the plane of the lens is parallel to the plane of the wire network, and the industrial camera is controlled to shoot images of a plurality of inclined wire networks to be detected according to the acquired shooting length, shooting width, vision length and vision width, so that complete image information of the inclined wire network to be detected is obtained;
wherein, the visual field length includes: starting from a first groove at the head of a main roller of the inclined wire net to be tested to a groove where a first wire of the inclined wire net to be tested is positioned; the field of view width includes: the tangential point position of the inclined wire net to be measured and the main roller is positioned at the position of the width range of the visual field.
7. An electronic device, comprising:
a memory storing computer executable instructions;
a processor configured to execute the computer-executable instructions,
wherein the computer executable instructions, when executed by the processor, implement the steps of the method of automatically identifying a diagonal network according to any one of claims 1-5.
8. A computer readable storage medium storing computer executable instructions which when executed by a processor perform the steps of the method of automatically identifying a diagonal network according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211042896.0A CN115424062B (en) | 2022-08-29 | 2022-08-29 | Method, device, equipment and storage medium for automatically identifying diagonal network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211042896.0A CN115424062B (en) | 2022-08-29 | 2022-08-29 | Method, device, equipment and storage medium for automatically identifying diagonal network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115424062A CN115424062A (en) | 2022-12-02 |
| CN115424062B true CN115424062B (en) | 2024-01-30 |
Family
ID=84199974
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211042896.0A Active CN115424062B (en) | 2022-08-29 | 2022-08-29 | Method, device, equipment and storage medium for automatically identifying diagonal network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115424062B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201514225U (en) * | 2009-09-14 | 2010-06-23 | 清华大学 | Light interference fringe reversible counting device with image display function |
| CN104240204A (en) * | 2014-09-11 | 2014-12-24 | 镇江苏仪德科技有限公司 | Solar silicon wafer and battery piece counting method based on image processing |
| CN107133922A (en) * | 2016-02-29 | 2017-09-05 | 孙智权 | A kind of silicon chip method of counting based on machine vision and image procossing |
| CN107599194A (en) * | 2017-08-24 | 2018-01-19 | 天津市环欧半导体材料技术有限公司 | A kind of silicon chip line blanking method |
-
2022
- 2022-08-29 CN CN202211042896.0A patent/CN115424062B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201514225U (en) * | 2009-09-14 | 2010-06-23 | 清华大学 | Light interference fringe reversible counting device with image display function |
| CN104240204A (en) * | 2014-09-11 | 2014-12-24 | 镇江苏仪德科技有限公司 | Solar silicon wafer and battery piece counting method based on image processing |
| CN107133922A (en) * | 2016-02-29 | 2017-09-05 | 孙智权 | A kind of silicon chip method of counting based on machine vision and image procossing |
| CN107599194A (en) * | 2017-08-24 | 2018-01-19 | 天津市环欧半导体材料技术有限公司 | A kind of silicon chip line blanking method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115424062A (en) | 2022-12-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI351654B (en) | Defect detecting device, image sensor device, imag | |
| CN113109368B (en) | Glass crack detection method, device, equipment and medium | |
| CN111612781A (en) | Screen defect detection method and device and head-mounted display equipment | |
| CN116559183B (en) | Method and system for improving defect judging efficiency | |
| CN111259891B (en) | Method, device, equipment and medium for identifying identity card in natural scene | |
| CN111192241B (en) | Quality evaluation method and device for face image and computer storage medium | |
| CN111626941A (en) | Document correction method based on deep learning semantic segmentation | |
| WO2017071406A1 (en) | Method and system for detecting pin of gold needle element | |
| CN115775244A (en) | Coil winding detection method and device, electronic equipment and storage medium | |
| KR102470422B1 (en) | Method of automatically detecting sewing stitch based on CNN feature map and system for the same | |
| CN108107611B (en) | Self-adaptive defect detection method and device and electronic equipment | |
| CN114581415A (en) | Method and device for detecting defects of PCB, computer equipment and storage medium | |
| CN112991159B (en) | Face illumination quality evaluation method, system, server and computer readable medium | |
| CN105229665A (en) | To the enhancing analysis of the snakelike belt wear assessment based on image | |
| CN115424062B (en) | Method, device, equipment and storage medium for automatically identifying diagonal network | |
| CN112669300A (en) | Defect detection method and device, computer equipment and storage medium | |
| CN115761270A (en) | Color card detection method and device, electronic equipment and storage medium | |
| CN115424181A (en) | Target object detection method and device | |
| JP6114559B2 (en) | Automatic unevenness detector for flat panel display | |
| US10958899B2 (en) | Evaluation of dynamic ranges of imaging devices | |
| WO2020107196A1 (en) | Photographing quality evaluation method and apparatus for photographing apparatus, and terminal device | |
| CN112730427B (en) | Product surface defect detection method and system based on machine vision | |
| CN113706401B (en) | Slide automatic shooting and intelligent editing method based on mobile phone camera | |
| CN116485748A (en) | Hole and slot identification detection method, system, equipment and medium | |
| JP2018021817A (en) | Inspection device and inspection method |
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 | ||
| CB02 | Change of applicant information |
Address after: 519000 unit B25, 2nd floor, building 4, No. 153, Rongao Road, Hengqin New District, Zhuhai City, Guangdong Province Applicant after: Gaojing Solar Co.,Ltd. Applicant after: Guangdong Jinwan Gaojing Solar Energy Technology Co.,Ltd. Address before: 519000 unit B25, 2nd floor, building 4, No. 153, Rongao Road, Hengqin New District, Zhuhai City, Guangdong Province Applicant before: Guangdong Gaojing Solar Energy Technology Co.,Ltd. Applicant before: Guangdong Jinwan Gaojing Solar Energy Technology Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |