CN116433664B - Panel defect detection method, device, storage medium, apparatus and program product - Google Patents
Panel defect detection method, device, storage medium, apparatus and program product Download PDFInfo
- Publication number
- CN116433664B CN116433664B CN202310692268.5A CN202310692268A CN116433664B CN 116433664 B CN116433664 B CN 116433664B CN 202310692268 A CN202310692268 A CN 202310692268A CN 116433664 B CN116433664 B CN 116433664B
- Authority
- CN
- China
- Prior art keywords
- image
- stretching
- convolution
- defect detection
- panel
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 138
- 238000001514 detection method Methods 0.000 title claims abstract description 96
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 230000004927 fusion Effects 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims description 54
- 238000010586 diagram Methods 0.000 claims description 30
- 230000015654 memory Effects 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000006073 displacement reaction Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 abstract description 10
- 230000008569 process Effects 0.000 description 23
- 238000012937 correction Methods 0.000 description 9
- 238000000605 extraction Methods 0.000 description 9
- 238000004519 manufacturing process Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 230000009467 reduction Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 230000007704 transition Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000013515 script Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06V10/32—Normalisation of the pattern dimensions
-
- 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
- G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
-
- 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
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
-
- 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
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30141—Printed circuit board [PCB]
-
- 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
Abstract
The embodiment of the application discloses a panel defect detection method, a device, a storage medium, equipment and a program product, relating to the technical field of image processing, comprising the following steps: stretching the image to be detected according to the stretching parameters to obtain a stretched image; deblurring the stretched image to obtain a target image; inputting the target image into a defect detection model to obtain defect information; the defect detection model comprises a convolution layer and a convolution kernel of the convolution layer, and is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers. The application corrects the size of the image to be detected by utilizing the stretching parameters obtained by the vibration parameters, and then the image is clear by deblurring treatment so as to facilitate the identification of the defect detection model.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a method, an apparatus, a storage medium, a device, and a program product for detecting a panel defect.
Background
In the industrial manufacturing process, the defect detection accuracy of the product often directly influences the economic benefit, the existing intelligent technology generally adopts a deep learning neural network algorithm to judge the defects of the product, and then manually re-judges the key defects, so that the labor cost is reduced. For example, in the detection of a PCB panel, the identification of defects can be completed by shooting an image directly above the PCB panel by an industrial camera and inputting the image into a detection model.
In the production and manufacture of PCB panels, the PCB panels are generally manufactured in a pipeline manner, and are sent to a fixed shooting area by a conveying device, and then sent back to the conveying device for further processing after shooting, and along with the movement of the conveying device and the operation of a driving device such as a motor, the panel can vibrate.
Disclosure of Invention
The application mainly aims to provide a panel defect detection method, a device, a storage medium, equipment and a program product, which aim to solve the problem that the defect detection on a PCB panel in a vibration state is difficult in the prior art.
In order to achieve the above object, the technical scheme adopted by the embodiment of the application is as follows:
in a first aspect, an embodiment of the present application provides a method for detecting a panel defect, including the steps of:
stretching the image to be detected according to the stretching parameters to obtain a stretched image; the method comprises the steps that an image to be detected is obtained based on the orthographic projection direction of a panel product, a stretching parameter is obtained based on the vibration parameter of the panel product, and the vibration parameter is obtained according to the vibration direction information and the vibration amplitude information of the panel product;
stretching the image to be detected according to the stretching parameters, and before obtaining the stretched image, the panel defect detection method further comprises the following steps:
acquiring a plurality of pixel points distributed along the vibration direction of the panel product on a standard image;
respectively obtaining horizontal offset information of a plurality of pixel points according to vibration amplitude information of the panel product;
constructing a horizontal offset point diagram of the pixel points according to the relative distances of the pixel points in the vibration direction and the horizontal offset information of the pixel points;
fitting the points on the horizontal offset point diagram to obtain a horizontal offset curve diagram;
obtaining a stretching parameter according to the horizontal offset curve chart;
stretching the image to be detected according to the stretching parameters to obtain a stretched image, comprising:
Stretching an image to be detected based on pixel points according to stretching parameters to obtain a stretched image;
deblurring the stretched image to obtain a target image;
inputting the target image into a defect detection model to obtain defect information; the defect detection model comprises a convolution layer and a convolution kernel of the convolution layer, and is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers.
The method comprises the steps of introducing vibration parameters of a panel product into a correction process of an image to be detected, correcting the size of the image to be detected by using stretching parameters obtained according to the vibration parameters, and then clearing the image by using deblurring treatment so as to facilitate the identification of a defect detection model.
In one possible implementation manner of the first aspect, the panel defect detection method further includes, before stretching the image to be detected according to the stretching parameter to obtain a stretched image:
and obtaining vibration parameters according to the vibration direction information and the vibration amplitude information of the panel product.
The vibration parameters are introduced and matched with the stretching parameters, and then the image to be detected is subjected to reduction stretching according to the stretching parameters, so that the correction of the size of the image to be detected is realized, in the practical situation, the vibration source is insufficient for vibrating and overturning the panel product, and from the perspective of a space coordinate system, the vibration generated by the panel product can be decomposed into the rotation of an X-axis square shaft and a Y-axis square shaft, namely the current state of the shot image to be detected can be obtained by overturning in the directions of the X-axis and the Y-axis respectively. The vibration parameters can be obtained by detecting the vibration condition of the sample, and the vibration direction information and the vibration amplitude information are included, so that the state of the corresponding product when shooting can be understood as the turning direction and the turning angle, and the larger the vibration amplitude is, the larger the turning angle of the panel product is in the vibration direction.
In a possible implementation manner of the first aspect, the method for detecting a panel defect further includes, before inputting the target image into the defect detection model to obtain defect information:
Acquiring three groups of parallel branch convolution layers; the three-component branch convolution layers respectively have convolution kernel sizes of 3*3, 1*3 and 3*1, the branch convolution layer with the convolution kernel size of 3*3 adopts distributed displacement convolution, and the branch convolution layers with the convolution kernel sizes of 1*3 and 3*1 adopt grouping convolution;
the convolution kernels of the three component branch convolution layers are fused into a 3*3 size convolution kernel to obtain the convolution kernels of the convolution layers.
The method is characterized in that the network characteristic extraction capability is enhanced, three layers of parallel branches of an asymmetric shuffling convolution structure are used for jointly extracting image characteristics, the original 3*3 convolution is replaced by convolution layers of 3*3, 1*3 and 3*1, wherein the branch convolution layers of 3*3 adopt distributed displacement convolution, the branch convolution layers of 1*3 and 3*1 adopt grouping convolution, the convolution layer of 3*3 is mainly used for increasing receptive fields to obtain richer characteristic information, the grouping convolution layer is mainly used for improving the generalization capability of a model to overturning or rotating, the dependence on space information is reduced, and model parameters can be greatly reduced.
In a possible implementation manner of the first aspect, the defect detection model further includes a channel shuffling layer, where after the branch convolution layer that adopts the group convolution, the channel shuffling layer is configured to:
Grouping the feature images extracted by the branch convolution layers to obtain a grouping matrix;
transposing the grouping matrix to obtain a transposed matrix;
and flattening the transposed matrix to regroup to obtain the target feature map.
The packet convolution can obstruct the information flow among channels, so that the channel shuffling layer is arranged and is arranged behind the packet convolution layer, and the relevance among different channels in the packet convolution can be effectively improved.
In a possible implementation manner of the first aspect, the defect detection model further includes an intermediate layer, which is obtained based on a feature layer of the feature pyramid network of the defect detection model.
Because the mapping in the feature pyramid is scaled in the X-axis and Y-axis directions and the scale difference between two adjacent layers is large, two similar objects are predicted and separated into different layers, the problem is solved by generating an intermediate layer in the feature layer of the feature pyramid network of the defect detection model, so that the transition between feature graphs with different sizes is smoother, and the detection effect is improved.
In a possible implementation manner of the first aspect, the method for detecting a panel defect further includes, before inputting the target image into the defect detection model to obtain defect information:
Respectively carrying out linear scaling on the feature layers of the feature pyramid network of the defect detection model to obtain a plurality of first feature layers;
adding the first characteristic layers pixel by pixel to obtain a second characteristic layer;
and carrying out convolution and fusion on the second characteristic layer to obtain an intermediate layer.
And linearly scaling adjacent layers of each layer in the feature pyramid network according to a certain coefficient, adding the obtained result, namely the first feature layer, pixel by pixel to obtain a second feature layer, and fusing the features from the original layer by the second feature layer through 3*3 convolution to generate an intermediate layer, so that the intermediate layers between the adjacent layers in the feature pyramid are in transition, the influence of size cutoff is reduced, the model performance is improved, and the defect extraction accuracy is further improved.
In a second aspect, an embodiment of the present application provides a panel defect detection apparatus, including:
the stretching module is used for stretching the image to be detected according to the stretching parameters to obtain a stretching image; the method comprises the steps that an image to be detected is obtained based on the orthographic projection direction of a panel product, a stretching parameter is obtained based on the vibration parameter of the panel product, and the vibration parameter is obtained according to the vibration direction information and the vibration amplitude information of the panel product;
The stretching module is also used for stretching the image to be detected according to the stretching parameters, and before the stretched image is obtained, a plurality of pixel points distributed along the vibration direction of the panel product on the standard image are obtained;
respectively obtaining horizontal offset information of a plurality of pixel points according to vibration amplitude information of the panel product;
constructing a horizontal offset point diagram of the pixel points according to the relative distances of the pixel points in the vibration direction and the horizontal offset information of the pixel points;
fitting the points on the horizontal offset point diagram to obtain a horizontal offset curve diagram;
obtaining a stretching parameter according to the horizontal offset curve chart;
stretching the image to be detected according to the stretching parameters to obtain a stretched image, comprising:
stretching an image to be detected based on pixel points according to stretching parameters to obtain a stretched image;
the deblurring module is used for deblurring the stretched image to obtain a target image;
the detection module is used for inputting the target image into the defect detection model to obtain defect information; the defect detection model comprises a convolution layer and a convolution kernel of the convolution layer, and is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers.
In a third aspect, an embodiment of the present application provides a computer readable storage medium storing a computer program, where the computer program when loaded and executed by a processor implements the panel defect detection method provided in any one of the first aspects above.
In a fourth aspect, an embodiment of the present application provides an electronic device, including a processor and a memory, where,
the memory is used for storing a computer program;
the processor is configured to load and execute a computer program to cause the electronic device to execute the panel defect detection method as provided in any one of the first aspects above.
Compared with the prior art, the application has the beneficial effects that:
the embodiment of the application provides a panel defect detection method, a device, a storage medium, equipment and a program product, wherein the method comprises the following steps: stretching the image to be detected according to the stretching parameters to obtain a stretched image; the image to be detected is obtained based on the orthographic projection direction of the panel product, and the stretching parameter is obtained based on the vibration parameter of the panel product; deblurring the stretched image to obtain a target image; inputting the target image into a defect detection model to obtain defect information; the defect detection model comprises a convolution layer and a convolution kernel of the convolution layer, and is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers. According to the method, vibration parameters of a panel product are introduced into a correction process of an image to be detected, the size of the image to be detected is corrected by utilizing stretching parameters obtained according to the vibration parameters, then the image is clarified by utilizing deblurring treatment so as to facilitate the identification of a defect detection model, and as a convolution kernel of a convolution layer of the defect detection model is obtained by fusion of convolution kernels of a plurality of parallel branch convolution layers, the model has no additional calculation amount, and can enhance the capability of feature extraction, so that the segmentation precision and generalization capability of the model are improved.
Drawings
FIG. 1 is a schematic diagram of an electronic device in a hardware operating environment according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of a method for detecting defects of a panel according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a panel product under a vibration condition in the panel defect detection method according to the embodiment of the present application;
FIG. 4 is a schematic diagram illustrating a horizontal offset of a single position point in a panel defect detection method according to an embodiment of the present application;
FIG. 5 is a diagram of a horizontal offset in a panel defect detection method according to an embodiment of the present application;
FIG. 6 is a graph showing a horizontal shift in a method for detecting a panel defect according to an embodiment of the present application;
FIG. 7 is a schematic block diagram of a panel defect detecting device according to an embodiment of the present application;
the marks in the figure: 101-processor, 102-communication bus, 103-network interface, 104-user interface, 105-memory.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The main solutions of the embodiments of the present application are: provided are a panel defect detection method, apparatus, storage medium, device, and program product, the method including: stretching the image to be detected according to the stretching parameters to obtain a stretched image; the image to be detected is obtained based on the orthographic projection direction of the panel product, and the stretching parameter is obtained based on the vibration parameter of the panel product; deblurring the stretched image to obtain a target image; inputting the target image into a defect detection model to obtain defect information; the defect detection model comprises a convolution layer and a convolution kernel of the convolution layer, and is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers.
In the industrial manufacturing process, the defect detection accuracy of the product often directly influences the economic benefit, the existing intelligent technology generally adopts a deep learning neural network algorithm to judge the defects of the product, and then manually re-judges the key defects, so that the labor cost is reduced. For example, in the detection of a PCB panel, the identification of defects can be completed by shooting an image directly above the PCB panel by an industrial camera and inputting the image into a detection model. However, because of many uncertain interference factors in the actual production environment, the background of the product is also complex, so that the effect of the deep learning algorithm on the inspection of small target defects of industrial products and dense multi-classification is still to be improved.
In the use of the deep neural network, after the image is turned over or rotated, the image features extracted by the k multiplied by k convolution layer change, and the recognition result of the model on the same target may deviate, so that the rotation of the model and the generalization capability of the turned-over target are reduced; the feature graphs in the feature pyramid can capture the visual features of objects on different scales, details such as texture edges, corners and the like are reserved in a shallow layer, more abstract semantic information is covered in a deep layer, in a real production environment, a plurality of targets with different sizes usually appear together, how to identify the different targets on one graph is a key problem, but the mapping scale of a conventional pyramid network in the X-axis direction and the Y-axis direction is large in the gap between adjacent layers, and is usually reduced in a multiple way, so that targets with similar scales can be classified on different prediction layers, a scale cut-off problem occurs, and the problems of inaccurate positioning of a prediction frame and reduced classification precision are caused.
In the production and manufacture of PCB panels, the PCB panels are generally manufactured in a pipeline manner, and are sent to a fixed shooting area by a conveying device, and then sent back to the conveying device for further processing after shooting, and along with the movement of the conveying device and the operation of a driving device such as a motor, the panel can vibrate.
Therefore, the application provides a solution, the vibration parameters of the panel products are introduced into the correction process of the image to be detected, the size of the image to be detected is corrected by utilizing the stretching parameters obtained according to the vibration parameters, then the image is clarified by utilizing the deblurring treatment, so that the defect detection model is conveniently identified, the convolution kernel of the convolution layer of the defect detection model is obtained by merging the convolution kernels of a plurality of parallel branch convolution layers, the model has no additional calculated amount, the capability of feature extraction is enhanced, the segmentation precision and the generalization capability of the model are improved, and the size deviation caused by vibration is eliminated by stretching in the steps, so that the defect information obtained by detecting the image can be highly matched with the defect information on the actual panel, and the accurate extraction of the defects on the panel under the vibration condition is realized.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an electronic device of a hardware running environment according to an embodiment of the present application, where the electronic device may include: a processor 101, such as a central processing unit (Central Processing Unit, CPU), a communication bus 102, a user interface 104, a network interface 103, a memory 105. Wherein the communication bus 102 is used to enable connected communication between these components. The user interface 104 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 104 may also include standard wired, wireless interfaces. The network interface 103 may alternatively comprise a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 105 may alternatively be a storage device independent of the foregoing processor 101, where the Memory 105 may be a high-speed random access Memory (Random Access Memory, RAM) Memory or may be a stable Non-Volatile Memory (NVM), such as at least one magnetic disk Memory; the processor 101 may be a general purpose processor including a central processing unit, a network processor, etc., as well as a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component.
It will be appreciated by those skilled in the art that the structure shown in fig. 1 is not limiting of the electronic device and may include more or fewer components than shown, or may combine certain components, or may be arranged in different components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and a panel defect detecting device may be included in the memory 105 as one storage medium.
In the electronic device shown in fig. 1, the network interface 103 is mainly used for data communication with a network server; the user interface 104 is mainly used for data interaction with a user; the processor 101 and the memory 105 in the present application may be provided in an electronic device, and the electronic device invokes the panel defect detection device stored in the memory 105 through the processor 101, and executes the panel defect detection method provided in the embodiment of the present application.
Referring to fig. 2, based on the hardware device of the foregoing embodiment, an embodiment of the present application provides a panel defect detection method, including the following steps:
s10: stretching the image to be detected according to the stretching parameters to obtain a stretched image; the image to be detected is obtained based on the orthographic projection direction of the panel product, and the stretching parameter is obtained based on the vibration parameter of the panel product.
In a specific implementation process, the image to be detected is an image obtained by shooting a panel product needing defect detection, for example, an industrial camera is fixedly arranged at a certain station, the panel product is conveyed to the orthographic projection direction under the station through a conveying device, the image of the panel product in the orthographic direction is obtained, in theory, under the condition that no artificial interference or vibration exists, the panel position cannot deviate, and the size of the image to be detected obtained by shooting the same panel product should be consistent. However, in the case of vibration, there may be relatively horizontal undulations in each direction of the panel, resulting in a change in the size of the image to be detected taken from the orthographic projection direction. Taking a change in one direction as an example, if the panel product vibrates in the left-right direction, which is equivalent to turning the panel in the left-right direction, the image to be detected can be considered to be compressed in the direction under the turning in the one direction, and the defects on the image can also change, so that accurate extraction of the defects is difficult.
Therefore, the vibration parameters are introduced and matched with the stretching parameters, and the image to be detected is subjected to reduction stretching according to the stretching parameters, so that the size of the image to be detected is corrected, namely: stretching the image to be detected according to the stretching parameters, and before obtaining the stretched image, the panel defect detection method further comprises the following steps:
And obtaining vibration parameters according to the vibration direction information and the vibration amplitude information of the panel product.
In the practical implementation process, the vibration source is insufficient to vibrate and overturn the panel product, and from the perspective of a space coordinate system, the vibration generated by the panel product can be decomposed into rotation in the directions of an X axis and a Y axis, namely the current state of the photographed image to be detected can be obtained by overturning in the directions of the X axis and the Y axis respectively. The vibration parameters can be obtained by detecting the vibration condition of the sample, and the vibration direction information and the vibration amplitude information are included, so that the state of the corresponding product when shooting can be understood as the turning direction and the turning angle, and the larger the vibration amplitude is, the larger the turning angle of the panel product is in the vibration direction. The stretching can be determined according to the vibration, that is, the stretching image to be detected according to the stretching parameters, and before the stretching image is obtained, the panel defect detecting method further comprises:
respectively obtaining stretching direction information and stretching length information according to the vibration direction information and the vibration amplitude information;
and obtaining stretching parameters according to the stretching direction information and the stretching length information.
In the specific implementation process, the stretching direction and the stretching length correspond to the vibration direction and the vibration amplitude respectively, the image to be detected is shot when the panel vibrates leftwards on the horizontal plane, the image vibrates leftwards, namely, the right side of the panel is lifted, the left side is taken as a rotation center to overturn, the image is compressed leftwards under the condition of relatively no vibration when seen from the shot image, the stretching direction is to enable the image to be stretched rightwards, the larger the amplitude is, the larger the overturning angle is, the greater the degree of compression of the image is, and therefore, the amount of the image to be stretched can be determined to be recovered, and when the size of the image is consistent with the size of the image shot when the image is not vibrated, the stretching can be considered to be completed, and the correction of the size is realized. Namely: stretching the image to be detected according to the stretching parameters to obtain a stretched image, comprising:
And stretching the image to be detected by the edge according to the stretching direction information until the stretching length information is met, so as to obtain a stretching image.
In the specific implementation process, the foregoing is only taken as an example of a single direction, and vibration may be multidirectional in actual situations, but based on the foregoing discussion, the multidirectional direction may also be gradually decomposed into a single direction, so that stretching may be performed based on the edges of the image, and after the image to be detected is stretched from the edges until the edges are stretched to the positions meeting the stretching length information, the dimensions of the image are corrected.
In one embodiment, according to the stretching direction information, stretching the image to be detected from the edge until the stretching length information is satisfied, and before obtaining the stretched image, the panel defect detection method further includes:
overlapping any edge of the image to be detected with the corresponding edge of the standard image;
stretching the image to be detected by the edge according to the stretching direction information until the stretching length information is met, and obtaining a stretching image, wherein the stretching image comprises the following steps:
and according to the stretching direction information, stretching the edge of the image to be detected, which is not overlapped with the standard image, and stretching the edge of the image to be detected, which is overlapped with the standard image, until the stretching length information is met, so as to obtain a stretching image.
In the specific implementation process, because the vibration is tiny, the influence is larger under the precision of the defect information to be extracted, but in the naked eye observation, the deflection of the panel product caused by the vibration is tiny and not easy to observe, that is to say, the deflection of the dimension is tiny, in the stretching reduction process, any edge of the image to be detected can be directly overlapped with the corresponding edge of the standard image, the stretching adjustment times are reduced, and the standard image is the image to be detected which is obtained by shooting at the same position under the condition of no vibration. In the stretching and restoring process, the rest edges are stretched based on the overlapped edges, and the overlapped edges are overlapped with the standard image edges, but the overlapped edges are compressed due to overturning caused by vibration in the direction parallel to the overlapped edges, so that the length of the overlapped edges is matched with the standard image after the length of the overlapped edges meets the stretching length information, and a stretching image is obtained.
In one embodiment, the panel defect detection method further includes, before stretching the image to be detected according to the stretching parameter to obtain a stretched image:
acquiring a plurality of pixel points distributed along the vibration direction of the panel product on a standard image;
Respectively obtaining horizontal offset information of a plurality of pixel points according to vibration amplitude information of the panel product;
constructing a horizontal offset point diagram of the pixel points according to the relative distances of the pixel points in the vibration direction and the horizontal offset information of the pixel points;
fitting the points on the horizontal offset point diagram to obtain a horizontal offset curve diagram;
from the horizontal offset plot, the stretch parameters are obtained.
In a specific implementation process, in order to further improve the precision of stretching correction, the stretching parameters are acquired based on pixel points. As is also apparent from the foregoing, in the case of analyzing the case when the panel is turned over to a certain position during vibration, if the panel is turned over by vibration around the left side, then the horizontal offset occurring at a position farther from the left side is larger than the horizontal offset occurring at the right side, that is, the image is compressed to a different extent, with respect to the left side, as shown in fig. 3, which is a schematic diagram of a panel product in the case of vibration, the horizontal solid line frame is the panel product, the dotted line frame is the position in the case of vibration, the point positions at different positions of the panel, the positions before and after vibration and the orthographic projection point after vibration form a right triangle, as shown in fig. 4, the horizontal offset schematic diagram of a single position point, where H represents the height difference before and after vibration, D represents the horizontal displacement before and after vibration, L represents the distance between the positions before and after vibration, H, D, L forms a right triangle, and the point vibration amplitude is the same, that the point positions on the panel is the same, that is the included angle between the panel vibrations is smaller as the point positions on the panel is the same, and the corresponding point positions are smaller as the point positions on the left side. Although the accuracy of the corrected defect information may be within the allowable range in the case where the vibration amplitude is small in the stretching of the same ratio, it is necessary to perform different stretching reduction for different positions according to the distance from the center of inversion in order to correct the image to be detected more accurately.
Therefore, in this embodiment, similar to the foregoing principle, stretching reduction is decomposed into each vibration direction, pixel points distributed in the vibration direction are obtained, and then the horizontal offset of each pixel point is obtained according to the amplitude information of the vibration. And then, constructing a horizontal offset point diagram by taking the relative distance and the horizontal offset as the horizontal and vertical coordinates, wherein the horizontal and vertical coordinates are based on pixels, and the relative distance is shown in the figure 5, the relative distance is the distance between the pixel point and the turning center turned by vibration, then, carrying out linear fitting on the points on the point diagram, and according to the curve on the fitted curve diagram, carrying out one-to-one correspondence on all the pixel points in the vibration direction and the horizontal displacement of the pixel point, wherein the obtained horizontal offset curve diagram is shown in the figure 6, and corresponds to the situation that the relative distance is far and the horizontal offset is larger, the slope of the position on the curve is larger, which shows that the horizontal offset of the position is far is faster, and the stretching parameter, namely the position with the larger horizontal offset, needs a greater degree of stretching reduction according to the curve diagram, so that the actual image to be detected is more accurate in a stretching correction reduction manner, and the defect information is more accurately extracted.
Stretching the image to be detected according to the stretching parameters based on the stretching parameters obtained by the pixel points in the previous step to obtain a stretched image, comprising:
and stretching the image to be detected based on the pixel points according to the stretching parameters to obtain a stretched image.
S20: and performing deblurring treatment on the stretched image to obtain a target image.
In a specific implementation process, deblurring treatment, namely, an image restoration technology or a deblurring technology is used for clearing an image, so that the problem of blurring of the image to be detected in a shooting vibration state is solved. In terms of technology, the blurred image processing method is mainly divided into three major categories, namely image enhancement, image restoration and super-resolution reconstruction. The image restoration function of the common office software such as PS and WPS can be utilized to perform deblurring, and common filtering processing or convolutional neural network can also be utilized, so that the principle of deblurring processing can refer to the prior art, and the description is omitted here.
S30: inputting the target image into a defect detection model to obtain defect information; the defect detection model comprises a convolution layer and a convolution kernel of the convolution layer, and is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers.
In the implementation process, the defect detection model is a model obtained by training a convolutional neural network, and the model can learn defect information through training so as to enable the model to have the capability of identifying defects from an input image. The method has the advantages that the advantages of strong feature extraction capability and strong generalization capability of the asymmetric structure are effectively utilized, and the network parameter quantity is reduced.
The embodiment of the application provides an implementation mode for replacing an original 3*3 convolution layer by a three-component convolution layer, in particular to a panel defect detection method before inputting a target image into a defect detection model to obtain defect information, which comprises the following steps:
acquiring three groups of parallel branch convolution layers; the three-component branch convolution layers respectively have convolution kernel sizes of 3*3, 1*3 and 3*1, the branch convolution layer with the convolution kernel size of 3*3 adopts distributed displacement convolution, and the branch convolution layers with the convolution kernel sizes of 1*3 and 3*1 adopt grouping convolution;
the convolution kernels of the three component branch convolution layers are fused into a 3*3 size convolution kernel to obtain the convolution kernels of the convolution layers.
In the specific implementation process, the image characteristics are extracted together by using three layers of parallel branches of an asymmetric shuffling convolution structure, and the original 3*3 convolution is replaced by using convolution layers of 3*3, 1*3 and 3*1, wherein the branch convolution layers of 3*3 adopt distributed displacement convolution, the branch convolution layers of 1*3 and 3*1 adopt grouping convolution, the convolution layer of 3*3 is mainly used for increasing receptive fields to obtain richer characteristic information, the grouping convolution layer is mainly used for improving the generalization capability of a model to overturning or rotating, the dependence on spatial information is reduced, and model parameters can be greatly reduced.
In this embodiment, the vibration parameters of the panel product are introduced into the correction process of the image to be detected, the size of the image to be detected is corrected by using the stretching parameters obtained according to the vibration parameters, and then the image is clarified by using the deblurring process, so that the defect detection model is identified.
In one embodiment, the defect detection model further comprises a channel shuffling layer, after the branch convolution layer employing the group convolution, the channel shuffling layer to:
grouping the feature images extracted by the branch convolution layers to obtain a grouping matrix;
transposing the grouping matrix to obtain a transposed matrix;
and flattening the transposed matrix to regroup to obtain the target feature map.
In the implementation process, the packet convolution can obstruct the information flow among channels, so that the channel shuffling layer is arranged and is arranged behind the packet convolution layer, and the relevance among different channels in the packet convolution can be effectively improved. The specific process of channel shuffling is that the input feature images are grouped into m groups, n is the number of channels in each group, and the input feature matrix vector is transformed into (m, n); performing transposition operation to transform into a transposed matrix of (n, m); and flattening the obtained result, and grouping the result again to n groups to form a new characteristic diagram as an output, namely a target characteristic diagram.
In one embodiment, the defect detection model further comprises an intermediate layer, the intermediate layer being obtained based on the feature layers of the feature pyramid network of the defect detection model.
In the implementation process, as the mapping in the feature pyramid is scaled in the X-axis direction and the Y-axis direction and the scale difference between two adjacent layers is large, two objects with similar sizes are predicted and separated into different layers, the problem is solved by generating an intermediate layer in the feature layer of the feature pyramid network of the defect detection model, so that the transition between feature graphs with different sizes is smoother, and the detection effect is improved. Specifically, an obtaining manner of the intermediate layer is provided, that is, a target image is input into a defect detection model, and before defect information is obtained, the panel defect detection method further includes:
respectively carrying out linear scaling on the feature layers of the feature pyramid network of the defect detection model to obtain a plurality of first feature layers;
adding the first characteristic layers pixel by pixel to obtain a second characteristic layer;
and carrying out convolution and fusion on the second characteristic layer to obtain an intermediate layer.
In the specific implementation process, the adjacent layers of each layer in the feature pyramid network are linearly scaled according to a certain coefficient, the obtained result, namely the first feature layer is added pixel by pixel to obtain the second feature layer, the second feature layer is convolved through 3*3 to fuse the features from the original layer, and an intermediate layer is generated, so that the adjacent layers in the feature pyramid are all in transition, the influence of size cutoff is reduced, the model performance is improved, and the defect extraction accuracy is further improved.
Referring to fig. 7, an embodiment of the present application also provides a panel defect detecting apparatus based on the same inventive concept as that of the previous embodiment, the apparatus including:
the stretching module is used for stretching the image to be detected according to the stretching parameters to obtain a stretching image; the image to be detected is obtained based on the orthographic projection direction of the panel product, and the stretching parameter is obtained based on the vibration parameter of the panel product;
the deblurring module is used for deblurring the stretched image to obtain a target image;
the detection module is used for inputting the target image into the defect detection model to obtain defect information; the defect detection model comprises a convolution layer and a convolution kernel of the convolution layer, and is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers.
It should be understood by those skilled in the art that the division of each module in the embodiment is merely a division of a logic function, and may be fully or partially integrated onto one or more actual carriers in practical application, and the modules may be fully implemented in a form of calling by a processing unit through software, may be fully implemented in a form of hardware, or may be implemented in a form of combining software and hardware, and it should be noted that each module in the panel defect detection apparatus in this embodiment is in one-to-one correspondence with each step in the panel defect detection method in the foregoing embodiment, so that a specific implementation of this embodiment may refer to an implementation manner of the foregoing panel defect detection method, and will not be repeated herein.
Based on the same inventive concept as in the previous embodiments, embodiments of the present application also provide a computer readable storage medium storing a computer program, which when loaded and executed by a processor, implements a panel defect detection method as provided in the embodiments of the present application.
Based on the same inventive concept as in the previous embodiments, an embodiment of the present application further provides an electronic device, including a processor and a memory, wherein,
the memory is used for storing a computer program;
the processor is used for loading and executing the computer program so as to enable the electronic equipment to execute the panel defect detection method provided by the embodiment of the application.
Furthermore, based on the same inventive concept as in the previous embodiments, embodiments of the present application also provide a computer program product including a computer program for executing the panel defect detection method as provided in the embodiments of the present application when the computer program is executed.
In some embodiments, the computer readable storage medium may be FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; but may be a variety of devices including one or any combination of the above memories. The computer may be a variety of computing devices including smart terminals and servers.
In some embodiments, the executable instructions may be in the form of programs, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and they may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.
As an example, the executable instructions may, but need not, correspond to files in a file system, may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a hypertext markup language (HTML, hyper Text Markup Language) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub-programs, or portions of code).
As an example, executable instructions may be deployed to be executed on one computing device or on multiple computing devices located at one site or, alternatively, distributed across multiple sites and interconnected by a communication network.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present application are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of embodiments, it will be clear to a person skilled in the art that the above embodiment method may be implemented by means of software plus a necessary general hardware platform, but may of course also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. read-only memory/random-access memory, magnetic disk, optical disk) comprising several instructions for causing a multimedia terminal device (which may be a mobile phone, a computer, a television receiver, or a network device, etc.) to perform the method according to the embodiments of the present application.
In summary, the method, apparatus, storage medium, device and program product for detecting panel defects provided by the present application include: stretching the image to be detected according to the stretching parameters to obtain a stretched image; the image to be detected is obtained based on the orthographic projection direction of the panel product, and the stretching parameter is obtained based on the vibration parameter of the panel product; deblurring the stretched image to obtain a target image; inputting the target image into a defect detection model to obtain defect information; the defect detection model comprises a convolution layer and a convolution kernel of the convolution layer, and is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers. According to the method, vibration parameters of a panel product are introduced into a correction process of an image to be detected, the size of the image to be detected is corrected by utilizing stretching parameters obtained according to the vibration parameters, then the image is clarified by utilizing deblurring treatment so as to facilitate the identification of a defect detection model, and as a convolution kernel of a convolution layer of the defect detection model is obtained by fusion of convolution kernels of a plurality of parallel branch convolution layers, the model has no additional calculation amount, and can enhance the capability of feature extraction, so that the segmentation precision and generalization capability of the model are improved.
The foregoing description of the preferred embodiments of the application is not intended to limit the application to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the application are intended to be included within the scope of the application.
Claims (9)
1. A method for detecting defects of a panel, comprising the steps of:
stretching the image to be detected according to the stretching parameters to obtain a stretched image; the image to be detected is obtained based on the orthographic projection direction of the panel product, the stretching parameter is obtained based on the vibration parameter of the panel product, and the vibration parameter is obtained according to the vibration direction information and the vibration amplitude information of the panel product;
the panel defect detection method further comprises the steps of:
acquiring a plurality of pixel points distributed along the vibration direction of the panel product on a standard image;
respectively obtaining horizontal offset information of a plurality of pixel points according to the vibration amplitude information of the panel product;
constructing a horizontal offset point diagram of the pixel points according to the relative distances of the pixel points in the vibration direction and the horizontal offset information of the pixel points;
Fitting the points on the horizontal offset point diagram to obtain a horizontal offset curve diagram;
obtaining the stretching parameter according to the horizontal offset curve chart;
stretching the image to be detected according to the stretching parameters to obtain a stretched image, including:
stretching the image to be detected based on the pixel points according to the stretching parameters to obtain a stretched image;
deblurring the stretched image to obtain a target image;
inputting the target image into a defect detection model to obtain defect information; the defect detection model comprises a convolution layer, wherein the convolution kernel of the convolution layer is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers.
2. The method for detecting a panel defect according to claim 1, wherein the method for detecting a panel defect further comprises, before stretching the image to be detected according to the stretching parameter to obtain a stretched image:
and obtaining the vibration parameters according to the vibration direction information and the vibration amplitude information of the panel product.
3. The method of detecting a panel defect according to claim 1, wherein before inputting the target image into a defect detection model to obtain defect information, the method further comprises:
Three groups of parallel branch convolution layers are obtained; the convolution kernel sizes of the three groups of branch convolution layers are 3*3, 1*3 and 3*1 respectively, the branch convolution layer with the convolution kernel size of 3*3 adopts distributed displacement convolution, and the branch convolution layers with the convolution kernel sizes of 1*3 and 3*1 adopt grouping convolution;
the convolution kernels of the three sets of branch convolution layers are fused into a 3*3-sized convolution kernel to obtain the convolution kernels of the convolution layers.
4. A method of panel defect detection according to claim 3 wherein the defect detection model further comprises a channel shuffle layer located after the branch convolution layer employing group convolution, the channel shuffle layer being to:
grouping the feature images extracted by the branch convolution layers to obtain a grouping matrix;
transposing the grouping matrix to obtain a transposed matrix;
and flattening the transposed matrix to regroup to obtain a target feature map.
5. The method of claim 1, wherein the defect detection model further comprises an intermediate layer, the intermediate layer being obtained based on a feature layer of a feature pyramid network of the defect detection model.
6. The method of detecting a panel defect according to claim 5, wherein before inputting the target image into a defect detection model to obtain defect information, the method further comprises:
respectively carrying out linear scaling on the feature layers of the feature pyramid network of the defect detection model to obtain a plurality of first feature layers;
adding the first characteristic layers pixel by pixel to obtain a second characteristic layer;
and carrying out convolution and fusion on the second characteristic layer to obtain the intermediate layer.
7. A panel defect detection apparatus, comprising:
the stretching module is used for stretching the image to be detected according to the stretching parameters to obtain a stretched image; the image to be detected is obtained based on the orthographic projection direction of the panel product, the stretching parameter is obtained based on the vibration parameter of the panel product, and the vibration parameter is obtained according to the vibration direction information and the vibration amplitude information of the panel product;
the stretching module is also used for obtaining a plurality of pixel points distributed along the vibration direction of the panel product on the standard image before stretching the image to be detected according to the stretching parameters to obtain a stretched image;
Respectively obtaining horizontal offset information of a plurality of pixel points according to the vibration amplitude information of the panel product;
constructing a horizontal offset point diagram of the pixel points according to the relative distances of the pixel points in the vibration direction and the horizontal offset information of the pixel points;
fitting the points on the horizontal offset point diagram to obtain a horizontal offset curve diagram;
obtaining the stretching parameter according to the horizontal offset curve chart;
stretching the image to be detected according to the stretching parameters to obtain a stretched image, including:
stretching the image to be detected based on the pixel points according to the stretching parameters to obtain a stretched image;
the deblurring module is used for deblurring the stretched image to obtain a target image;
the detection module is used for inputting the target image into a defect detection model to obtain defect information; the defect detection model comprises a convolution layer, wherein the convolution kernel of the convolution layer is obtained based on the fusion of the convolution kernels of a plurality of parallel branch convolution layers.
8. A computer readable storage medium storing a computer program, wherein the computer program, when loaded and executed by a processor, implements the panel defect detection method according to any one of claims 1-6.
9. An electronic device comprising a processor and a memory, wherein,
the memory is used for storing a computer program;
the processor is configured to load and execute the computer program to cause the electronic device to perform the panel defect detection method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310692268.5A CN116433664B (en) | 2023-06-13 | 2023-06-13 | Panel defect detection method, device, storage medium, apparatus and program product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310692268.5A CN116433664B (en) | 2023-06-13 | 2023-06-13 | Panel defect detection method, device, storage medium, apparatus and program product |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116433664A CN116433664A (en) | 2023-07-14 |
| CN116433664B true CN116433664B (en) | 2023-09-01 |
Family
ID=87091108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310692268.5A Active CN116433664B (en) | 2023-06-13 | 2023-06-13 | Panel defect detection method, device, storage medium, apparatus and program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116433664B (en) |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103620928A (en) * | 2011-03-17 | 2014-03-05 | 联合活跃驱动公司 | Asymmetric and general vibration waveforms from a plurality of synchronized vibration actuators |
| CN105046695A (en) * | 2015-07-01 | 2015-11-11 | 华南理工大学 | Projective capacitive touch panel ITO (Indium Tin Oxide) circuit defect detection method based on one-dimensional image |
| CN105510348A (en) * | 2015-12-31 | 2016-04-20 | 南京协辰电子科技有限公司 | Flaw detection method and device of printed circuit board and detection equipment |
| CN107123107A (en) * | 2017-03-24 | 2017-09-01 | 广东工业大学 | Cloth defect inspection method based on neutral net deep learning |
| CN109472760A (en) * | 2019-02-01 | 2019-03-15 | 深兰人工智能芯片研究院（江苏）有限公司 | A kind of method, apparatus of correcting distorted image |
| CN109658823A (en) * | 2019-02-27 | 2019-04-19 | 上海天马微电子有限公司 | Display panel and display device |
| CN110321908A (en) * | 2018-03-29 | 2019-10-11 | 华为技术有限公司 | Image-recognizing method, terminal device and computer readable storage medium |
| CN112508826A (en) * | 2020-11-16 | 2021-03-16 | 哈尔滨工业大学（深圳） | Printed matter defect detection method based on feature registration and gradient shape matching fusion |
| CN113256485A (en) * | 2021-05-21 | 2021-08-13 | 百果园技术(新加坡)有限公司 | Image stretching method, device, electronic equipment and storage medium |
| CN113255654A (en) * | 2020-02-12 | 2021-08-13 | 斯凯孚公司 | Imaging processing method of characteristic data and application thereof |
| CN113496279A (en) * | 2020-04-08 | 2021-10-12 | 脸谱公司 | Packet convolution for channel convolution engine using point-to-point connections |
| WO2021217857A1 (en) * | 2020-04-27 | 2021-11-04 | 平安科技（深圳）有限公司 | Slice defect detection method and apparatus, and electronic device and readable storage medium |
| CN113887491A (en) * | 2021-10-21 | 2022-01-04 | 江南大学 | Human skeleton behavior recognition system and method based on cross-space-time graph convolutional network |
| CN114075751A (en) * | 2020-08-19 | 2022-02-22 | 天津海尔洗涤电器有限公司 | Device with vibration detection and water level sensing functions and washing machine |
| CN114364075A (en) * | 2021-12-29 | 2022-04-15 | 合肥维信诺科技有限公司 | Stretchable display panel, stretchable display device and preparation method |
| CN114577906A (en) * | 2022-02-23 | 2022-06-03 | 韶关东阳光自动化设备有限公司 | Positive corrosion foil micro-perforation defect ultrasonic guided wave detection method and system |
| CN115393252A (en) * | 2021-05-25 | 2022-11-25 | 武汉Tcl集团工业研究院有限公司 | Defect detection method and device for display panel, electronic equipment and storage medium |
| CN115471439A (en) * | 2021-06-11 | 2022-12-13 | 武汉Tcl集团工业研究院有限公司 | Method and device for identifying defects of display panel, electronic equipment and storage medium |
| CN115497076A (en) * | 2022-10-09 | 2022-12-20 | 江苏智能无人装备产业创新中心有限公司 | High-precision and high-efficiency signal identification detection method, device and medium |
| CN115661161A (en) * | 2022-12-29 | 2023-01-31 | 成都数联云算科技有限公司 | Method, device, storage medium, equipment and program product for detecting defects of parts |
| CN116229199A (en) * | 2022-12-15 | 2023-06-06 | 北京市地铁运营有限公司技术创新研究院分公司 | Target detection method based on model light weight |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IT201900005826A1 (en) * | 2019-04-16 | 2020-10-16 | Santex Rimar Group S R L | DEVICE AND METHOD FOR REAL TIME DETECTION OF DEFECTS IN FABRICS, DURING WEAVING |
| CN111768386B (en) * | 2020-06-30 | 2024-02-20 | 北京百度网讯科技有限公司 | Product defect detection method, device, electronic equipment and storage medium |
| AU2022346778A1 (en) * | 2021-09-15 | 2024-04-18 | Amsted Rail Company, Inc. | Mobile railway asset monitoring apparatus and methods |
-
2023
- 2023-06-13 CN CN202310692268.5A patent/CN116433664B/en active Active
Patent Citations (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103620928A (en) * | 2011-03-17 | 2014-03-05 | 联合活跃驱动公司 | Asymmetric and general vibration waveforms from a plurality of synchronized vibration actuators |
| CN105046695A (en) * | 2015-07-01 | 2015-11-11 | 华南理工大学 | Projective capacitive touch panel ITO (Indium Tin Oxide) circuit defect detection method based on one-dimensional image |
| CN105510348A (en) * | 2015-12-31 | 2016-04-20 | 南京协辰电子科技有限公司 | Flaw detection method and device of printed circuit board and detection equipment |
| CN107123107A (en) * | 2017-03-24 | 2017-09-01 | 广东工业大学 | Cloth defect inspection method based on neutral net deep learning |
| CN110321908A (en) * | 2018-03-29 | 2019-10-11 | 华为技术有限公司 | Image-recognizing method, terminal device and computer readable storage medium |
| CN109472760A (en) * | 2019-02-01 | 2019-03-15 | 深兰人工智能芯片研究院（江苏）有限公司 | A kind of method, apparatus of correcting distorted image |
| CN109658823A (en) * | 2019-02-27 | 2019-04-19 | 上海天马微电子有限公司 | Display panel and display device |
| CN113255654A (en) * | 2020-02-12 | 2021-08-13 | 斯凯孚公司 | Imaging processing method of characteristic data and application thereof |
| CN113496279A (en) * | 2020-04-08 | 2021-10-12 | 脸谱公司 | Packet convolution for channel convolution engine using point-to-point connections |
| EP3968237A1 (en) * | 2020-04-08 | 2022-03-16 | Facebook, Inc. | Grouped convolution using point-to-point connected channel convolution engines |
| WO2021217857A1 (en) * | 2020-04-27 | 2021-11-04 | 平安科技（深圳）有限公司 | Slice defect detection method and apparatus, and electronic device and readable storage medium |
| CN114075751A (en) * | 2020-08-19 | 2022-02-22 | 天津海尔洗涤电器有限公司 | Device with vibration detection and water level sensing functions and washing machine |
| CN112508826A (en) * | 2020-11-16 | 2021-03-16 | 哈尔滨工业大学（深圳） | Printed matter defect detection method based on feature registration and gradient shape matching fusion |
| CN113256485A (en) * | 2021-05-21 | 2021-08-13 | 百果园技术(新加坡)有限公司 | Image stretching method, device, electronic equipment and storage medium |
| CN115393252A (en) * | 2021-05-25 | 2022-11-25 | 武汉Tcl集团工业研究院有限公司 | Defect detection method and device for display panel, electronic equipment and storage medium |
| CN115471439A (en) * | 2021-06-11 | 2022-12-13 | 武汉Tcl集团工业研究院有限公司 | Method and device for identifying defects of display panel, electronic equipment and storage medium |
| CN113887491A (en) * | 2021-10-21 | 2022-01-04 | 江南大学 | Human skeleton behavior recognition system and method based on cross-space-time graph convolutional network |
| CN114364075A (en) * | 2021-12-29 | 2022-04-15 | 合肥维信诺科技有限公司 | Stretchable display panel, stretchable display device and preparation method |
| CN114577906A (en) * | 2022-02-23 | 2022-06-03 | 韶关东阳光自动化设备有限公司 | Positive corrosion foil micro-perforation defect ultrasonic guided wave detection method and system |
| CN115497076A (en) * | 2022-10-09 | 2022-12-20 | 江苏智能无人装备产业创新中心有限公司 | High-precision and high-efficiency signal identification detection method, device and medium |
| CN116229199A (en) * | 2022-12-15 | 2023-06-06 | 北京市地铁运营有限公司技术创新研究院分公司 | Target detection method based on model light weight |
| CN115661161A (en) * | 2022-12-29 | 2023-01-31 | 成都数联云算科技有限公司 | Method, device, storage medium, equipment and program product for detecting defects of parts |
Non-Patent Citations (1)
| Title |
|---|
| "基于无人机视觉的起重机表面裂纹检测方法";周前飞等;《测控技术》;第41卷(第4期);第28-34页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116433664A (en) | 2023-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108009543B (en) | License plate recognition method and device | |
| CN109840477B (en) | Method and device for recognizing shielded face based on feature transformation | |
| CN110033009B (en) | Method for processing image data in a connected network | |
| CN112348116B (en) | Target detection method and device using space context and computer equipment | |
| CN111179270A (en) | Image co-segmentation method and device based on attention mechanism | |
| CN110070503A (en) | Scale calibration method, system and medium based on convolutional neural networks | |
| CN113112511B (en) | Method and device for correcting test paper, storage medium and electronic equipment | |
| CN116433664B (en) | Panel defect detection method, device, storage medium, apparatus and program product | |
| CN109523570B (en) | Motion parameter calculation method and device | |
| CN108734712B (en) | Background segmentation method and device and computer storage medium | |
| CN112883983A (en) | Feature extraction method and device and electronic system | |
| CN115661156B (en) | Image generation method, image generation device, storage medium, image generation apparatus, and computer program product | |
| CN115345895A (en) | Image segmentation method and device for visual detection, computer equipment and medium | |
| CN111815628B (en) | Display panel defect detection method, device, equipment and readable storage medium | |
| CN110853087B (en) | Parallax estimation method, device, storage medium and terminal | |
| CN111401365B (en) | OCR image automatic generation method and device | |
| CN112529943B (en) | Object detection method, object detection device and intelligent equipment | |
| CN115063813A (en) | Training method and training device of alignment model aiming at character distortion | |
| CN110636373B (en) | Image processing method and device and electronic equipment | |
| CN114140429A (en) | Real-time parking space detection method and device for vehicle end | |
| CN114758145A (en) | Image desensitization method and device, electronic equipment and storage medium | |
| CN114119376A (en) | Image processing method and device, electronic equipment and storage medium | |
| CN116740385B (en) | Equipment quality inspection method, device and system | |
| CN110738212B (en) | Method and device for acquiring convolutional network model and processing equipment | |
| CN117132767B (en) | Small target detection method, device, equipment and readable storage medium |
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 |