CN109871895B - Method and device for detecting defects of circuit board - Google Patents
Method and device for detecting defects of circuit board Download PDFInfo
- Publication number
- CN109871895B CN109871895B CN201910133359.9A CN201910133359A CN109871895B CN 109871895 B CN109871895 B CN 109871895B CN 201910133359 A CN201910133359 A CN 201910133359A CN 109871895 B CN109871895 B CN 109871895B
- Authority
- CN
- China
- Prior art keywords
- image
- circuit board
- classification model
- pixel
- defect
- 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
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The application provides a method and a device for detecting defects of a circuit board, wherein the method comprises the following steps: the method comprises the steps of obtaining a shot image of a circuit board to be tested, carrying out pixel information difference on the shot image and pixels corresponding to a standard image to obtain a difference value of the corresponding pixels, synthesizing the pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image, inputting the input image into a trained classification model, and determining defect information of the circuit board to be tested according to the output of the classification model, wherein the classification model learns the image characteristics of the input image corresponding to the circuit board with various defects. Because the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects, the input data of the classification model can be enriched by adding the difference value into the pixel information of the shot image for image characteristic extraction, and the accuracy and the efficiency of the defect detection of the circuit boards to be detected are improved.
Description
Technical Field
The application relates to the technical field of deep learning and image processing, in particular to a method and a device for detecting defects of a circuit board.
Background
At present, the circuit board is manufactured by depending on an automatic industrial production line, because the integration level of electronic components of the circuit board is continuously increased, the production process of the circuit board is more and more complex, and a defective circuit board can be inevitably generated in the process of generating the circuit board, so that the defect detection is carried out on the circuit board, and the damage of the circuit board caused by the defect can be avoided.
In the prior art, a circuit board generation enterprise mainly adopts a manual detection method to detect the defects of the circuit board. However, manual detection requires a worker to check with naked eyes, and has the disadvantages of high detection cost, low accuracy, low efficiency and the like.
Disclosure of Invention
The present application is directed to solving, at least to some extent, one of the technical problems in the related art.
The application provides a method and a device for detecting the defects of a circuit board, which solve the technical problems of low accuracy, high detection cost and low efficiency when the defect detection of the circuit board depends on manual detection in the prior art.
An embodiment of a first aspect of the present application provides a method for detecting defects of a circuit board, including:
acquiring a shot image of a circuit board to be tested;
carrying out pixel information difference on the corresponding pixels of the shot image and the standard image to obtain a difference value of the corresponding pixels; the standard image is obtained by shooting a reference circuit board without defects;
synthesizing the pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image;
inputting the input image into a trained classification model to determine the defect information of the circuit board to be tested according to the output of the classification model; and the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects.
As a first possible implementation manner of the embodiment of the application, the pixel information of each pixel in the captured image and the standard image includes a depth value and a grayscale value of each color channel.
As a second possible implementation manner of the embodiment of the present application, the differentiating the pixel information of the corresponding pixels of the captured image and the standard image to obtain a differential value of the corresponding pixels includes:
carrying out depth value difference on the corresponding pixels in the shot image and the standard image to obtain the depth difference value of the corresponding pixels;
and carrying out gray value difference on the corresponding pixels in the shot image and the standard image to obtain gray value difference values of the corresponding pixels in each color channel.
As a third possible implementation manner of the embodiment of the present application, the synthesizing the difference value between the pixel information of each pixel of the captured image and the corresponding pixel to obtain the input image includes:
and taking the depth difference value, the depth value, and the gray difference value and the gray value of each color channel as the pixel information of the corresponding pixel in the input image.
As a fourth possible implementation manner of the embodiment of the present application, the classification model includes a convolution layer, a pooling layer, and a full-link layer;
the convolutional layer is used for extracting image characteristics of pixel information of each pixel in the input image;
the pooling layer is used for performing dimension reduction operation on the features extracted from the convolutional layer;
and the full connection layer is used for classifying according to the image characteristics after the dimensionality reduction of the pooling layer.
As a fifth possible implementation manner of the embodiment of the application, the classification model is obtained by synthesizing a difference value obtained by performing pixel information difference on a sample image and a pixel corresponding to the standard image with pixel information of a corresponding pixel in the sample image, and training the synthesized sample image; the sample images are marked with defect information, and the defect information is used for indicating whether the circuit board displayed by the corresponding sample image has defects or not and indicating the defect types;
and when the difference between the defect information output by the classification model and the defect information labeled by the sample image is minimized, finishing the training of the classification model.
As a sixth possible implementation manner of the embodiment of the present application, after the inputting the input image into a trained classification model to determine the defect information of the tested circuit board according to the output of the classification model, the method further includes:
if the circuit board to be tested has defects, displaying the shot image on a control interface;
acquiring defect information obtained by marking the type of the artificial defect of the shot image;
generating a first training sample according to the shot image and the defect information;
and training the classification model by adopting the first training sample.
As a seventh possible implementation manner of the embodiment of the present application, after the inputting the input image into a trained classification model to determine the defect information of the tested circuit board according to the output of the classification model, the method further includes:
selecting part of circuit boards from the circuit boards to be tested without defects, and displaying the shot images of the selected part of circuit boards on a control interface;
and acquiring the defect information of the corresponding shot image manual marking for the selected partial circuit boards so as to generate a second training sample for training the classification model.
As an eighth possible implementation manner of the embodiment of the present application, after the inputting the input image into the trained classification model to determine the defect information of the tested circuit board according to the output of the classification model, the method further includes:
if the tested circuit board has defects, controlling a production line so as to place the tested circuit board with defects in a set area.
The defect detection method of the embodiment of the application comprises the steps of obtaining a shot image of a circuit board to be detected, carrying out pixel information difference on corresponding pixels of the shot image and a standard image to obtain a difference value of the corresponding pixels, wherein the standard image is obtained by shooting a reference circuit board without defects, synthesizing pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image, inputting the input image into a trained classification model, and determining the defect information of the circuit board to be detected according to the output of the classification model. Because the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects, the input data of the classification model can be enriched by synthesizing the difference value and the pixel information of the shot image and then extracting the image characteristics, and the accuracy and the efficiency of the defect detection of the circuit boards to be detected are improved.
The embodiment of the second aspect of the present application provides a defect detecting apparatus for a circuit board, including:
the acquisition module is used for acquiring a shot image of the circuit board;
the difference module is used for carrying out pixel information difference on the corresponding pixels of the shot image and the standard image to obtain a difference value of the corresponding pixels; the standard image is obtained by shooting a reference circuit board without defects;
the processing module is used for synthesizing the pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image;
the detection module is used for inputting the input image into the trained classification model so as to determine the defect information of the circuit board to be tested according to the output of the classification model; and the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects.
The defect detection device of the embodiment of the application obtains the shot image of the circuit board to be detected; the method comprises the steps of carrying out pixel information difference on corresponding pixels of a shot image and a standard image to obtain a difference value of the corresponding pixels, wherein the standard image is obtained by shooting a reference circuit board without defects, synthesizing pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image, inputting the input image into a trained classification model, and determining defect information of the circuit board to be tested according to the output of the classification model. Because the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects, the input data of the classification model can be enriched by synthesizing the difference value and the pixel information of the shot image and then extracting the image characteristics, and the accuracy and the efficiency of the defect detection of the circuit boards to be detected are improved.
An embodiment of a third aspect of the present application provides a computer device, including: the circuit board defect detection method comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein when the processor executes the program, the defect detection method of the circuit board is realized.
A fourth aspect of the present application provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for detecting defects of a circuit board as described in the above embodiments.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flowchart of a method for detecting defects of a circuit board according to an embodiment of the present disclosure;
fig. 2 is a schematic flowchart of a classification model training method according to an embodiment of the present application;
FIG. 3 is a diagram illustrating a classification model training method according to an embodiment of the present disclosure;
FIG. 4 is a schematic flowchart illustrating another method for detecting defects of a circuit board according to an embodiment of the present disclosure;
fig. 5 is a schematic flowchart of another method for detecting defects of a circuit board according to an embodiment of the present disclosure;
fig. 6 is a schematic structural diagram of a defect detection apparatus for a circuit board according to an embodiment of the present disclosure;
FIG. 7 illustrates a block diagram of an exemplary computer device suitable for use to implement embodiments of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
The embodiment of the application provides a method for detecting the defects of the circuit board, aiming at the technical problems that in the prior art, the defect detection of the circuit board depends on a pure manual detection method, the detection cost is high, the accuracy of the detection result is low, and the real-time performance is poor.
The method for detecting the defects of the circuit board comprises the steps of obtaining a shot image of the circuit board to be detected, carrying out pixel information difference on corresponding pixels of the shot image and a standard image to obtain a difference value of the corresponding pixels, wherein the standard image is obtained by shooting a reference circuit board without defects, synthesizing pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image, inputting the input image into a trained classification model, and determining the defect information of the circuit board to be detected according to the output of the classification model.
A defect detection method and apparatus of a circuit board according to an embodiment of the present application are described below with reference to the drawings.
Fig. 1 is a schematic flow chart of a method for detecting defects of a circuit board according to an embodiment of the present disclosure.
As shown in fig. 1, the method for detecting defects of a circuit board includes the following steps:
The tested circuit board is a circuit board which needs to be subjected to defect detection.
In the embodiment of the application, when the defect detection is performed on the circuit board, the defect detection can be performed on the circuit board to be detected according to the shot image by obtaining the shot image of the circuit board to be detected. Specifically, when the shot image of the circuit board to be tested is obtained, the shot image of the circuit board to be tested can be obtained through real-time collection by utilizing a high-precision camera of the image collection system and adjusting the angle, light, a filter, a double lens, focusing and other parameters of the camera.
As an example, if the light of a workshop for producing the circuit board to be tested is dark, a clear image of the circuit board to be tested can be acquired by adjusting parameters such as the angle and the light of the high-precision camera.
102, carrying out pixel information difference on corresponding pixels of the shot image and the standard image to obtain a difference value of the corresponding pixels; the standard image is obtained by shooting a reference circuit board without defects.
In the embodiment of the application, the standard image is an image obtained by shooting a reference circuit board without defects by using a high-precision camera. After the shot image of the circuit board to be tested is obtained, difference calculation of pixel information is carried out on each pixel of the shot image and the corresponding pixel in the standard image, and a difference value of the corresponding pixel is obtained.
In the embodiment of the present application, the pixel information of each pixel in the captured image and the standard image includes a depth value and a gray value of each color channel. The pixel information of the shot image and the standard image respectively comprises depth values and gray values corresponding to three color channels of R, G and B.
The depth value refers to a distance between an object imaged by the corresponding pixel and the camera. The gray scale value of each color channel refers to the brightness of each color channel, i.e., the shade of each color. The gray values typically range from 0 to 255, 255 for white and 0 for black.
As a possible implementation manner, a depth value of each pixel in the pixel information of the captured image may be differentially calculated from a depth value of a corresponding pixel in the pixel information of the standard image, so as to obtain a depth differential value of the corresponding pixel.
As another possible implementation manner, the difference between the gray-scale value of the pixel in each color channel in the pixel information of the captured image and the gray-scale value of the corresponding pixel in each color channel in the standard image may be calculated to obtain the gray-scale difference value of the corresponding pixel in each color channel.
As another possible implementation manner, the difference between the gray-scale value of the pixel in each color channel in the pixel information of the captured image and the gray-scale value of the corresponding pixel in each color channel in the standard image may be calculated to obtain the gray-scale difference value of the corresponding pixel in each color channel; and carrying out difference calculation on the depth value of each pixel in the pixel information of the shot image and the depth value of the corresponding pixel in the pixel information of the standard image to obtain the depth difference value of the corresponding pixel.
In the embodiment of the present application, after each pixel of the captured image obtained by the pixel difference calculation is subjected to the pixel information difference with the pixel corresponding to the standard image to obtain the difference value of the corresponding pixel, the pixel information of the captured image and the difference value of the corresponding pixel are synthesized to obtain the input image.
Specifically, the depth difference value of the corresponding pixel obtained by performing pixel information difference calculation on the corresponding pixel of the captured image and the standard image and the gray scale difference value in each color channel may be added to the pixel information of each pixel in the captured image, so as to obtain the pixel information of the corresponding pixel in the input image. At this time, the pixel information of each pixel in the input image includes a depth value of the pixel, a gray value of each color channel, a depth difference value, and a gray difference value in each color channel.
It should be noted that, the tested circuit board may have some defects due to limitations of processes, equipment, raw materials, and the like in the production process, and therefore, in the embodiment of the present application, the defect information of the tested circuit board is used to indicate whether the circuit board has defects and the defect types. The defect types can be defects of less tin in chip welding, more tin in chip welding, continuous tin in chip welding, wrong chip installation and the like. However, in the embodiment of the present application, the defect information of the tested circuit board may be in other situations, and is not limited to the defect information.
In an embodiment of the application, the classification model may be obtained by training a training sample image through a Deep Convolutional Neural Network (DCNN), and the classification model obtained through training has learned image characteristics of input images corresponding to circuit boards with various defects, so that after the input images corresponding to the circuit boards to be tested are input to the classification model, the defect information of the circuit boards to be tested can be determined according to the output of the classification model.
The pixel information of the sample image used for training is added with a difference value obtained by carrying out pixel information difference on the sample image and the corresponding pixel of the standard image.
The defect detection method of the embodiment of the application comprises the steps of obtaining a shot image of a circuit board to be detected, carrying out pixel information difference on corresponding pixels of the shot image and a standard image to obtain a difference value of the corresponding pixels, wherein the standard image is obtained by shooting a reference circuit board without defects, synthesizing the pixel information of the shot image and the difference value of the corresponding pixels to obtain an input image, inputting the synthesized input image into a trained classification model to determine the defect information of the circuit board to be detected according to the output of the classification model, and the classification model learns the image characteristics of the input image corresponding to the circuit board with various defects. Because the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects, the input data of the classification model can be enriched by adding the difference value into the pixel information of the shot image for image characteristic extraction, and the accuracy and the efficiency of the defect detection of the circuit boards to be detected are improved.
In a possible implementation form of the embodiment of the present application, training of a classification model may be performed by synthesizing a difference value obtained by performing corresponding pixel difference on a sample image and a standard image with pixel information of a corresponding pixel in the sample image, and then training with the synthesized sample image, where a specific model training process is shown in fig. 2, and fig. 2 is a flowchart of a classification model training method provided in the embodiment of the present application.
As shown in fig. 2, the model training method may include the steps of:
The sample image is marked with defect information, and the defect information is used for indicating whether the circuit board displayed by the corresponding sample image has defects or not and indicating the defect type.
In the embodiment of the application, each pixel in the pixel information of the acquired sample image and the corresponding pixel in the pixel information of the standard image are subjected to pixel difference to obtain a difference value of the corresponding pixel. The sample image is obtained by shooting the sample circuit board through the high-precision camera.
Specifically, the depth value of each pixel in the pixel information of the sample image and the depth value of the corresponding pixel in the pixel information of the standard image are subjected to difference calculation to obtain a depth difference value of the corresponding pixel. And carrying out difference calculation on the gray value of the pixel in each color channel in the pixel information of the sample image and the gray value of the corresponding pixel in each color channel in the standard image to obtain the gray difference value of the corresponding pixel in each color channel.
Furthermore, a difference value obtained by pixel information difference between corresponding pixels of the sample image and the standard image is synthesized with pixel information of corresponding pixels in the sample image, and the pixel information of each pixel in the synthesized sample image comprises a depth value, a gray value of each color channel, a depth difference value obtained by corresponding pixel difference between the sample image and the standard image, and a gray difference value of each color channel.
And step 202, training by using the synthesized sample image.
As a possible scenario, the classification model in the embodiment of the present application may include a convolutional layer, a pooling layer, and a fully-connected layer. The convolution layer is used for carrying out image feature extraction on pixel information of each pixel in the input classification model image; the pooling layer is used for performing dimension reduction operation on the features extracted from the convolutional layer; and the full connection layer is used for classifying according to the image characteristics after the dimensionality reduction of the pooling layer.
In the embodiment of the application, the defect information of the marked circuit board of the sample image is used as the characteristic of model training, the synthesized sample image is input into the classification model, the classification model is trained, and then the defect information of the circuit board is output.
It should be explained that the defect information of the labeled circuit board of the sample image is used as the feature of the model training, because the defect information can clearly show whether the circuit board shown by the corresponding sample image has defects and the defect types, the defect information is used as the feature to train the classification model, and after the corresponding input image of the tested circuit board is input into the classification model, the defect information of the tested circuit board can be used to determine whether the tested circuit board has defects and the existing defect types.
Specifically, a sample image obtained by synthesizing pixel information is input into a classification model, and a convolution layer of the classification model performs image feature extraction on the pixel information of each pixel in the input sample image according to the labeled defect information of the sample image. Two methods are available for extracting the characteristics of the image, one method is that firstly, a sample image is automatically segmented, object or color areas contained in the sample image are divided, then the image characteristics are extracted according to the areas, and an index is established; another method simply divides the sample image evenly into regular sub-blocks, then extracts features for each sample image sub-block, and builds an index.
Furthermore, the pooling layer performs dimension reduction operation on the image features extracted from the convolutional layer, only main features in the image features are reserved, and the image features subjected to dimension reduction of the pooling layer are classified through the full-connection layer, so that the classification model outputs classified defect information.
And step 203, when the difference between the defect information output by the classification model and the defect information labeled by the sample image is minimized, the training of the classification model is finished.
In the embodiment of the application, when the difference between the defect information output by the classification model and the defect information labeled by the sample image is large, the classification model is further trained in a signal forward propagation and error backward propagation mode until the difference between the defect information output by the classification model and the defect information labeled by the sample image is minimized, and the training of the classification model is finished.
As a possible case, when the difference between the defect information output by the classification model and the defect information labeled by the sample image reaches a preset difference threshold, the training of the classification model is finished.
Furthermore, the trained classification model gradually replaces the classification model which is running on line in a small-flow online mode, so that the purpose of dynamic expansion and generalization of the classification model along with the service is achieved.
As an example, referring to fig. 3, fig. 3 is an exemplary diagram for outputting defect information of a tested circuit board according to a classification model provided in an embodiment of the present application. The classification model in fig. 3 includes convolutional layers, pooling layers, and fully-connected layers. The method comprises the steps of synthesizing a difference value obtained by carrying out pixel information difference on a shot image of a tested circuit board and a corresponding pixel of a standard image with pixel information of the shot image, determining an input image, inputting the input image into a classification model shown in figure 3, carrying out image feature extraction on the pixel information of each pixel in the input image by a convolution layer of the classification model, further carrying out dimension reduction operation on the extracted image features by a pooling layer through an image down-sampling method to obtain main image features of the input image, and finally classifying the dimension-reduced image features through a full connection layer, so that whether the tested circuit board has defects or not and the defect type with the defects are determined according to output defect information.
It should be noted that, for the characteristics of different circuit board production scenarios and data, a deep neural network model composed of different depths, different numbers of neurons, and different convolution pooling organization modes can be designed based on the classification model shown in fig. 3, and then the model is trained by adopting a sample image labeled with defect information and adopting a signal forward propagation and error backward propagation mode to meet the requirements of different service scenarios.
In the embodiment of the application, a differential value obtained by pixel information difference between a sample image and a corresponding pixel of a standard image is synthesized with pixel information of the corresponding pixel in the sample image, the sample image of the synthesized pixel information is trained, and when the difference between defect information output by a classification model and the defect information labeled by the sample image is minimized, the classification model training is finished. Therefore, through training of the classification model, after the input image is input into the trained classification model, whether the circuit board to be detected has defect information or not and the defect type can be accurately determined according to the output of the classification model, and therefore the accuracy of circuit board detection is improved.
In the embodiment of the application, on the basis of the embodiment shown in fig. 1, after the defect information of the circuit board to be tested is determined according to the output of the classification model, whether the circuit board to be tested has defects is further judged according to the defect information of the circuit board to be tested.
As a possible situation, after the circuit board to be tested is determined to have defects according to the defect information of the circuit board to be tested, which is output by the classification model, a first training sample is obtained according to the shot image of the circuit board with defects and the defect information obtained by artificial defect type marking, and then the classification model is trained by adopting the first training sample. Fig. 4 is a schematic flow chart of another method for detecting defects of a circuit board according to an embodiment of the present application.
As shown in fig. 4, the following steps are also included after step 104 of the previous embodiment:
In the embodiment of the application, a difference value obtained by pixel information difference between a shot image of a tested circuit board and a corresponding pixel of a standard image is synthesized with pixel information of the shot image to determine an input image, and then the input image is input into a classification model to determine the defect information of the tested circuit board. And when the circuit board to be tested has defects according to the defect information of the circuit board to be tested, displaying the shot image of the circuit board to be tested on the control interface.
In the embodiment of the application, when the circuit board to be tested has defects according to the defect information of the circuit board to be tested, the circuit board to be tested with the defects can be placed in the set area by controlling the production line.
For example, when the circuit board under test is determined to have defects according to the defect information of the circuit board under test, the mechanical arm can be controlled to take out the circuit board from the production line, and the conveyor belt can be controlled to convey the circuit board to a set position, so that a normal circuit board and an abnormal circuit board can be separated conveniently, and the circuit board with defects can be processed in a centralized manner. The manner of separating the abnormal circuit boards is only an example, and the specific separation manner is determined according to the actual production line.
When the normal circuit board and the abnormal circuit board are separated, different areas can be preset according to the defect types of the circuit boards, so that the circuit boards with different defect types are placed in different areas, and the circuit boards with different defect types can be conveniently processed by adopting different processing modes. The circuit boards with defects can also be placed in the same area for centralized processing. The specific processing manner is determined according to actual conditions, and the method is not limited in the embodiment of the application.
As another possible situation, when it is determined that the tested circuit board has a defect according to the defect information of the tested circuit board, a preset alarm system may be set up to place the tested circuit board having the defect in the set area, or the defect information of the tested circuit board may be stored in the production database to update the production database, so as to train the classification model according to the data in the updated production database.
In the embodiment of the application, when the circuit board to be tested has defects, the defect information obtained by marking the type of the artificial defects of the shot image can be obtained by comparing the difference between the shot image and the standard image after the shot image is displayed on the control interface.
For example, assume that after the circuit board under test is labeled with the artificial defect type, the circuit board under test has defect type a, defect type B, and defect type C. The defect type B of the circuit board can be acquired through the shot image of the circuit board to be detected.
In the embodiment of the application, after a first training sample is generated according to a shot image and defect information, the sample image of the training sample is already marked with the defect information, that is, the sample image is marked with the defect and the defect type of the training sample.
And step 304, training the classification model by using the first training sample.
In the embodiment of the present application, the method for training the classification model by using the first training sample may refer to the classification model training method described in the above embodiment, and details are not repeated here.
In the embodiment of the application, when the defect of the circuit board to be tested is determined by the defect information of the circuit board to be tested, the shot image is displayed on the control interface, the defect information obtained by marking the type of the artificial defect of the shot image is obtained, a first training sample is generated according to the shot image and the defect information, and the classification model is trained by adopting the first training sample. Therefore, the training sample is obtained according to the shot image of the circuit board with the defect and the defect information so as to train the classification model, the classification model can be updated, and the detection precision of the circuit board is improved.
In order to avoid the model training deviation caused by the fact that the circuit board with the defects is over-learned, on the basis of the embodiment of fig. 4, after the fact that the circuit board to be tested does not have the defects is determined according to the defect information of the circuit board to be tested, which is output by the classification model, a second training sample is generated according to the shot image of part of the circuit board and the defect information manually marked in the image, and then the classification model is trained by adopting the first training sample and the second training sample. Fig. 5 is a schematic flow chart of another method for detecting defects of a circuit board according to an embodiment of the present application.
As shown in fig. 5, the following steps are also included after step 104 of the previous embodiment:
In the embodiment of the application, whether the corresponding circuit board has defects or not can be determined according to the defect information of the circuit board to be tested, a part of circuit boards are selected from the determined circuit boards to be tested without defects, and furthermore, the shot images of the selected part of circuit boards are displayed on the control interface.
It should be noted that, because there are many tested circuit boards without defects in the production process, if displaying the shot images of the tested circuit boards without defects on the control interface one by one and manually detecting the shot images would be a huge project, only a part of the circuit boards without defects needs to be selected for displaying, so as to facilitate manual re-detection.
And 402, acquiring defect information of the manual marking of the corresponding shot image for the selected partial circuit board to generate a second training sample for training the classification model.
In the embodiment of the application, part of circuit boards confirmed by manual reinspection are selected from circuit boards without defects, corresponding defect information is marked, namely marks without defects are marked, and then a second training sample is generated so as to train the classification model by adopting the generated second training sample.
It should be noted that, in the process of training the classification model by using the second training sample, reference may also be made to the method for training the classification model in the foregoing embodiment, and details are not described here again.
In the embodiment of the application, part of circuit boards are selected from the circuit boards to be tested without defects, the shot images of the selected part of circuit boards are displayed on the control interface, and the defect information of the corresponding shot images marked manually is acquired for the selected part of circuit boards, so that a second training sample for training the classification model is generated. Therefore, the classification model is trained through the shot image corresponding to the circuit board without the defect and the training sample generated by the defect information, the classification model can be updated in real time, and the detection precision of the circuit board is improved.
In order to implement the above embodiments, the present application further provides a defect detection apparatus for a circuit board.
Fig. 6 is a schematic structural diagram of a defect detection apparatus for a circuit board according to an embodiment of the present application.
As shown in fig. 6, the defect detecting apparatus 100 of the circuit board includes: an acquisition module 110, a difference module 120, a processing module 130, and a detection module 140.
And an obtaining module 110, configured to obtain a captured image of the circuit board.
A difference module 120, configured to perform pixel information difference on the corresponding pixels of the captured image and the standard image to obtain a difference value of the corresponding pixel; the standard image is obtained by shooting a reference circuit board without defects.
And the processing module 130 is configured to synthesize the difference value between the pixel information of each pixel of the captured image and the corresponding pixel to obtain an input image.
A detection module 140, configured to input the input image into the trained classification model, so as to determine defect information of the circuit board to be tested according to the output of the classification model; the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects.
As a possible case, the pixel information of each pixel point in the captured image and the standard image includes a depth value and a gray value of each color channel.
As another possible scenario, the difference module 120 is specifically configured to:
carrying out depth value difference on corresponding pixels in the shot image and the standard image to obtain a depth difference value of the corresponding pixels; and carrying out gray value difference on corresponding pixels in the shot image and the standard image to obtain gray value difference values of the corresponding pixels in each color channel.
As another possible scenario, the processing module 130 is specifically configured to: and taking the depth difference value, the depth value, and the gray difference value and the gray value of each color channel as the pixel information of the corresponding pixel in the input image.
As another possible scenario, the classification model includes a convolutional layer, a pooling layer, and a fully-connected layer;
the convolution layer is used for carrying out image feature extraction on pixel information of each pixel in an input image;
the pooling layer is used for performing dimension reduction operation on the features extracted from the convolutional layer;
and the full connection layer is used for classifying according to the image characteristics after the dimensionality reduction of the pooling layer.
As another possible situation, the classification model is obtained by synthesizing a difference value obtained by pixel information difference between corresponding pixels of the sample image and the standard image with pixel information of corresponding pixels in the sample image and training the synthesized sample image; the sample image is marked with defect information, and the defect information is used for indicating whether the circuit board displayed by the corresponding sample image has defects or not and indicating the defect type;
and when the difference between the defect information output by the classification model and the defect information labeled by the sample image is minimized, finishing the training of the classification model.
As another possible case, the defect detecting apparatus 100 further includes:
the first display module is used for displaying the shot image on the control interface if the circuit board to be tested has defects.
And the second acquisition module is used for acquiring the defect information obtained by marking the type of the artificial defect on the shot image.
And the first generation module is used for generating a first training sample according to the shot image and the defect information.
And the training module is used for training the classification model by adopting the first training sample.
As another possible case, the defect detecting apparatus 100 further includes:
and the second display module is used for selecting part of circuit boards from the circuit boards to be tested without defects and displaying the shot images of the selected part of circuit boards on the control interface.
And the second generation module is used for acquiring the defect information of the corresponding shot image manual marking for the selected partial circuit board so as to generate a second training sample for training the classification model.
As another possible case, the defect detecting apparatus 100 further includes:
and the control module is used for controlling the production line to place the tested circuit board with defects in the set area if the tested circuit board has defects.
It should be noted that the foregoing explanation of the embodiment of the defect detection method is also applicable to the defect detection apparatus of this embodiment, and is not repeated herein.
The defect detection device of the embodiment of the application obtains the shot image of the circuit board to be detected; the method comprises the steps of carrying out pixel information difference on a shot image and pixels corresponding to a standard image to obtain a difference value of the corresponding pixels, wherein the standard image is obtained by shooting a reference circuit board without defects, synthesizing pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image, inputting the input image into a trained classification model to determine defect information of a circuit board to be tested according to the output of the classification model, and the classification model learns the image characteristics of the input image corresponding to the circuit board with various defects. Because the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects, the input data of the classification model can be enriched by adding the difference value into the pixel information of the shot image for image characteristic extraction, and the accuracy and the efficiency of the defect detection of the circuit boards to be detected are improved.
In order to implement the foregoing embodiments, the present application also provides a computer device, including: the circuit board defect detection method comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein when the processor executes the program, the defect detection method of the circuit board is realized.
In order to implement the above embodiments, the present application also proposes a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of defect detection of a circuit board as described in the above embodiments.
FIG. 7 illustrates a block diagram of an exemplary computer device suitable for use to implement embodiments of the present application. The computer device 12 shown in fig. 7 is only an example, and should not bring any limitation to the function and the scope of use of the embodiments of the present application.
As shown in FIG. 7, computer device 12 is in the form of a general purpose computing device. The components of computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that couples various system components including the system memory 28 and the processing unit 16.
Computer device 12 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.
A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 42 generally perform the functions and/or methodologies of the embodiments described herein.
Computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), with one or more devices that enable a user to interact with the computer system/server 12, and/or with any devices (e.g., network card, modem, etc.) that enable the computer device 12 to communicate with one or more other computing devices. Such communication may be through an input/output (I/O) interface 22. Moreover, computer device 12 may also communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public Network such as the Internet) via Network adapter 20. As shown in FIG. 7, the network adapter 20 communicates with the other modules of the computer device 12 via the bus 18. It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
The processing unit 16 executes various functional applications and data processing by running a program stored in the system memory 28, for example, implementing the defect detection method of the circuit board mentioned in the foregoing embodiments.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (14)
1. A method for detecting defects of a circuit board, the method comprising the steps of:
acquiring a shot image of a circuit board to be tested;
carrying out pixel information difference on the corresponding pixels of the shot image and the standard image to obtain a difference value of the corresponding pixels; the standard image is obtained by shooting a reference circuit board without defects; the pixel information of each pixel in the shot image and the standard image comprises a depth value and a gray value of each color channel; carrying out depth value difference on the corresponding pixels in the shot image and the standard image to obtain depth difference values of the corresponding pixels; carrying out gray value difference on the corresponding pixels in the shot image and the standard image to obtain gray value difference values of the corresponding pixels in the color channels;
synthesizing the pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image; the depth difference value, the depth value, and the gray difference value and the gray value of each color channel are used as pixel information of corresponding pixels in the input image;
inputting the input image into a trained classification model to determine the defect information of the circuit board to be tested according to the output of the classification model; and the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects.
2. The defect detection method of claim 1, wherein the classification model comprises a convolutional layer, a pooling layer, and a fully-connected layer;
the convolutional layer is used for extracting image characteristics of pixel information of each pixel in the input image;
the pooling layer is used for performing dimension reduction operation on the features extracted from the convolutional layer;
and the full connection layer is used for classifying according to the image characteristics after the dimensionality reduction of the pooling layer.
3. The defect detection method of claim 1, wherein the classification model is obtained by synthesizing a difference value obtained by pixel information difference between a sample image and a corresponding pixel of the standard image with pixel information of a corresponding pixel in the sample image, and training the synthesized sample image; the sample images are marked with defect information, and the defect information is used for indicating whether the circuit board displayed by the corresponding sample image has defects or not and indicating the defect types;
and when the difference between the defect information output by the classification model and the defect information labeled by the sample image is minimized, finishing the training of the classification model.
4. The method of claim 1, wherein the inputting the input image into a trained classification model to determine the defect information of the circuit board under test according to the output of the classification model, further comprises:
if the circuit board to be tested has defects, displaying the shot image on a control interface;
acquiring defect information obtained by marking the type of the artificial defect of the shot image;
generating a first training sample according to the shot image and the defect information;
and training the classification model by adopting the first training sample.
5. The method of claim 1, wherein the inputting the input image into a trained classification model to determine the defect information of the circuit board under test according to the output of the classification model, further comprises:
selecting part of circuit boards from the circuit boards to be tested without defects, and displaying the shot images of the selected part of circuit boards on a control interface;
and acquiring the defect information of the corresponding shot image manual marking for the selected partial circuit boards so as to generate a second training sample for training the classification model.
6. The method of claim 1, wherein the inputting the input image into a trained classification model to determine the defect information of the circuit board under test according to the output of the classification model, further comprises:
if the tested circuit board has defects, controlling a production line so as to place the tested circuit board with defects in a set area.
7. A defect inspection apparatus for a circuit board, the apparatus comprising:
the acquisition module is used for acquiring a shot image of the circuit board to be detected;
the difference module is used for carrying out pixel information difference on the corresponding pixels of the shot image and the standard image to obtain a difference value of the corresponding pixels; the standard image is obtained by shooting a reference circuit board without defects; the pixel information of each pixel in the shot image and the standard image comprises a depth value and a gray value of each color channel; the difference module is specifically configured to: carrying out depth value difference on the corresponding pixels in the shot image and the standard image to obtain the depth difference value of the corresponding pixels; carrying out gray value difference on the corresponding pixels in the shot image and the standard image to obtain gray value difference values of the corresponding pixels in the color channels;
the processing module is used for synthesizing the pixel information of each pixel of the shot image and the difference value of the corresponding pixel to obtain an input image; the processing module is specifically configured to: taking the depth difference value, the depth value, and the gray difference value and the gray value of each color channel as pixel information of corresponding pixels in the input image;
the detection module is used for inputting the input image into the trained classification model so as to determine the defect information of the circuit board to be tested according to the output of the classification model; and the classification model learns the image characteristics of the input images corresponding to the circuit boards with various defects.
8. The defect detection apparatus of claim 7, wherein the classification model comprises a convolutional layer, a pooling layer, and a fully-connected layer;
the convolutional layer is used for extracting image characteristics of pixel information of each pixel in the input image;
the pooling layer is used for performing dimension reduction operation on the features extracted from the convolutional layer;
and the full connection layer is used for classifying according to the image characteristics after the dimensionality reduction of the pooling layer.
9. The defect detection apparatus according to claim 7, wherein the classification model is obtained by synthesizing a difference value obtained by differentiating pixel information between a sample image and a corresponding pixel of the standard image with pixel information of a corresponding pixel in the sample image, and training the synthesized sample image; the sample images are marked with defect information, and the defect information is used for indicating whether the circuit board displayed by the corresponding sample image has defects or not and indicating the defect types;
and when the difference between the defect information output by the classification model and the defect information labeled by the sample image is minimized, finishing the training of the classification model.
10. The defect detection apparatus of claim 7, further comprising:
the first display module is used for displaying the shot image on a control interface if the circuit board to be tested has defects;
the second acquisition module is used for acquiring defect information obtained by marking the type of the artificial defect on the shot image;
the first generation module is used for generating a first training sample according to the shot image and the defect information;
and the training module is used for training the classification model by adopting the first training sample.
11. The defect detection apparatus of claim 7, further comprising:
the second display module is used for selecting part of circuit boards from the circuit boards to be tested without defects and displaying the shot images of the selected part of circuit boards on the control interface;
and the second generation module is used for acquiring the defect information of the corresponding shot image manual marking for the selected partial circuit board so as to generate a second training sample for training the classification model.
12. The defect detection apparatus of claim 7, further comprising:
and the control module is used for controlling a production line to place the tested circuit board with defects in a set area if the tested circuit board has defects.
13. Computer device, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing a method for defect detection of a circuit board according to any of claims 1-6 when executing the program.
14. A non-transitory computer-readable storage medium on which a computer program is stored, the program, when executed by a processor, implementing a method for defect detection of a circuit board according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910133359.9A CN109871895B (en) | 2019-02-22 | 2019-02-22 | Method and device for detecting defects of circuit board |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910133359.9A CN109871895B (en) | 2019-02-22 | 2019-02-22 | Method and device for detecting defects of circuit board |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109871895A CN109871895A (en) | 2019-06-11 |
| CN109871895B true CN109871895B (en) | 2021-03-16 |
Family
ID=66919116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910133359.9A Active CN109871895B (en) | 2019-02-22 | 2019-02-22 | Method and device for detecting defects of circuit board |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109871895B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110441315B (en) * | 2019-08-02 | 2022-08-05 | 英特尔产品（成都）有限公司 | Electronic component testing apparatus and method |
| JP7536517B2 (en) * | 2019-10-08 | 2024-08-20 | キヤノン株式会社 | Method for generating training data, trained learning model, and system |
| CN111078912B (en) * | 2019-12-18 | 2024-02-20 | 国网上海市电力公司 | Power equipment image data warehouse and power equipment defect detection method |
| CN111223077A (en) * | 2019-12-30 | 2020-06-02 | 浙江力创自动化科技有限公司 | Test method and test system of circuit board |
| CN111696092B (en) * | 2020-06-11 | 2023-08-25 | 深圳市华汉伟业科技有限公司 | Defect detection method and system based on feature comparison and storage medium |
| CN111986178B (en) * | 2020-08-21 | 2024-06-18 | 北京百度网讯科技有限公司 | Product defect detection method, device, electronic equipment and storage medium |
| CN112026350A (en) * | 2020-09-17 | 2020-12-04 | 广州夕千科技有限公司 | Printing machine for processing electronic components |
| CN112396599B (en) * | 2020-12-04 | 2021-08-31 | 高视科技(苏州)有限公司 | Visual inspection method for transparent packaged IC (integrated circuit) defects |
| CN114723651A (en) * | 2020-12-22 | 2022-07-08 | 东方晶源微电子科技（北京）有限公司 | Defect detection model training method, defect detection method, device and equipment |
| CN112834518A (en) * | 2021-01-06 | 2021-05-25 | 优刻得科技股份有限公司 | Particle defect detection method, system, device and medium |
| CN113111828B (en) * | 2021-04-23 | 2022-03-18 | 中国科学院宁波材料技术与工程研究所 | Three-dimensional defect detection method and system for bearing |
| CN113222913B (en) * | 2021-04-28 | 2024-04-12 | 南京南瑞继保电气有限公司 | Circuit board defect detection positioning method, device and storage medium |
| CN114581362B (en) * | 2021-07-22 | 2023-11-07 | 正泰集团研发中心(上海)有限公司 | Photovoltaic module defect detection method and device, electronic equipment and readable storage medium |
| CN113466261B (en) * | 2021-07-26 | 2023-04-07 | 鸿安（福建）机械有限公司 | PCB board automatic checkout device |
| CN114152627B (en) * | 2022-02-09 | 2022-05-03 | 季华实验室 | Chip circuit defect detection method and device, electronic equipment and storage medium |
| CN114708266A (en) * | 2022-06-07 | 2022-07-05 | 青岛通产智能科技股份有限公司 | Tool, method and device for detecting card defects and medium |
| TWI803393B (en) * | 2022-07-19 | 2023-05-21 | 神通資訊科技股份有限公司 | Defect detection system based on unsupervised learning and method thereof |
| CN115239734A (en) * | 2022-09-23 | 2022-10-25 | 成都数之联科技股份有限公司 | Model training method, device, storage medium, equipment and computer program product |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106204618A (en) * | 2016-07-20 | 2016-12-07 | 南京文采科技有限责任公司 | Product surface of package defects detection based on machine vision and sorting technique |
| CN106897994A (en) * | 2017-01-20 | 2017-06-27 | 北京京仪仪器仪表研究总院有限公司 | A kind of pcb board defect detecting system and method based on layered image |
| CN107966448A (en) * | 2017-11-17 | 2018-04-27 | 福建工程学院 | A kind of 2 dimension detection methods for PCB paste solder printing quality |
| CN108982544A (en) * | 2018-06-20 | 2018-12-11 | 青岛联合创智科技有限公司 | A kind of printed circuit board flaw component detection method |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001304842A (en) * | 2000-04-25 | 2001-10-31 | Hitachi Ltd | Method and device of pattern inspection and treatment method of substrate |
| US6642726B2 (en) * | 2001-06-29 | 2003-11-04 | Kla-Tencor Corporation | Apparatus and methods for reliable and efficient detection of voltage contrast defects |
| CN101339607B (en) * | 2008-08-15 | 2012-08-01 | 北京中星微电子有限公司 | Human face recognition method and system, human face recognition model training method and system |
| CN104458755B (en) * | 2014-11-26 | 2017-02-22 | 吴晓军 | Multi-type material surface defect detection method based on machine vision |
| CN106570511B (en) * | 2016-11-02 | 2021-04-09 | 上海大学 | Keyboard character defect online detection system and method based on computer vision |
| CN109300151B (en) * | 2018-07-02 | 2021-02-12 | 浙江商汤科技开发有限公司 | Image processing method and device and electronic equipment |
| CN109345524A (en) * | 2018-09-26 | 2019-02-15 | 深圳市鑫汇达机械设计有限公司 | A kind of bearing open defect detection system of view-based access control model |
-
2019
- 2019-02-22 CN CN201910133359.9A patent/CN109871895B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106204618A (en) * | 2016-07-20 | 2016-12-07 | 南京文采科技有限责任公司 | Product surface of package defects detection based on machine vision and sorting technique |
| CN106897994A (en) * | 2017-01-20 | 2017-06-27 | 北京京仪仪器仪表研究总院有限公司 | A kind of pcb board defect detecting system and method based on layered image |
| CN107966448A (en) * | 2017-11-17 | 2018-04-27 | 福建工程学院 | A kind of 2 dimension detection methods for PCB paste solder printing quality |
| CN108982544A (en) * | 2018-06-20 | 2018-12-11 | 青岛联合创智科技有限公司 | A kind of printed circuit board flaw component detection method |
Non-Patent Citations (3)
| Title |
|---|
| Detecting PCB component placement defects by genetic programming;Feng Xie et al.;《2013 IEEE Congress on Evolutionary Computation》;20131231;全文 * |
| ONLINE PCB DEFECT DETECTOR ON A NEW PCB DEFECT DATASET;Sanli Tang et al.;《arXiv》;20190217;全文 * |
| 基于形态学的PCB缺陷快速检测技术;王栋等;《计算机科学》;20160615;第43卷(第6A期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109871895A (en) | 2019-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109871895B (en) | Method and device for detecting defects of circuit board | |
| US20230419472A1 (en) | Defect detection method, device and system | |
| CN109949305B (en) | Product surface defect detection method and device and computer equipment | |
| CN112348787B (en) | Training method of object defect detection model, object defect detection method and device | |
| CN111699499B (en) | Inspection, image recognition, recognizer generation system, and learning data generation device | |
| CN111833306A (en) | Defect detection method and model training method for defect detection | |
| JP2017049974A (en) | Discriminator generator, quality determine method, and program | |
| CN110992349A (en) | Underground pipeline abnormity automatic positioning and identification method based on deep learning | |
| CN111598913B (en) | Image segmentation method and system based on robot vision | |
| CN111583114B (en) | Automatic measuring device and measuring method for pipeline threads | |
| CN112070746A (en) | Steel strip defect detection method and device | |
| CN112070747A (en) | LED lamp bead defect detection method and device | |
| CN113780484B (en) | Industrial product defect detection method and device | |
| JP2020181532A (en) | Image determination device and image determination method | |
| CN110738644A (en) | automobile coating surface defect detection method and system based on deep learning | |
| CN116830157A (en) | Image enhancement techniques for automated visual inspection | |
| CN113516652A (en) | Battery surface defect and adhesive detection method, device, medium and electronic equipment | |
| CN115937147B (en) | Defect detection parameter determining method, device, equipment and storage medium | |
| CN112102253A (en) | Non-woven fabric surface defect automatic detection method and system based on machine vision | |
| CN113610054B (en) | Underwater structure disease depth detection method, system, device and storage medium | |
| CN113554645A (en) | Industrial anomaly detection method and device based on WGAN | |
| CN115375608A (en) | Detection method and device, detection equipment and storage medium | |
| CN112070748A (en) | Metal oil pipe defect detection method and device | |
| CN114419078B (en) | Surface defect region segmentation method and device based on convolutional neural network | |
| CN112001388B (en) | Method for detecting circular target in PCB based on YOLOv3 improved model |
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 |