CN113804704A - Circuit board detection method, visual detection equipment and device with storage function - Google Patents
Circuit board detection method, visual detection equipment and device with storage function Download PDFInfo
- Publication number
- CN113804704A CN113804704A CN202010531361.4A CN202010531361A CN113804704A CN 113804704 A CN113804704 A CN 113804704A CN 202010531361 A CN202010531361 A CN 202010531361A CN 113804704 A CN113804704 A CN 113804704A
- Authority
- CN
- China
- Prior art keywords
- tested
- circuit board
- component
- recognition model
- image
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/956—Inspecting patterns on the surface of objects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/956—Inspecting patterns on the surface of objects
- G01N2021/95638—Inspecting patterns on the surface of objects for PCB's
Abstract
The application discloses a circuit board detection method, a visual detection device and a device with a storage function, wherein the circuit board detection method comprises the following steps: acquiring an image to be detected, wherein the image to be detected comprises a circuit board to be detected and at least one component to be detected, which is arranged on the circuit board to be detected; inputting the image to be tested into a pre-trained recognition model to obtain a prediction result of the installation direction of the component to be tested on the circuit board to be tested, wherein the recognition model can predict the installation directions of different types of components to be tested; and determining whether the component to be tested is correctly mounted on the circuit board to be tested according to the comparison result of the mounting direction predicted by the recognition model and the preset standard mounting direction. The circuit board detection method provided by the application can judge whether the installation direction of the components on the circuit board is correct.
Description
Technical Field
The present disclosure relates to the field of circuit board technologies, and in particular, to a circuit board detection method, a visual inspection apparatus, and a device with a storage function.
Background
In recent years, as electronic technology has been developed, circuit boards have been rapidly developed as important components of electronic technology, wherein whether components are mounted on the circuit boards correctly or not is one of important factors determining the quality of the circuit boards.
In the process of installing components on a circuit board, due to the diversity of circuit board types, the difference of incoming materials of different components and the error of operators, the phenomenon that the installation direction of the components on the circuit board is wrong can be caused, so that the defective rate of the circuit board is increased, and serious influence is brought to enterprises and factories.
Disclosure of Invention
The technical problem mainly solved by the application is to provide a circuit board detection method, a visual detection device and a device with a storage function, and whether the installation direction of components on a circuit board is correct or not can be judged.
In order to solve the technical problem, the application adopts a technical scheme that: provided is a circuit board detection method, comprising the following steps: acquiring an image to be detected, wherein the image to be detected comprises a circuit board to be detected and at least one component to be detected, which is arranged on the circuit board to be detected; inputting the image to be tested into a pre-trained recognition model to obtain a prediction result of the installation direction of the component to be tested on the circuit board to be tested, wherein the recognition model can predict the installation directions of different types of components to be tested; and determining whether the component to be tested is correctly mounted on the circuit board to be tested according to the comparison result of the mounting direction predicted by the recognition model and the preset standard mounting direction.
In order to solve the above technical problem, another technical solution adopted by the present application is: there is provided a visual inspection apparatus comprising a processor, a memory and a communication circuit, the memory having stored therein program data, the processor being coupled to the memory and the communication circuit, respectively, the processor implementing the steps of the method by executing the program data in the memory.
In order to solve the above technical problem, another technical solution adopted by the present application is: there is provided a device having a storage function, the device having a storage function storing program data executable by a processor to implement the steps in the above method.
The beneficial effect of this application is: this application utilizes the recognition model that trains in advance to treat in the image of awaiting measuring each components and parts installation direction on the circuit board that awaits measuring to predict, and the installation direction that will predict and the standard installation direction of predetermineeing carry out the comparison, thereby can judge on the one hand whether the installation direction of components and parts that await measuring on the circuit board that awaits measuring is correct, on the other hand because the recognition model that trains in advance can predict the installation direction of the components and parts that await measuring of different grade type, consequently only need use in whole testing process one recognition model can, thereby also can reduce cost when improving detection efficiency.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts. Wherein:
FIG. 1 is a schematic flow chart diagram illustrating an embodiment of a circuit board inspection method according to the present application;
FIG. 2 is a schematic flow chart diagram illustrating another embodiment of a circuit board inspection method according to the present application;
FIG. 3 is a schematic flow chart illustrating a variation of an image to be measured according to an embodiment;
FIG. 4 is a schematic partial flow chart diagram of another embodiment of a circuit board inspection method according to the present application;
FIG. 5 is a schematic flow chart for training a recognition model according to an embodiment;
FIG. 6 is a schematic flow chart of training a recognition model according to another embodiment;
FIG. 7 is a schematic diagram of a configuration file structure;
FIG. 8 is a schematic structural diagram of an embodiment of the vision inspection apparatus of the present application;
fig. 9 is a schematic structural diagram of an embodiment of the device with a storage function according to the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Referring to fig. 1, fig. 1 is a schematic flow chart diagram of an embodiment of a circuit board detection method according to the present application, where the detection method includes:
s110: and acquiring an image to be detected, wherein the image to be detected comprises a circuit board to be detected and at least one component to be detected, which is arranged on the circuit board to be detected.
In an application scenario, the vision inspection apparatus obtains an image to be inspected by shooting a circuit board to be inspected, for example, a camera is installed on the vision inspection apparatus, the circuit board to be inspected is shot by the camera, or in order to improve the definition of the image to be inspected, a vision inspection system composed of hardware such as an industrial camera, a lens, a coaxial light source and a photoelectric sensor is installed on the vision inspection apparatus, and the circuit board to be inspected is shot by the vision inspection system to obtain the image to be inspected. In another application scenario, the visual inspection device does not shoot the circuit board to be detected, but directly receives the image to be detected sent by other devices.
The circuit board to be tested in the image to be tested may be the whole circuit board to be tested or may be a part of the circuit board to be tested, for example, when only the component to be tested in the local area on the circuit board to be tested needs to be correctly mounted, the obtained image to be tested may only include the local area corresponding to the circuit board to be tested.
At least one component to be tested, which is arranged on the circuit board to be tested, can be the same type of component or different types of components, and the installation directions of the components to be tested, which are arranged on the circuit board to be tested, can be different.
S120: and inputting the image to be tested into a pre-trained recognition model to obtain a prediction result of the installation direction of the component to be tested on the circuit board to be tested, wherein the recognition model can predict the installation directions of different types of components to be tested.
The recognition model is trained in advance, and can predict the installation directions of different types of components to be tested, specifically, after the image to be tested obtained in step S110 is input into the recognition model, the recognition model predicts the installation directions of the components to be tested on the circuit board to be tested in the image to be tested, and then the recognition model outputs the installation directions of the components to be tested on the circuit board to be tested, wherein the installation directions predicted by the recognition model are the actual installation directions of the components to be tested on the circuit board to be tested.
S130: and determining whether the component to be tested is correctly mounted on the circuit board to be tested according to the comparison result of the mounting direction predicted by the recognition model and the preset standard mounting direction.
The standard installation direction of each component to be tested on the circuit board to be tested is pre-stored in the visual inspection equipment. After the installation direction of each component to be tested on the circuit board to be tested is predicted by the recognition model, the respective installation direction of the component to be tested is compared with the corresponding standard installation direction, and then whether the component to be tested is installed on the circuit board to be tested correctly is judged according to the comparison result.
In an application scene, in response to the fact that the installation direction predicted by the recognition model is consistent with the preset standard installation direction, whether the component to be tested is installed on the circuit board to be tested correctly is judged, otherwise, whether the component to be tested is installed on the circuit board to be tested incorrectly is judged, namely, the component to be tested is determined to be installed on the circuit board to be tested correctly according to the result that the installation direction predicted by the recognition model is consistent with the preset standard installation direction; and determining that the component to be tested is incorrectly installed on the circuit board to be tested according to the result that the installation direction predicted by the recognition model is inconsistent with the preset standard installation direction.
It can be seen from the above contents that, in the present embodiment, the pre-trained recognition model is used to predict the installation direction of each component to be tested on the circuit board to be tested in the image to be tested, and the predicted installation direction is compared with the preset standard installation direction, so that on one hand, whether the installation direction of the component to be tested on the circuit board to be tested is correct can be determined, and on the other hand, because the pre-trained recognition model can predict the installation directions of the components to be tested of different types, only one recognition model needs to be used in the whole processing process, different recognition models do not need to be trained for different circuit boards, and thus the cost can be reduced.
In this embodiment, step S120 specifically includes: and extracting the outline of the component to be detected by the recognition model, and predicting the installation direction of the component to be detected according to the extracted outline.
Specifically, the surface shape of each type of component is not a regular shape, and the direction in which the component has a gap is generally defined as the direction of the component. It can be understood that the types of the components are different, the surface shapes of the components are also different, and the outlines of the components are also different, so that after the identification model receives the image to be detected, the outlines of the components to be detected are firstly extracted, then the installation direction of the components to be detected can be predicted according to the extracted outlines, and in an application scene, the direction with the gap in the extracted outlines is identified as the installation direction of the components to be detected on the circuit board to be detected.
The identification model directly determines the installation direction of the component to be detected according to the extracted outline, and the type of the component to be detected does not influence the prediction process and the prediction result of the component to be detected, so that the universality of the identification model can be realized, namely, the identification model can predict the installation directions of different types of components to be detected.
Referring to fig. 2, fig. 2 is a schematic flow chart of another embodiment of the circuit board inspection method of the present application. The detection method comprises the following steps:
s210: and acquiring an image to be detected, wherein the image to be detected comprises a circuit board to be detected and at least one component to be detected, which is arranged on the circuit board to be detected.
S220: and carrying out segmentation processing on the image to be detected so as to obtain a plurality of sub-images to be detected which respectively comprise a single component to be detected.
S230: and respectively inputting the sub-images to be tested into a pre-trained recognition model to obtain a prediction result of the installation direction of the component to be tested on the circuit board to be tested.
S240: and determining whether the component to be tested is correctly mounted on the circuit board to be tested according to the comparison result of the mounting direction predicted by the recognition model and the preset standard mounting direction.
Step S210 and step S240 are the same as step S110 and step S130 in the above embodiments, respectively, and refer to the above embodiments for details, which are not described herein again.
With reference to fig. 3, different from the foregoing embodiment, in this embodiment, the direction of the component 111 to be tested in the image 100 to be tested is not directly predicted by using the recognition model, but the image 100 to be tested is first divided into a plurality of sub-images 110 to be tested, each sub-image 110 to be tested includes a single component 111 to be tested, then the recognition model is used to predict the installation direction of the component 111 to be tested in the sub-images 110 to be tested, and finally, the result of comparing the installation direction predicted by the recognition model with the preset standard installation direction is used to determine whether the component 111 to be tested is correctly installed on the circuit board to be tested.
In the embodiment, the recognition model only predicts the sub-image 110 to be tested including the single component to be tested 111 in each prediction, that is, only predicts the direction of the single component to be tested 111 in each prediction, compared with the above embodiment in which the recognition model predicts the installation direction of at least one component to be tested 111 at the same time, the embodiment can improve the accuracy of the prediction result and avoid the phenomenon of prediction error due to too much single-processing data of the recognition model.
In an application scenario, when the image 100 to be measured is segmented, a high-precision positioning vision technology is used for segmentation.
Specifically, during the segmentation, the component 111 to be measured in the image 100 to be measured is positioned with high precision, and then is segmented, for example, a configuration file corresponding to the circuit board to be measured is generated in advance, the position of each component 111 to be measured on the circuit board to be measured (for example, the coordinate position of the center of the component 111 to be measured) is stored in the configuration file, and then the component 111 to be measured is positioned with high precision according to the configuration file, and then is segmented.
Referring to fig. 4, fig. 4 is a schematic partial flow chart of another embodiment of the circuit board inspection method according to the present application, in the embodiment, before the step of obtaining the image to be inspected, the method further includes:
s310: a plurality of sample images are acquired, wherein the sample images include a sample circuit board and sample components mounted on the sample circuit board.
S320: and obtaining the marking information of the sample component, wherein the marking information comprises the outline and the installation direction of the sample component.
S330: and taking the sample image as input, and taking the annotation information as a truth label to train the recognition model.
The method for acquiring the sample image is the same as the method for acquiring the image to be measured in the above embodiment, which is described in detail in the above embodiment.
The sample circuit board and the circuit board to be tested can be the same in type or different in type, and the type of the sample component and the type of the component to be tested can be the same or different.
With reference to fig. 5, after obtaining a plurality of sample images 200, manually framing the outline of the sample component 211 in each sample image 200 (the thick line in the figure is the outline of the sample component 211) in a manual manner, and labeling the installation directions corresponding to the sample components 211, and then acquiring labeling information of the outline and the installation directions of the sample components 211 by the visual inspection equipment.
When generating the recognition model, the recognition model is obtained by training using the sample image 200 as an input and using the labeling information of the outline and the mounting direction of each sample component 211 in the sample image 200 as a true label.
Specifically, the contour labeling information of the sample components 211 included in each of the plurality of sample images 200 and the mounting direction labeling information corresponding to the sample components 211 are used as training sets, the training sets are trained by using an algorithm to obtain recognition models, and in an application scenario, the training sets are trained by using a segmentation algorithm to obtain recognition models.
It should be noted that the method for training the training set may be deep learning based on a neural network, or machine learning not based on a neural network, and is not limited herein.
In an application scenario, before step S330, the method further includes: and respectively carrying out segmentation processing on the plurality of sample images to obtain a plurality of sub-sample images respectively containing a single sample component. At this time, step S330 specifically includes: and taking the sub-sample image as input, and taking the annotation information as a truth label to train the recognition model.
Specifically, with reference to fig. 6, after the sample image 200 is obtained, the sample image 200 is firstly divided to obtain a plurality of sub-sample images 210 respectively including a single sample component 211, then the user manually frames the outline of the sample component 211 in each sub-sample image 210, and labels the installation direction corresponding to each sample component 211, and then the visual inspection equipment obtains the outline of the sample component 211 and the labeling information of the installation direction.
When the recognition model is generated, the plurality of sub-sample images 210 are used as input, and training is performed with the labeling information of the outline and the mounting direction of the sample component 211 in the plurality of sub-sample images 210 as a true label to obtain the recognition model.
In the application scenario, after the sample image 200 is divided, the divided sub-sample image 210 is used as an input to generate the recognition model, so that the complexity of data processing can be reduced, the difficulty of generating the recognition model is further reduced, and the accuracy of the trained recognition model is ensured.
By the aid of the training method for the recognition model in the two embodiments, the trained recognition model can predict the installation directions of all types of components to be detected, the universality of the recognition model is realized, the recognition model does not need to be retrained no matter how the components to be detected on the circuit board to be detected are updated, the workload of development can be reduced, the difficulty of the whole detection method is reduced, an independent model does not need to be established for each type of components to be detected, and the calculation amount and the cost of a processor can be reduced.
In any one of the above embodiments, the standard mounting direction of the component to be tested on the circuit board to be tested is stored in a preset configuration file; the step of judging whether the component to be tested is correctly installed on the circuit board to be tested according to the comparison result of the installation direction predicted by the recognition model and the preset standard installation direction comprises the following steps: the installation direction predicted by the recognition model is compared with the corresponding standard installation direction in the configuration file.
Specifically, the mounting direction of each device to be tested on each circuit board to be tested is preset and stored to form a configuration file, wherein the stored mounting direction is the standard mounting direction of the device to be tested on the circuit board to be tested. After the image to be tested is obtained, a configuration file corresponding to the model of the circuit board to be tested is determined according to the model of the circuit board to be tested, then the installation direction of the component to be tested on the circuit board to be tested is predicted by using the recognition model, the installation direction of the component to be tested in the configuration file is read, the installation direction of the component to be tested predicted by the recognition model is compared with the standard installation direction corresponding to the component to be tested, and whether the component to be tested is installed on the circuit board to be tested correctly is finally determined.
Meanwhile, in order to determine the standard installation direction of the component to be tested on the circuit board to be tested in the configuration file, before comparing the installation direction predicted by the recognition model with the corresponding standard installation direction in the configuration file, the method further comprises the following steps: and searching the corresponding standard installation direction from the configuration file according to the component name pre-allocated to the component to be tested.
Specifically, as shown in fig. 7, in addition to the standard mounting direction of the component to be tested (for example, "direction left" indicates that the standard mounting direction of the component to be tested is left), a component name corresponding to the component to be tested is also stored in the configuration file (for example, "RY 2" indicates that the component name of the component to be tested is RY2), so that the corresponding standard mounting direction is subsequently searched in the configuration file according to the component name of each component to be tested in the image to be tested.
In other embodiments, the standard mounting direction corresponding to the position of each component to be tested in the image to be tested may be searched in the configuration file.
Specifically, at this time, the configuration file stores the positions of the components to be tested on the circuit board to be tested, in addition to the standard mounting directions of the components to be tested, and then after the images to be tested are obtained, the positions of the components to be tested on the circuit board to be tested are determined by using technologies such as image recognition, and then the corresponding standard mounting directions are searched in the configuration file according to the positions.
In any one of the above embodiments, in order to prompt an operator in time when it is determined that the component to be tested is not correctly mounted on the circuit board to be tested, the circuit board detection method further includes: when the component to be tested is determined to be incorrectly installed on the circuit board to be tested, prompt information is sent, and the prompt information can be voice prompt information, light prompt information or a combination of the voice prompt information and the light prompt information, which is not limited herein.
Meanwhile, when it is determined that the components to be tested are incorrectly installed on the circuit board to be tested, in order to enable an operator to determine which component to be tested is incorrectly installed, the sent prompt information can also carry identification information of the incorrectly-installed component to be tested, and the identification information can be the position of the component to be tested on the circuit board to be tested or the name of the component assigned to the component to be tested.
Of course, when it is determined that the component to be tested is correctly mounted on the circuit board to be tested, prompt information can be sent, and only the prompt information is distinguished from the prompt information.
Referring to fig. 8, fig. 8 is a schematic structural diagram of an embodiment of the visual inspection apparatus of the present application. The visual inspection apparatus 700 includes a processor 710, a memory 720 and a communication circuit 730, wherein the processor 710 is coupled to the memory 720 and the communication circuit 730 respectively, the memory 720 stores program data, and the processor 710 implements any of the steps of the method by executing the program data in the memory 720.
In the actual operation process, the visual inspection apparatus 700 may perform inspection on each circuit board to be inspected in the operation process, perform spot inspection on the circuit boards to be inspected at a certain time interval, or perform inspection on a specific circuit board to be inspected after receiving an instruction sent by a user.
Referring to fig. 9, fig. 9 is a device with a storage function according to the present application, the device with a storage function 800 stores program data 810, and the program data 810 can be executed by a processor to implement the steps in any of the methods, where the detailed methods can refer to the above embodiments and are not described herein again.
The device 800 with a storage function may be a server, a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
In summary, the circuit board detection method in the application can automatically judge whether the installation of the component to be detected on the circuit board to be detected is correct, the method is simple, and the purposes of reducing cost and improving the popularization can be achieved.
The above description is only for the purpose of illustrating embodiments of the present application and is not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application or are directly or indirectly applied to other related technical fields, are also included in the scope of the present application.
Claims (10)
1. A circuit board detection method is characterized by comprising the following steps:
acquiring an image to be detected, wherein the image to be detected comprises a circuit board to be detected and at least one component to be detected, which is arranged on the circuit board to be detected;
inputting the image to be tested into a pre-trained recognition model to obtain a prediction result of the installation direction of the component to be tested on the circuit board to be tested, wherein the recognition model can predict the installation directions of different types of components to be tested;
and determining whether the component to be tested is correctly mounted on the circuit board to be tested according to the comparison result of the mounting direction predicted by the recognition model and the preset standard mounting direction.
2. The inspection method according to claim 1, wherein the step of inputting the image to be inspected to a pre-trained recognition model to obtain a result of predicting the mounting direction of the component on the circuit board to be inspected includes:
and extracting the outline of the component to be detected by the identification model, and predicting the installation direction of the component to be detected according to the extracted outline.
3. The inspection method according to claim 1, wherein before the step of inputting the image to be inspected to a pre-trained recognition model to obtain a result of predicting the mounting direction of the component to be inspected on the circuit board to be inspected, the method further comprises:
the image to be detected is subjected to segmentation processing to obtain a plurality of sub-images to be detected which respectively comprise a single component to be detected;
the step of inputting the image to be tested into a pre-trained recognition model to obtain a prediction result of the installation direction of the component to be tested on the circuit board to be tested comprises the following steps:
and respectively inputting the sub-images to be tested into the pre-trained recognition model to obtain a prediction result of the installation direction of the component to be tested on the circuit board to be tested.
4. The inspection method of claim 1, wherein the step of obtaining the image to be inspected is preceded by the step of:
acquiring a plurality of sample images, wherein the sample images comprise a sample circuit board and sample components mounted on the sample circuit board;
obtaining marking information of the sample component, wherein the marking information comprises the outline and the installation direction of the sample component;
and taking the sample image as input, and taking the annotation information as a truth label to train the recognition model.
5. The detection method according to claim 4, wherein the step of training the recognition model with the sample image as input and the annotation information as a truth label further comprises:
respectively carrying out segmentation processing on the plurality of sample images to obtain a plurality of sub-sample images respectively containing a single sample component;
the step of training the recognition model by using the sample image as input and the annotation information as a truth label includes:
and taking the sub-sample image as input, and taking the annotation information as a truth label to train the recognition model.
6. The method of claim 1, wherein the standard installation direction is stored in a preset configuration file;
the step of determining whether the component to be tested is correctly mounted on the circuit board to be tested according to the comparison result of the mounting direction predicted by the recognition model and the preset standard mounting direction comprises the following steps:
comparing the installation direction predicted by the recognition model with the corresponding standard installation direction in the configuration file.
7. The method of claim 6, wherein before comparing the installation direction predicted by the recognition model with the corresponding standard installation direction in the configuration file, further comprising:
and searching the corresponding standard installation direction from the configuration file according to the component name pre-assigned to the component to be tested.
8. The method according to claim 1, wherein the step of determining whether the component to be tested is correctly mounted on the circuit board to be tested according to the comparison result between the mounting direction predicted by the recognition model and a preset standard mounting direction comprises:
determining that the component to be tested is correctly installed on the circuit board to be tested according to the result that the installation direction predicted by the recognition model is consistent with the preset standard installation direction;
and determining that the component to be tested is incorrectly installed on the circuit board to be tested according to the result that the installation direction predicted by the recognition model is inconsistent with the preset standard installation direction.
9. A visual inspection apparatus comprising a processor, a memory and a communication circuit, the processor being coupled to the memory and the communication circuit respectively, the memory having stored therein program data, the processor implementing the steps of the method of any one of claims 1-8 by executing the program data in the memory.
10. An apparatus with storage function, characterized in that the apparatus with storage function stores program data executable by a processor to implement the steps of the method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010531361.4A CN113804704A (en) | 2020-06-11 | 2020-06-11 | Circuit board detection method, visual detection equipment and device with storage function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010531361.4A CN113804704A (en) | 2020-06-11 | 2020-06-11 | Circuit board detection method, visual detection equipment and device with storage function |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113804704A true CN113804704A (en) | 2021-12-17 |
Family
ID=78943742
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010531361.4A Pending CN113804704A (en) | 2020-06-11 | 2020-06-11 | Circuit board detection method, visual detection equipment and device with storage function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113804704A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114417626A (en) * | 2022-01-24 | 2022-04-29 | 深圳市云采网络科技有限公司 | Method and device for detecting assemblability, method and medium for checking bill of material |
-
2020
- 2020-06-11 CN CN202010531361.4A patent/CN113804704A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114417626A (en) * | 2022-01-24 | 2022-04-29 | 深圳市云采网络科技有限公司 | Method and device for detecting assemblability, method and medium for checking bill of material |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111814850A (en) | Defect detection model training method, defect detection method and related device | |
| CN109146873B (en) | Learning-based intelligent detection method and device for defects of display screen | |
| CN110136116B (en) | Injection pump defect detection method, device, equipment and storage medium | |
| TWI716012B (en) | Sample labeling method, device, storage medium and computing equipment, damage category identification method and device | |
| CN110335313B (en) | Audio acquisition equipment positioning method and device and speaker identification method and system | |
| CN111428374A (en) | Part defect detection method, device, equipment and storage medium | |
| CN113409250A (en) | Solder joint detection method based on convolutional neural network | |
| CN110346704A (en) | Determination method, apparatus, equipment and the storage medium of test file in board test | |
| CN114266764A (en) | Character integrity detection method and device for printed label | |
| CN113077416A (en) | Welding spot welding defect detection method and system based on image processing | |
| CN111652145A (en) | Formula detection method and device, electronic equipment and storage medium | |
| CN113804704A (en) | Circuit board detection method, visual detection equipment and device with storage function | |
| CN111368824A (en) | Instrument identification method, mobile device and storage medium | |
| CN117274245B (en) | AOI optical detection method and system based on image processing technology | |
| CN110817674B (en) | Method, device and equipment for detecting step defect of escalator and storage medium | |
| CN116978638A (en) | Automatic cable wrapping control method, device and medium | |
| CN114693554B (en) | Big data image processing method and system | |
| CN115718830A (en) | Method for training information extraction model, information extraction method and corresponding device | |
| CN114841255A (en) | Detection model training method, device, equipment, storage medium and program product | |
| CN113808067A (en) | Circuit board detection method, visual detection equipment and device with storage function | |
| CN112052883B (en) | Clothes detection method, device and storage medium | |
| CN111768439B (en) | Method, device, electronic equipment and medium for determining experiment scores | |
| CN107341830A (en) | A kind of method and device for determining shell hole ring value | |
| CN115184378B (en) | Concrete structure disease detection system and method based on mobile equipment | |
| CN116952166B (en) | Method, device, equipment and medium for detecting parts of automobile door handle assembly |
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 |