Positive and negative sample data balancing method in factory PCB defect detection
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A PCB board and defect detection technology, applied in the field of data processing, can solve problems such as model overfitting, positive and negative sample imbalance, etc., and achieve the effect of strong robustness, broad market application prospects, and superior synthesis effect
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
In the PCB defect detection project, the number of negative samples with defects is only a thousand, while the number of positive samples without defects is on the order of 100,000. It can be seen
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0053] Such as Figure 1 to Figure 5 As shown, the present embodiment discloses a positive and negative sample data balance method in a factory PCB board defect detection, and the balance method mainly includes the following specific steps:
[0054] 1) Collect, organize and classify data sets: collect image data sets of factory PCB boards, organize the data sets, and manually classify according to whether there are defects.
[0055] Specifically, in step 1) data preprocessing needs to set specific and clear classification rules, and the classification boundaries cannot be blurred. A positive sample refers to a PCB without defects, and a negative sample refers to a defective PCB.
[0056] Preferably, for simplicity, the input images are all regularized to a size of 256*256.
[0057] 2) Construct a generative adversarial network model to solve the problem of imbalance between positive and negative samples, aiming to design a network structure based on two groups of unidirectio...
Embodiment 2
[0077] combine Figure 1 to Figure 8 As shown, this embodiment discloses a method for balancing positive and negative sample data in a factory PCB board defect detection, including the following specific implementation steps:
[0078] 1) Collect the image data set of the PCB board in the factory, organize the data set, and manually classify it according to whether there are defects.
[0079] 2) Construct a generative adversarial network model to solve the problem of imbalance between positive and negative samples. The envisaged method is unsupervised image-to-image translation, aiming to design a network structure based on two sets of unidirectional GANs and achieve bidirectional image generation.
[0080] 3) The model is divided into generator and discriminator. Firstly, the generator module is designed. Generators include encoders, converters and decoders.
[0081] 4) Design of the encoder. Different channels of the encoder output feature map combine different features ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention discloses a positive and negative sample data balancing method in factory PCB defect detection, which is a data balancing method in PCB positive and negative sample classification basedon an adversarial generative network, and mainly comprises the following steps: collecting, sorting and classifying a data set; designing an encoder which is composed of five convolution layers, and extracting features from the input image by the encoder; designing a converter which is composed of eight residual blocks and converts the feature vector from a source domain to a target domain; designing a decoder, wherein the decoder is composed of five deconvolution layers; designing a discriminator, wherein the discriminator is composed of seven convolution layers; designing a loss function, wherein the loss function comprises four parts; preparing a training set for model training; the obtained weight file is used for a test set, and a negative sample needing to be amplified is synthesized. The method is high in robustness, wide in application range and excellent in synthesis effect. And by means of cyclic consistency conditions, the effect of standardizing the model is achieved, and the generation effect of the shape and texture of the synthesized image is flexibly controlled to a certain extent.
Description
technical field [0001] The invention relates to the technical field of data processing, in particular to a method for balancing positive and negative samples in a factory assembly line product based on an adversarial generation network. Background technique [0002] Printed Circuit Board (PCB) welding machines are mainly used on equipment in the electronics industry. The industry has higher quality requirements for PCB boards. Therefore, the quality of PCB boards needs to be tested. PCB board has the characteristics of high density, high precision and high reliability. As the number of PCB board layers increases, the density increases, and the volume of the PCB board decreases, making it more difficult to inspect the quality of the PCB board. [0003] In the production line, due to reasons such as immature welding technology and manual operation errors, quality problems such as welding defects often occur, which greatly affect the use of products. [0004] In the past, aut...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more