scholarly journals The Impact of a Number of Samples on Unsupervised Feature Extraction, Based on Deep Learning for Detection Defects in Printed Circuit Boards

2021 ◽  
Vol 14 (1) ◽  
pp. 8
Author(s):  
Ihar Volkau ◽  
Abdul Mujeeb ◽  
Wenting Dai ◽  
Marius Erdt ◽  
Alexei Sourin
Keyword(s):  
Deep Learning ◽  
Defect Detection ◽  
Printed Circuit Boards ◽  
Learning Model ◽  
Circuit Boards ◽  
Printed Circuit ◽  
Unsupervised Deep Learning ◽  
The Impact ◽  
Deep Learning Model ◽  
Unsupervised Feature Extraction

Deep learning provides new ways for defect detection in automatic optical inspections (AOI). However, the existing deep learning methods require thousands of images of defects to be used for training the algorithms. It limits the usability of these approaches in manufacturing, due to lack of images of defects before the actual manufacturing starts. In contrast, we propose to train a defect detection unsupervised deep learning model, using a much smaller number of images without defects. We propose an unsupervised deep learning model, based on transfer learning, that extracts typical semantic patterns from defect-free samples (one-class training). The model is built upon a pre-trained VGG16 model. It is further trained on custom datasets with different sizes of possible defects (printed circuit boards and soldered joints) using only small number of normal samples. We have found that the defect detection can be performed very well on a smooth background; however, in cases where the defect manifests as a change of texture, the detection can be less accurate. The proposed study uses deep learning self-supervised approach to identify if the sample under analysis contains any deviations (with types not defined in advance) from normal design. The method would improve the robustness of the AOI process to detect defects.

scholarly journals Applying deep learning to defect detection in printed circuit boards via a newest model of you-only-look-once

2021 ◽  
Vol 18 (4) ◽  
pp. 4411-4428
Author(s):  
Venkat Anil Adibhatla ◽  
◽  
Huan-Chuang Chih ◽  
Chi-Chang Hsu ◽  
Joseph Cheng ◽  
...  
Keyword(s):  
Deep Learning ◽  
Defect Detection ◽  
Printed Circuit Boards ◽  
Circuit Boards ◽  
Printed Circuit

scholarly journals Defect Detection in Printed Circuit Boards Using You-Only-Look-Once Convolutional Neural Networks

Electronics ◽  
2020 ◽  
Vol 9 (9) ◽  
pp. 1547
Author(s):  
Venkat Anil Adibhatla ◽  
Huan-Chuang Chih ◽  
Chi-Chang Hsu ◽  
Joseph Cheng ◽  
Maysam F. Abbod ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Defect Detection ◽  
Printed Circuit Boards ◽  
Learning Algorithms ◽  
Detection Accuracy ◽  
Quality Inspection ◽  
Circuit Boards ◽  
Printed Circuit

In this study, a deep learning algorithm based on the you-only-look-once (YOLO) approach is proposed for the quality inspection of printed circuit boards (PCBs). The high accuracy and efficiency of deep learning algorithms has resulted in their increased adoption in every field. Similarly, accurate detection of defects in PCBs by using deep learning algorithms, such as convolutional neural networks (CNNs), has garnered considerable attention. In the proposed method, highly skilled quality inspection engineers first use an interface to record and label defective PCBs. The data are then used to train a YOLO/CNN model to detect defects in PCBs. In this study, 11,000 images and a network of 24 convolutional layers and 2 fully connected layers were used. The proposed model achieved a defect detection accuracy of 98.79% in PCBs with a batch size of 32.

scholarly journals Dismantling of Waste Printed Circuit Boards with the Simultaneous Recovery of Copper: Experimental Study and Process Modeling

Materials ◽  
2021 ◽  
Vol 14 (18) ◽  
pp. 5186
Author(s):  
Szabolcs Fogarasi ◽  
Árpád Imre-Lucaci ◽  
Florica Imre-Lucaci
Keyword(s):  
Environmental Performance ◽  
Printed Circuit Boards ◽  
Electrochemical Process ◽  
Circuit Boards ◽  
Electronic Components ◽  
Printed Circuit ◽  
Waste Printed Circuit Boards ◽  
Different Types ◽  
Leaching Solution ◽  
The Impact

The study was carried out with the aim to demonstrate the applicability of a combined chemical–electrochemical process for the dismantling of waste printed circuit boards (WPCBs) created from different types of electronic equipment. The concept implies a simple and less polluting process that allows the chemical dismantling of WPCBs with the simultaneous recovery of copper from the leaching solution and the regeneration of the leaching agent. In order to assess the performance of the dismantling process, various tests were performed on different types of WPCBs using the 0.3 M FeCl3 in 0.5 M HCl leaching system. The experimental results show that, through the leaching process, the electronic components (EC) together with other fractions can be efficiently dismounted from the surface of WPCBs, with the parallel electrowinning of copper from the copper rich leaching solution. In addition, the process was scaled up for the dismantling of 100 kg/h WPCBs and modeled and simulated using process flow modelling software ChemCAD in order to assess the impact of all steps and equipment on the technical and environmental performance of the overall process. According to the results, the dismantling of 1 kg of WPCBs requires a total energy of 0.48 kWh, and the process can be performed with an overall low environmental impact based on the obtained general environmental indexes (GEIs) values.

Nanovoid Formation at Cu/Cu/Cu Interconnections of Blind Microvias: A Field Study

2019 ◽  
Vol 2019 (1) ◽  
pp. 000492-000502 ◽  
Author(s):  
T. Bernhard ◽  
L. Gregoriades ◽  
S. Branagan ◽  
L. Stamp ◽  
E. Steinhäuser ◽  
...  
Keyword(s):  
Printed Circuit Boards ◽  
Structural Integrity ◽  
Weak Link ◽  
Critical Factor ◽  
Circuit Boards ◽  
Printed Circuit ◽  
Different Types ◽  
Key Factor ◽  
Electrical Reliability ◽  
The Impact

Abstract A key factor for a high electrical reliability of multilayer High Density Interconnection Printed Circuit Boards (HDI PCBs) is the thermomechanical stability of stacked microvia interconnections. With decreasing via sizes and higher numbers of interconnected layers, the structural integrity of these interconnections becomes a critical factor and is a topic of high interest in current research. The formation of nanovoids and inhibited Cu recrystallization across the interfaces are the two main indications of a weak link from the target pad to the filled via. Based on TEM/EDX measurements on a statistically relevant number of stacked and blind microvias produced in the industrial field, different types of nanovoid phenomena are revealed at the Cu/Cu/Cu junction. The types of nanovoids were categorized relating to the time of appearance (before or after thermal treatment), the affected interfaces or layers and the impact on the Cu recrystallization. The main root causes for each void type are identified and the expected impact on the thermomechanical stability of the via junction is discussed.

scholarly journals The Impact of Emotional Music on Active ROI in Patients with Depression Based on Deep Learning: A Task-State fMRI Study

2019 ◽  
Vol 2019 ◽  
pp. 1-14
Author(s):  
Renzhou Gui ◽  
Tongjie Chen ◽  
Han Nie
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Correlation Matrix ◽  
Nearest Neighbor ◽  
Learning Model ◽  
Fmri Data ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
The Impact ◽  
Deep Learning Model

With the continuous development of science, more and more research results have proved that machine learning is capable of diagnosing and studying the major depressive disorder (MDD) in the brain. We propose a deep learning network with multibranch and local residual feedback, for four different types of functional magnetic resonance imaging (fMRI) data produced by depressed patients and control people under the condition of listening to positive- and negative-emotions music. We use the large convolution kernel of the same size as the correlation matrix to match the features and obtain the results of feature matching of 264 regions of interest (ROIs). Firstly, four-dimensional fMRI data are used to generate the two-dimensional correlation matrix of one person’s brain based on ROIs and then processed by the threshold value which is selected according to the characteristics of complex network and small-world network. After that, the deep learning model in this paper is compared with support vector machine (SVM), logistic regression (LR), k-nearest neighbor (kNN), a common deep neural network (DNN), and a deep convolutional neural network (CNN) for classification. Finally, we further calculate the matched ROIs from the intermediate results of our deep learning model which can help related fields further explore the pathogeny of depression patients.

scholarly journals Effect of Graphite on Copper Bioleaching from Waste Printed Circuit Boards

Minerals ◽  
2020 ◽  
Vol 10 (1) ◽  
pp. 79 ◽  
Author(s):  
Linlin Tong ◽  
Qianfei Zhao ◽  
Ali Kamali ◽  
Wolfgang Sand ◽  
Hongying Yang
Keyword(s):  
Printed Circuit Boards ◽  
Acid Leaching ◽  
Circuit Boards ◽  
Printed Circuit ◽  
Waste Printed Circuit Boards ◽  
Copper Leaching ◽  
Initial Ph ◽  
Efficient Extraction ◽  
Bacteria Culture ◽  
The Impact

The efficient extraction of copper as a valuable metal from waste printed circuit boards (WPCBs) is currently attracting growing interest. Here, we systematically investigated the impact of bacteria on the efficiency of copper leaching from WPCBs, and evaluated the effect of graphite on bioleaching performance. The HQ0211 bacteria culture containing Acidithiobacillus ferrooxidans, Ferroplasma acidiphilum, and Leptospirillum ferriphilum enhanced Cu-leaching performance in either ferric sulfate and sulfuric acid leaching, so a final leaching of up to 76.2% was recorded after 5 days. With the addition of graphite, the percentage of copper leaching could be increased to 80.5%. Single-factor experiments confirmed the compatibility of graphite with the HQ0211 culture, and identified the optimal pulp density of WPCBs, the initial pH, and the graphite content to be 2% (w/v), 1.6, and 2.5 g/L, respectively.

Sign in / Sign up

Export Citation Format

Share Document