scholarly journals Compilation of training datasets for use of convolutional neural networks supporting automatic inspection processes in industry 4.0 based electronic manufacturing

2020 ◽  
Vol 9 (1) ◽  
pp. 167-178 ◽  
Author(s):  
Alida Ilse Maria Schwebig ◽  
Rainer Tutsch
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Printed Circuit Boards ◽  
Recognition Performance ◽  
Inspection System ◽  
Automotive Electronics ◽  
Automatic Optical Inspection ◽  
Optical Inspection ◽  
Competitive Prices

Abstract. Ensuring the highest quality standards at competitive prices is one of the greatest challenges in the manufacture of electronic products. The identification of flaws has the uppermost priority in the field of automotive electronics, particularly as a failure within this field can result in damages and fatalities. During assembling and soldering of printed circuit boards (PCBs) the circuit carriers can be subject to errors. Hence, automatic optical inspection (AOI) systems are used for real-time detection of visible flaws and defects in production. This article introduces an application strategy for combining a deep learning concept with an optical inspection system based on image processing. Above all, the target is to reduce the risk of error slip through a second inspection. The concept is to have the inspection results additionally evaluated by a convolutional neural network. For this purpose, different training datasets for the deep learning procedures are examined and their effects on the classification accuracy for defect identification are assessed. Furthermore, a suitable compilation of image datasets is elaborated, which ensures the best possible error identification on solder joints of electrical assemblies. With the help of the results, convolutional neural networks can achieve a good recognition performance, so that these can support the automatic optical inspection in a profitable manner. Further research aims at integrating the concept in a fully automated way into the production process in order to decide on the product quality autonomously without human interference.

scholarly journals Intelligent fault detection of electrical assemblies using hierarchical convolutional networks for supporting automatic optical inspection systems

2020 ◽  
Vol 9 (2) ◽  
pp. 363-374
Author(s):  
Alida Ilse Maria Schwebig ◽  
Rainer Tutsch
Keyword(s):  
Neural Networks ◽  
Recognition Accuracy ◽  
Early Stage ◽  
Classification Systems ◽  
Inspection System ◽  
Automotive Electronics ◽  
Automatic Optical Inspection ◽  
Convolutional Networks ◽  
Optical Inspection ◽  
Inspection Systems

Abstract. Electrical assemblies are the core of many electronic devices and therefore represent a crucial part of the overall product, which must be carefully checked before integration into its functional environment. For this reason, automatic optical inspection systems are required in electronic manufacturing to detect visible errors in products at an early stage. In particular, the automotive electronics production area is one of the sectors in which quality assurance has uppermost priority, as undetected defects can pose a danger to life. However, most optical inspection processes still have error slippage rates, which are responsible for delivering faulty electrical assemblies to customers. Therefore, this article shows how an application strategy of deep learning, based on neural networks, can be combined with an automatic optical inspection system to further increase the recognition accuracy of the process. The additional use of artificial intelligence supported classification systems provides a way to find out the exact details about the manufacturing-related errors of electrical assemblies. However, due to the high number of different error categories, a single classification algorithm is usually not sufficient to provide reliable visual inspection results with high robustness against error slip. For this reason, a hierarchical model with multiple classifiers is proposed in this article. The principle is based on the hierarchical description of the quality status and fault types using several combined neural networks. In this context, the original classification task is distributed over different subnetworks. These subnetworks, which interact as an overall model, only verify certain error and quality features of the electrical assemblies, which means that higher recognition accuracy and robustness can be achieved compared to a single network.

scholarly journals A Performance Evaluation of Classic Convolutional Neural Networks for 2D and 3D Palmprint and Palm Vein Recognition

2020 ◽  
Author(s):  
Wei Jia ◽  
Jian Gao ◽  
Wei Xia ◽  
Yang Zhao ◽  
Hai Min ◽  
...  
Keyword(s):  
Neural Networks ◽  
Performance Evaluation ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Recognition Performance ◽  
Mixed Data ◽  
Palmprint Recognition ◽  
Vein Recognition ◽  
Palm Vein ◽  
2D And 3D

AbstractPalmprint recognition and palm vein recognition are two emerging biometrics technologies. In the past two decades, many traditional methods have been proposed for palmprint recognition and palm vein recognition, and have achieved impressive results. However, the research on deep learning-based palmprint recognition and palm vein recognition is still very preliminary. In this paper, in order to investigate the problem of deep learning based 2D and 3D palmprint recognition and palm vein recognition in-depth, we conduct performance evaluation of seventeen representative and classic convolutional neural networks (CNNs) on one 3D palmprint database, five 2D palmprint databases and two palm vein databases. A lot of experiments have been carried out in the conditions of different network structures, different learning rates, and different numbers of network layers. We have also conducted experiments on both separate data mode and mixed data mode. Experimental results show that these classic CNNs can achieve promising recognition results, and the recognition performance of recently proposed CNNs is better. Particularly, among classic CNNs, one of the recently proposed classic CNNs, i.e., EfficientNet achieves the best recognition accuracy. However, the recognition performance of classic CNNs is still slightly worse than that of some traditional recognition methods.

scholarly journals AE-RTISNet: Aeronautics Engine Radiographic Testing Inspection System Net with Improved Fast Region-based Convolutional Neural Networks Framework

2020 ◽  
Author(s):  
Zhi-Hao Chen ◽  
Jyh-Ching Juang
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Defect Detection ◽  
Training Model ◽  
Inspection System ◽  
Destructive Testing ◽  
X Ray ◽  
Nondestructive Examination ◽  
Fast Region

To ensure the safety in aircraft flying, we aim use of the deep learning methods of nondestructive examination with multiple defect detection paradigms for X-ray image detection posed. The use of the Fast Region-based Convolutional Neural Networks (Fast R-CNN) driven model seeks to augment and improve existing automated Non-Destructive Testing (NDT) diagnosis. Within the context of X-ray screening, limited numbers insufficient types of X-ray aeronautics engine defect data samples can thus pose another problem in training model tackling multiple detections perform accuracy. To overcome this issue, we employ a deep learning paradigm of transfer learning tackling both single and multiple detection. Overall the achieve result get more then 90% accuracy based on the AE-RTISNet retrained with 8 types of defect detection. Caffe structure software to make networks tracking detection over multiples Fast R-CNN. We consider the AE-RTISNet provide best results to the more traditional multiple Fast R-CNN approaches simpler translate to C++ code and installed in the Jetson™ TX2 embedded computer. With the use of LMDB format, all images using input images of size 640 × 480 pixel. The results scope achieves 0.9 mean average precision (mAP) on 8 types of material defect classifiers problem and requires approximately 100 microseconds.

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 Deep convolutional neural networks for cardiovascular vulnerable plaque detection

2019 ◽  
Vol 277 ◽  
pp. 02024 ◽  
Author(s):  
Lincan Li ◽  
Tong Jia ◽  
Tianqi Meng ◽  
Yizhe Liu
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Vulnerable Plaque ◽  
Recall Rate ◽  
Superior Performance ◽  
Learning Approaches ◽  
Deep Convolutional Neural Networks ◽  
Vulnerable Plaques ◽  
Plaque Detection

In this paper, an accurate two-stage deep learning method is proposed to detect vulnerable plaques in ultrasonic images of cardiovascular. Firstly, a Fully Convonutional Neural Network (FCN) named U-Net is used to segment the original Intravascular Optical Coherence Tomography (IVOCT) cardiovascular images. We experiment on different threshold values to find the best threshold for removing noise and background in the original images. Secondly, a modified Faster RCNN is adopted to do precise detection. The modified Faster R-CNN utilize six-scale anchors (122,162,322,642,1282,2562) instead of the conventional one scale or three scale approaches. First, we present three problems in cardiovascular vulnerable plaque diagnosis, then we demonstrate how our method solve these problems. The proposed method in this paper apply deep convolutional neural networks to the whole diagnostic procedure. Test results show the Recall rate, Precision rate, IoU (Intersection-over-Union) rate and Total score are 0.94, 0.885, 0.913 and 0.913 respectively, higher than the 1st team of CCCV2017 Cardiovascular OCT Vulnerable Plaque Detection Challenge. AP of the designed Faster RCNN is 83.4%, higher than conventional approaches which use one-scale or three-scale anchors. These results demonstrate the superior performance of our proposed method and the power of deep learning approaches in diagnose cardiovascular vulnerable plaques.

Sign in / Sign up

Export Citation Format

Share Document