scholarly journals Construction of Defect Detection System for Image DataUsing Machine Learning and Image Processing

2017 ◽  
Vol 3 (2) ◽  
pp. 46-58 ◽  
Author(s):  
Kenta YOSHIDA ◽  
Tatsuya IWASAWA ◽  
Natsuki SANO ◽  
Mirai TANAKA ◽  
Tomomichi SUZUKI
Keyword(s):  
Machine Learning ◽  
Image Processing ◽  
Defect Detection ◽  
Detection System ◽  
Image Data

scholarly journals Lightweight Convolutional Neural Network Based Intrusion Detection System

2020 ◽  
pp. 808-817
Author(s):  
Vinh Pham ◽  
◽  
Eunil Seo ◽  
Tai-Myoung Chung
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Intrusion Detection ◽  
Convolutional Neural Network ◽  
Network Traffic ◽  
Detection System ◽  
Image Data ◽  
Training Data ◽  
Deep Packet Inspection ◽  
Detection Model

Identifying threats contained within encrypted network traffic poses a great challenge to Intrusion Detection Systems (IDS). Because traditional approaches like deep packet inspection could not operate on encrypted network traffic, machine learning-based IDS is a promising solution. However, machine learning-based IDS requires enormous amounts of statistical data based on network traffic flow as input data and also demands high computing power for processing, but is slow in detecting intrusions. We propose a lightweight IDS that transforms raw network traffic into representation images. We begin by inspecting the characteristics of malicious network traffic of the CSE-CIC-IDS2018 dataset. We then adapt methods for effectively representing those characteristics into image data. A Convolutional Neural Network (CNN) based detection model is used to identify malicious traffic underlying within image data. To demonstrate the feasibility of the proposed lightweight IDS, we conduct three simulations on two datasets that contain encrypted traffic with current network attack scenarios. The experiment results show that our proposed IDS is capable of achieving 95% accuracy with a reasonable detection time while requiring relatively small size training data.

scholarly journals Study on Machine Learning Based Intelligent Defect Detection System

2018 ◽  
Vol 201 ◽  
pp. 01010 ◽  
Author(s):  
Chung-Chi Huang ◽  
Xin-Pu Lin
Keyword(s):  
Machine Learning ◽  
Machine Vision ◽  
Defect Detection ◽  
Detection System ◽  
Vision System ◽  
National Development ◽  
Image Preprocessing ◽  
Testing Procedures ◽  
Metal Products

In the paper, it is proposed to develop a machine learning based intelligent defect detection system for metal products. The common machine vision system has the surface (stain, shallow pit, shallow tumor, scratches, Edge defects, pattern defects) detection, or for the processing of the size, diameter, diameter, eccentricity, height, thickness and other parts of the non-contact numerical parameters of detection. Considering the quality of the work piece and the defects of the standard, so for the quality of customized testing requirements, the study is the development of machine vision and machine learning metal products defect detection system, mainly composed of three procedures: Image preprocessing, training procedures and testing procedures. The system architecture consists of three parts: (1) Image preprocessing: we first use the machine vision. OPENCV to carry out the image pre-processing part of the product before the detection. (2) Training procedures: The algorithm of the machine learning includes the convolution neural network (CNN), chunk-max pooling is used to train the program, and the generative adversarial network (GAN) based architecture is used to solve the problem of small datasets for surface defects. (3) Testing procedures：The Python language is used to write the program and implement the testing procedures with the GPU-Based embedded hardware In industries, collecting training dataset is usually costly and related methods are highly dataset-dependent. So most companies cannot provide Big-data to be analyzed or applied. By the experimental results, the recognition accuracy can be obviously improved as increasing data augmentation by GAN-Based samples maker. Manual inspection is labor intensive, costly and less in efficiency. Therefore, this study will contribute to technological innovation, industry, national development and other applications. (1) The use of intelligent machine learning technology will make the industry 4.0 technology more sophisticated. (2) It will make the development of equipment industry be better by the machine learning applications. (3) It will increase the economics and productivity of countries for the aging of the population by machine learning.

705 Development of defect detection system by image processing using wavelet transform

2010 ◽  
Vol 2010.18 (0) ◽  
pp. 185-186
Author(s):  
Kaoru TAKAMORI ◽  
Masashi ONO ◽  
Kazutaka NONOMURA ◽  
Libo ZHOU ◽  
Hirotaka OJIMA
Keyword(s):  
Image Processing ◽  
Wavelet Transform ◽  
Defect Detection ◽  
Detection System

The Electron Microscope and the Computer: Present Trends and Future Prospects

1990 ◽  
Vol 48 (1) ◽  
pp. 116-117
Author(s):  
Peter Rez ◽  
W.J. de Ruijter
Keyword(s):  
Image Processing ◽  
Dynamic Range ◽  
Detection System ◽  
Image Data ◽  
Analytical Techniques ◽  
Digital Data ◽  
External Control ◽  
Major Barrier ◽  
Computer Based ◽  
Electron Microscopes

In the present generation of electron microscopes the roles of computers or microprocessors can be divided into control, acquisition and analysis of both spectral and image data. Not much, however, has been done to realise the full power of computer based systems and integrate all these functions with the electron optics. The control of practically all microscope columns is performed by an 8-bit or 16-bit processor usually running a program that repeatedly scans for user inputs and changes lens currents or alignment settings if necessary. External control for special experiments can either be implemented by scanning an additional user input from a serial port or by allowing external analog signals to replace those generated by the microscope control scheme. As the program loop typically takes 0.1 sec to complete it is preferable to implement some functions such as external beam scanning by providing analog ramps (even if generated by another computer).Computer acquistion of data was introduced to electron microscopy with analytical techniques, such as EDX, in which the computer was the basis of a multi channel analyser. In the case of energy loss and Auger spectroscopy, computer scanning of the spectrometer and acquisition of single electron pulse-counted data quickly displaced chart recorders as a means of collecting data. A computer based system not only could perform acquisition more efficiently, it could also provide a convenient means for processing the results and doing quantitative analysis. Furthermore digitally stored data could easily be transferred to other systems on disks or by direct link (such as ethernet) and analysed elsewhere. However for image acquisition there has been very little use of computers in acquiring data. Although microscopists are happy to consider a spectrum as an array of numbers they still prefer to deal with images as pictures rather than digital data sets. The problems are not entirely psychological since a major barrier to the widespread use of image processing in microscopy is the lack of a suitable detection system. TV cameras compare unfavourably with photographic plates in terms of both dynamic range and "resolution" as defined loosely in terms of pixel size or lines/mm. This argument does not apply to scanning microscopes but, even in scanning systems, frame buffers and powerful computer systems have only been integrated as part of the microscope electronics in the last few years. Microscopists still prefer to measure quantities from exposed photographs rather than work with digitized data in a workstation environment using high level image processing software. Until recently cost may have been a consideration but now computing platforms of sufficient capability are less than 1/5 of the cost of an average SEM.

scholarly journals Micro pin header defect detection system based on OpenCV

2021 ◽  
Vol 2137 (1) ◽  
pp. 012037
Author(s):  
Houcheng Yang ◽  
Yinxin Yan ◽  
Zhangsi Yu ◽  
Zhang Ning
Keyword(s):  
Image Processing ◽  
Success Rate ◽  
Real Time ◽  
Light Source ◽  
Defect Detection ◽  
Detection Efficiency ◽  
Detection System ◽  
Experimental Tests ◽  
Quality Inspection ◽  
Detection Error

Abstract In order to solve the problems of low detection efficiency and large detection error in the process of manual quality inspection, a full-automatic defect detection system is built. The system uses an industrial camera, selects a suitable light source for image acquisition, uses the open source OpenCV visual library for image processing and defect contour recognition, and sets the screening conditions for unqualified products. The system can detect whether the needle arrangement has defects in real time and classify them according to different defect categories, It can greatly improve the detection efficiency of needle arranging production enterprises. Through a large number of experimental tests, the detection success rate can reach 98.67%, which shows that the system is feasible.

Rapid Defect Detection and Classification in Images Using Convolutional Neural Networks

2021 ◽  
Author(s):  
Peter Warren ◽  
Hessein Ali ◽  
Hossein Ebrahimi ◽  
Ranajay Ghosh
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Defect Detection ◽  
Visual Inspection ◽  
Surface Defects ◽  
Image Data ◽  
Rapid Identification ◽  
Machine Learning Techniques ◽  
Great Success

Abstract Several image processing methods have been implemented over recent years to assist and partially replace on-site technician visual inspection of both manufactured parts and operational equipments. Convolutional neural networks (CNNs) have seen great success in their ability to both identify and classify anomalies within images, in some cases they do this to a higher degree of accuracy than an expert human. Several parts that are manufactured for various aspects of turbomachinery operation must undergo a visual inspection prior to qualification. Machine learning techniques can streamline these visual inspection processes and increase both efficiency and accuracy of defect detection and classification. The adoption of CNNs to manufactured part inspection can also help to improve manufacturing methods by rapidly retrieving data for overall system improvement. In this work a dataset of images with a variety of surface defects and some without defects will be fed through varying CNN set-ups for the rapid identification and classification of the flaws within the images. This work will examine the techniques used to create CNNs and how they can best be applied to part surface image data, and determine the most accurate and efficient techniques that should be implemented. By combining machine learning with non-destructive evaluation methods component health can be rapidly determined and create a more robust system for manufactured parts and operational equipment evaluation.

scholarly journals Image Processing Based Automatic Detection of Asphalt Pavement Rutting Using a Novel Metaheuristic Optimized Machine Learning Approach

2021 ◽  
Author(s):  
Minh-Tu Cao ◽  
Kuan-Tsung Chang ◽  
Ngoc-Mai Nguyen ◽  
Van-Duc Tran ◽  
Xuan-Linh Tran ◽  
...  
Keyword(s):  
Machine Learning ◽  
Image Processing ◽  
Prediction Model ◽  
Gabor Filter ◽  
Asphalt Pavement ◽  
Model Performance ◽  
Image Data ◽  
Support Vector ◽  
Dynamic Feature ◽  
Metaheuristic Optimization

Abstract This study presents a novel computer vision based approach to automatically identify rutting appeared on asphalt pavement of road. The developed model is established base on a hybridization of image processing techniques and an advanced machine learning model with support of a metaheuristic optimization engine. Gabor filter and discrete cosine transform are employed to implement context computation for image data, accordingly generate initially extracted features of rutting and non-rutting. Least Squares Support Vector Classification (LSSVC) is then used to learn categorization of rutting and non-rutting based on the extracted features. The final LSSVC prediction model is constructed via a loop of optimization process which is controlled by a novel metaheuristic optimization algorithm, called forensic-based investigation (FBI), to attain optimal model’s configuration with ultimate prediction accuracy. This study further utilized a dynamic feature selection (FS) method to integrate in the searching loop to appropriately remove redundant features that provide inconsistent information leading to the compromising of model performance. A dataset of 2000 image samples has been collected during field trip of pavement survey in Da Nang city to form and verify the newly developed model. The statistical results of 20 run times using k-fold cross validation method have demonstrated the hybrid model of FBI-LSSVC-FS to achieve the most desired rutting detection performance with classification accuracy rate, precision, recall, and F1 score of 98.9%, 0.994, 0.984 and 0.989, respectively. Hence, this paper contributes to the core body of knowledge a novel AI-based prediction model to assist transportation agencies in the task of periodic asphalt pavement survey.

scholarly journals Defect Detection in Printed Board Circuit using Image Processing

2019 ◽  
Vol 9 (2) ◽  
pp. 3005-3010
Keyword(s):  
Machine Learning ◽  
Image Processing ◽  
Defect Detection ◽  
Printed Circuit Board ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Circuit Board ◽  
Basic Part ◽  
Printed Circuit ◽  
Related Information

A printed circuit board without connecting with any components called as a bare PCB. Consider a PCB as a basic part which has been settled with more electronic units. In order to display the manufacturing process, the drawbacks have been taken by PCB individually. The reflection of this separation process impacts the performance of the circuits. Also, we have examined about classification methodologies as well as referential based PCB detection. From the input images, the needed and related information has been pulled out using image processing methodologies by the referential based PCB detection. Comparing with the un-defected PCB images, this was used to find out the defects. To meet the goal of the PCB defect detection, several feature extraction and pre-processing methods are derived in this article. The PCB defects have been classified by those features using the machine learning algorithms. Moreover, several types of machine learning algorithms are derived in this article. This paper helps the researchers for achieving a better solution for image processing and machine learning-based printed circuit board the defect classification

Sign in / Sign up

Export Citation Format

Share Document