Development of Machine Vision System Based on BP Neural Network Self-learning

2008 ◽  
Author(s):  
Ge Dongyuan ◽  
Yao Xifan ◽  
Zhang Qing
Keyword(s):  
Neural Network ◽  
Machine Vision ◽  
Bp Neural Network ◽  
Vision System ◽  
Machine Vision System ◽  
Self Learning

scholarly journals Design and development of a machine vision system using artificial neural network-based algorithm for automated coal characterization

2020 ◽  
Author(s):  
Amit Kumar Gorai ◽  
Simit Raval ◽  
Ashok Kumar Patel ◽  
Snehamoy Chatterjee ◽  
Tarini Gautam
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Machine Vision ◽  
Vision System ◽  
Color Space ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Machine Vision System ◽  
Coal Characterization ◽  
Artificial Neural

Abstract Coal is heterogeneous in nature, and thus the characterization of coal is essential before its use for a specific purpose. Thus, the current study aims to develop a machine vision system for automated coal characterizations. The model was calibrated using 80 image samples that are captured for different coal samples in different angles. All the images were captured in RGB color space and converted into five other color spaces (HSI, CMYK, Lab, xyz, Gray) for feature extraction. The intensity component image of HSI color space was further transformed into four frequency components (discrete cosine transform, discrete wavelet transform, discrete Fourier transform, and Gabor filter) for the texture features extraction. A total of 280 image features was extracted and optimized using a step-wise linear regression-based algorithm for model development. The datasets of the optimized features were used as an input for the model, and their respective coal characteristics (analyzed in the laboratory) were used as outputs of the model. The R-squared values were found to be 0.89, 0.92, 0.92, and 0.84, respectively, for fixed carbon, ash content, volatile matter, and moisture content. The performance of the proposed artificial neural network model was also compared with the performances of performances of Gaussian process regression, support vector regression, and radial basis neural network models. The study demonstrates the potential of the machine vision system in automated coal characterization.

Identification of Peanut Pods with Three or More Kernels by Machine Vision and Neural Network

2014 ◽  
Vol 10 (1) ◽  
pp. 97-102
Author(s):  
Jason Wang ◽  
Wade W. Yang ◽  
Lloyd T. Walker ◽  
Taha Rababah
Keyword(s):  
Neural Network ◽  
Machine Vision ◽  
Vision System ◽  
Major Axis ◽  
Machine Vision System ◽  
Discrimination Accuracy ◽  
Projected Area ◽  
Niche Marketing ◽  
Axis Length ◽  
Physical Features

Abstract Separation of unshelled peanuts containing three or more kernels and then niche marketing them can potentially increase the value of unshelled peanuts and thus the profit of peanut producers or processors. Effective identification of peanut pods with three or more kernels is a critical step prior to separation. In this study, a machine vision system was teamed up with neural network technique to discriminate unshelled peanuts into two groups: one with three or more kernels and the other with two or less kernels. A set of physical features including the number of bumps, projected area, length and perimeter, etc., were extracted from the images taken and used to train an artificial neural network for discriminating the peanuts. It was found that among all the selected features, the length, the major axis length and perimeter have the best correlation with the number of kernels (correlation coefficient r = 0.87–0.88); the area and convex area have good correlation (r = 0.85); the eccentricity, number of bumps, and the compactness have relatively lower correction (r = 0.77–0.80); the solidity and the minor axis length have the least correlation to the number of kernels (r = −0.415–0.26). The best discrimination accuracy obtained for peanut pods with three or more kernels was 92.5% for the conditions used in this study.

UAV Navigation System Autonomous Correction Algorithm Based on Road and River Network Recognition

2020 ◽  
Vol 28 (3) ◽  
pp. 32-42
Author(s):  
A.P. Tanchenko ◽  
◽  
A.M. Fedulin ◽  
R.R. Bikmaev ◽  
R.N. Sadekov ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Vision ◽  
Convolutional Neural Network ◽  
Navigation System ◽  
Vision System ◽  
River Network ◽  
Machine Vision System ◽  
Topographic Map ◽  
Correction Algorithm ◽  
Uav Navigation

The paper considers an original autonomous correction algorithm for UAV navigation system based on comparison between terrain images obtained by onboard machine vision system and vector topographic map images. Comparison is performed by calculating the homography of vision system images segmented using the convolutional neural network and the vector map images. The presented results of mathematical and flight experiments confirm the algorithm effectiveness for navigation applications.

Unmanned machine vision system for automated recognition of mechanical parts

2018 ◽  
Vol 6 (4) ◽  
pp. 184-196
Author(s):  
Tushar Jain ◽  
Meenu Gupta ◽  
H.K. Sardana
Keyword(s):  
Neural Network ◽  
Machine Vision ◽  
Classification Accuracy ◽  
Network Architecture ◽  
Vision System ◽  
Confusion Matrix ◽  
Machine Vision System ◽  
Content Type ◽  
Recognition Of Objects ◽  
Hidden Nodes

Purpose The field of machine vision, or computer vision, has been growing at fast pace. The growth in this field, unlike most established fields, has been both in breadth and depth of concepts and techniques. Machine vision techniques are being applied in areas ranging from medical imaging to remote sensing, industrial inspection to document processing and nanotechnology to multimedia databases. The goal of a machine vision system is to create a model of the real world from images. Computer vision recognition has attracted the attention of researchers in many application areas and has been used to solve many ranges of problems. The purpose of this paper is to consider recognition of objects manufactured in mechanical industry. Mechanically manufactured parts have recognition difficulties due to manufacturing process including machine malfunctioning, tool wear and variations in raw material. This paper considers the problem of recognizing and classifying the objects of such parts. RGB images of five objects are used as an input. The Fourier descriptor technique is used for recognition of objects. Artificial neural network (ANN) is used for classification of five different objects. These objects are kept in different orientations for invariant rotation, translation and scaling. The feed forward neural network with back-propagation learning algorithm is used to train the network. This paper shows the effect of different network architecture and numbers of hidden nodes on the classification accuracy of objects. Design/methodology/approach The overall goal of this research is to develop algorithms for feature-based recognition of 2D parts from intensity images. Most present industrial vision systems are custom-designed systems, which can only handle a specific application. This is not surprising, since different applications have different geometry, different reflectance properties of the parts. Findings Classification accuracy is affected by the changing network architecture. ANN is computationally demanding and slow. A total of 20 hidden nodes network structure produced the best results at 500 iterations (90 percent accuracy based on overall accuracy and 87.50 percent based on κ coefficient). So, 20 hidden nodes are selected for further analysis. The learning rate is set to 0.1, and momentum term used is 0.2 that give the best results architectures. The confusion matrix also shows the accuracy of the classifier. Hence, with these results the proposed system can be used efficiently for more objects. Originality/value After calculating the variation of overall accuracy with different network architectures, the results of different configuration of the sample size of 50 testing images are taken. Table II shows the results of the confusion matrix obtained on these testing samples of objects.

scholarly journals Artificial Neural Network Controlled GMAW System: Penetration and Quality Assurance in a Multi-Pass Butt Weld Application

2019 ◽  
Vol 105 (7-8) ◽  
pp. 3369-3385 ◽  
Author(s):  
Sakari Penttilä ◽  
Paul Kah ◽  
Juho Ratava ◽  
Harri Eskelinen
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Decision Making ◽  
Artificial Neural Network ◽  
Machine Vision ◽  
Vision System ◽  
Welding Parameter ◽  
Parameter Control ◽  
Machine Vision System ◽  
Profile Data

Abstract Intelligent welding parameter control is fast becoming a key instrument for attaining quality consistency in automated welding. Recent scientific breakthroughs in intelligent systems have turned the focus of adaptive welding control to artificial intelligence-based welding parameter control. The aim of this study is to combine artificial neural network (ANN) decision-making software and a machine vision system to develop an adaptive artificial intelligence (AI)-based gas metal arc welding (GMAW) parameter control system. The machine vision system uses a laser sensor to scan the upcoming seam and gather seam profile data. Based on further processing of the seam profile data, welding parameters are optimized by the decision-making system. In this work, the developed system is tested in a multivariable welding condition environment and its performance is evaluated. The quality of the welds was consistent and surpassed the required quality level. Additionally, the heat-affected zone (HAZ) was evaluated by microscopy, X-ray, and scanning electron microscope (SEM) imaging. It is concluded that the developed ANN system is suitable for implementation in automated applications, can improve quality consistency and cost efficiency, and reduce required workpiece preparation and handling.

Extracting External Features of Sweet Peppers Using Machine Vision System on Mobile Fruits Grading Robot

2012 ◽  
Vol 8 (3) ◽  
Author(s):  
Qiao Jun ◽  
Akira Sasao ◽  
Sakai Shibusawa ◽  
Naoshi Kondo
Keyword(s):  
Neural Network ◽  
Machine Vision ◽  
Vision System ◽  
Sweet Pepper ◽  
Maximum Diameter ◽  
Equivalent Diameter ◽  
Machine Vision System ◽  
Test Results ◽  
Agreement Rate ◽  
Fruit Mass

Abstract Algorithms to extract the external features of sweet peppers were developed using the machine vision system. The objectives were to sort color, estimate size, classify shape, detect bruises and predict mass of sweet peppers. A test was performed using 372 samples of sweet pepper variety “TosahikariD”. The results showed that the three main unacceptable colors were recognized and their percentages were calculated by the distribution of hue by saturation. The size of each sample was estimated by three parameters: Feret’s V, maximum diameter and equivalent diameter, and they ranged from 55.3mm to 116.7 mm, 23.2mm to 61.6mm and 21.8mm to 73.1mm, respectively. Five parameters were selected and a neural network model was developed to classify the shape of sweet peppers. Test results indicated that the agreement rate for Shapes A, B, and total samples was 95.12, 100 and 95.70%, respectively. All 11 samples with bruises were successfully detected. The fruit mass was predicted using projection areas; the determination and correlation coefficient was 0.93 and 0.96, respectively.

Sign in / Sign up

Export Citation Format

Share Document