scholarly journals Chart Classification Using Siamese CNN

2021 ◽  
Vol 7 (11) ◽  
pp. 220
Author(s):  
Filip Bajić ◽  
Josip Job
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Classification Accuracy ◽  
Support Vector ◽  
Network Architectures ◽  
Average Classification Accuracy ◽  
Multiple Network ◽  
Network Verification ◽  
Type Classification ◽  
Better Than

In recovering information from the chart image, the first step should be chart type classification. Throughout history, many approaches have been used, and some of them achieve results better than others. The latest articles are using a Support Vector Machine (SVM) in combination with a Convolutional Neural Network (CNN), which achieve almost perfect results with the datasets of few thousand images per class. The datasets containing chart images are primarily synthetic and lack real-world examples. To overcome the problem of small datasets, to our knowledge, this is the first report of using Siamese CNN architecture for chart type classification. Multiple network architectures are tested, and the results of different dataset sizes are compared. The network verification is conducted using Few-shot learning (FSL). Many of described advantages of Siamese CNNs are shown in examples. In the end, we show that the Siamese CNN can work with one image per class, and a 100% average classification accuracy is achieved with 50 images per class, where the CNN achieves only average classification accuracy of 43% for the same dataset.

A visual terrain classification method for mobile robots’ navigation based on convolutional neural network and support vector machine

2021 ◽  
pp. 014233122098791
Author(s):  
Wanli Wang ◽  
Botao Zhang ◽  
Kaiqi Wu ◽  
Sergey A Chepinskiy ◽  
Anton A Zhilenkov ◽  
...  
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Mobile Robots ◽  
Convolutional Neural Network ◽  
Hybrid Method ◽  
Classification Accuracy ◽  
Support Vector ◽  
High Classification Accuracy ◽  
Enhance Efficiency ◽  
Multi Class Classification

In this paper, a hybrid method based on deep learning is proposed to visually classify terrains encountered by mobile robots. Considering the limited computing resource on mobile robots and the requirement for high classification accuracy, the proposed hybrid method combines a convolutional neural network with a support vector machine to keep a high classification accuracy while improve work efficiency. The key idea is that the convolutional neural network is used to finish a multi-class classification and simultaneously the support vector machine is used to make a two-class classification. The two-class classification performed by the support vector machine is aimed at one kind of terrain that users are mostly concerned with. Results of the two classifications will be consolidated to get the final classification result. The convolutional neural network used in this method is modified for the on-board usage of mobile robots. In order to enhance efficiency, the convolutional neural network has a simple architecture. The convolutional neural network and the support vector machine are trained and tested by using RGB images of six kinds of common terrains. Experimental results demonstrate that this method can help robots classify terrains accurately and efficiently. Therefore, the proposed method has a significant potential for being applied to the on-board usage of mobile robots.

Application of Support Vector Machine in Springback Forecasting for Steel Structure

2012 ◽  
Vol 166-169 ◽  
pp. 1366-1369
Author(s):  
Jian Guo Chen ◽  
Zhao Guang Li
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Friction Coefficient ◽  
Bp Neural Network ◽  
Prediction Error ◽  
Steel Structure ◽  
Support Vector ◽  
Springback Prediction ◽  
Better Than

Support vector machine is applied to springback forecasting for steel structure in the paper. In the steel structure, pressure-pad-force, friction coefficient and die filleted corner have a certain influence on springback amount.We employ BP neural network to compare with support vector machine to show the superiority of support vector machine in this study. Finally,we give the comparison of the prediction error of springback for steel structure between support vector machine and BP neural network. Evidently,the springback prediction for steel structure of support vector machine is better than that of BP neural network.

Steel Pip Corrosion Forecasting Based on Support Vector Machine

2012 ◽  
Vol 166-169 ◽  
pp. 1002-1006
Author(s):  
Guang Yue Ma
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Bp Neural Network ◽  
Learning Algorithm ◽  
Support Vector ◽  
Risk Minimization ◽  
Prediction Ability ◽  
Structural Risk ◽  
Better Than ◽  
Local Extreme

BP neural network has some shorcomings,such as local extreme. Support vector machine is a novel statistical learning algorithm,which is based on the principle of structural risk minimization. In the paper, support vector machine is used to perform steel pip corrosion forecasting.The collected steel pip corrosion forecasting experimental data are given,among which corrosion deeps from 8ths to 11ths are used to test the proposed prediction model. BP neural network is applied to steel pip corrosion deep forecasting,which is used to compare with support vector machine to show the superiority of support vector machine in steel pip corrosion forecasting.The comparison of the prediction error of steel pip corrosion deep between support vector machine and BP neural network is given. It can be seen that the prediction ability for steel pip corrosion deep of support vector machine is better than that of BP neural network

Study on Detection Method for Crack in Eggs Based on Computer Vision and Support Vector Machine Neural Network

2014 ◽  
Vol 472 ◽  
pp. 176-179 ◽  
Author(s):  
Jian Yang ◽  
Ying Shi ◽  
Wei Zhou ◽  
Yong Shun Che
Keyword(s):  
Neural Network ◽  
Computer Vision ◽  
Support Vector Machine ◽  
Traditional Method ◽  
Detection Method ◽  
Support Vector ◽  
Characteristic Parameters ◽  
Sample Data ◽  
Better Than

To improve the accuracy of detection and classification of egg with cracks, this paper is to add Support Vector Machine to neural network to automatically identify and classify the eggs with cracks. Firstly process the egg images with light-transmitting were obtained by the computer vision device including denoising, threshold segmentation. Five characteristic parameters of crack areas and noise areas were acquired. Secondly train SVM Neural Network and identify the eggs with cracks by five parameters data as the sample data. The correct discerning rate of grading table eggs is 98.07%. It proves better than traditional method in terms of prediction accuracy and robustness. The generalization ability of SVM Neural Network is strengthened.

scholarly journals Protein Structure Classification Based on Distance Feature

2020 ◽  
pp. 263-269
Author(s):  
Sheshang Degadwala ◽  
Dhairya Vyas ◽  
Harsh S Dave
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Support Vector Machine ◽  
Protein Structure ◽  
Random Forest ◽  
Machine Intelligence ◽  
Support Vector ◽  
Bayes Classifier ◽  
Protein Structure Classification ◽  
Better Than

In Bioinformatics field Protein Structure Classification is the hugest undertaking. The realized proteins have been requested subject to their level, feature, work, amino destructive and family and superfamily. Protein structure segregated into four sorts: all ? protein structure, all ? protein structure, ?+? protein structure, and ?/? protein structure. The use of a standard way to deal with perform plan is a very inconvenient and dreary task. The quantity of cutting edge Machine Intelligence enrolling strategies such Support Vector Machine, Random Forest, Artificial Neural Network, Decision Tree and Naïve Bayes Classifier had been proposed in the composition. Our objective right currently is to develop a system that performs better than anything past markers for protein structure gathering by thinking about the separation among the distinctive Amino Acid buildups. To take a gander at the display of proposed work particular datasets are used.

scholarly journals An Artificial Intelligence Method for Asphalt Pavement Pothole Detection Using Least Squares Support Vector Machine and Neural Network with Steerable Filter-Based Feature Extraction

2018 ◽  
Vol 2018 ◽  
pp. 1-12 ◽  
Author(s):  
Nhat-Duc Hoang
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Support Vector Machine ◽  
Least Squares ◽  
Classification Accuracy ◽  
Asphalt Pavement ◽  
Support Vector ◽  
Accuracy Rate ◽  
Pothole Detection ◽  
Steerable Filter

This study establishes an artificial intelligence (AI) model for detecting pothole on asphalt pavement surface. Image processing methods including Gaussian filter, steerable filter, and integral projection are utilized for extracting features from digital images. A data set consisting of 200 image samples has been collected to train and validate the predictive performance of two machine learning algorithms including the least squares support vector machine (LS-SVM) and the artificial neural network (ANN). Experimental results obtained from a repeated subsampling process with 20 runs show that both LS-SVM and ANN are capable methods for pothole detection with classification accuracy rate larger than 85%. In addition, the LS-SVM has achieved the highest classification accuracy rate (roughly 89%) and the area under the curve (0.96). Accordingly, the proposed AI approach used with LS-SVM can be very potential to assist transportation agencies and road inspectors in the task of pavement pothole detection.

A Wavelet Kernel-Based Support Vector Machine for Communication Network Intrusion Detection

2014 ◽  
Vol 989-994 ◽  
pp. 4474-4477
Author(s):  
Ying Zhan
Keyword(s):  
Neural Network ◽  
Communication Network ◽  
Support Vector Machine ◽  
Intrusion Detection ◽  
Bp Neural Network ◽  
Network Intrusion Detection ◽  
Support Vector ◽  
Detection Accuracy ◽  
Network Intrusion ◽  
Better Than

This study is to propose a wavelet kernel-based support vector machine (SVM) for communication network intrusion detection. The common intrusion types of communication network mainly include DOS, R2L, U2R and Probing. SVM, BP neural network are used to compare with the proposed wavelet kernel-based SVM method to show the superiority of wavelet kernel-based SVM. The detection accuracy for communication network intrusion of wavelet kernel-based SVM is 96.67 %, the detection accuracy for communication network intrusion of SVM is 90.83%, and the detection accuracy for communication network intrusion of BP neural network is 86.67%.It can be seen that the detection accuracy for communication network intrusion of wavelet kernel-based SVM is better than that of SVM or BP neural network.

scholarly journals A CAD System for the Early Detection of Lung Nodules Using Computed Tomography Scan Images

2019 ◽  
Vol 15 (04) ◽  
pp. 40
Author(s):  
Hanan M. Amer ◽  
Fatma E. Abou-Chadi ◽  
Sherif S. Kishk ◽  
Marwa I. Obayya
Keyword(s):  
Neural Network ◽  
Computed Tomography ◽  
Lung Cancer ◽  
Support Vector Machine ◽  
Classification Accuracy ◽  
Support Vector ◽  
Lung Nodules ◽  
Computed Tomography Scan ◽  
Accuracy Rate ◽  
Tomography Scan

<p>In this paper,  a computer-aided detection system is developed to detect lung nodules at an early stage using Computed Tomography (CT) scan images where lung nodules are one of the most important indicators to predict lung cancer. The developed system consists of four stages. First, the raw Computed Tomography lung  images were preprocessed to enhance the image contrast and eliminate noise. Second, an automatic segmentation procedure for human's lung and pulmonary nodule canddates (nodules, blood vessels) using a two-level thresholding technique and morphological operations. Third, a feature fusion technique that fuses four feature extraction techniques: the statistical features of first and second order, value histogram features, histogram of oriented gradients features, and texture features of gray level co-occurrence matrix based on wavelet coefficients was utilised to extract the main features. The fourth stage is the classifier. Three classifiers were used and their performance was compared in order to obtain the highest classification accuracy. These are; multi-layer feed-forward neural network, radial basis function neural network and support vector machine. The  performance of the proposed system was assessed using three quantitative parameters. These are: the classification accuracy rate, the sensitivity and the specificity. Forty standard computed tomography images containing 320 regions of interest obtained from an early lung cancer action project association were used to test and evaluate the developed system. The images consists of 40 computed tomography scan images. The results have shown that the fused features vector resulting from genetic algorithm as a feature selection technique and the support vector machine classifier give the highest classification accuracy rate, sensitivity and specificity values of 99.6%, 100% and 99.2%, respectively.</p>

Diabetes prediction model based on deep belief network

2020 ◽  
pp. 1-12
Author(s):  
Li-Ying Lang ◽  
Zheng Gao ◽  
Xue-Guang Wang ◽  
Hui Zhao ◽  
Yan-Ping Zhang ◽  
...  
Keyword(s):  
Support Vector Machine ◽  
Logistic Regression ◽  
Classification Accuracy ◽  
Support Vector ◽  
Hospital Data ◽  
Related Factors ◽  
Diabetes Prediction ◽  
Logistic Regression Algorithm ◽  
Complications And Mortality ◽  
Better Than

Diabetes is a disease that seriously endangers human health. Early detection and early treatment can reduce the likelihood of complications and mortality. The predictive model can effectively solve the above problems and provide helpful information for the clinic. Based on this, it is proposed to apply the idea of integrated algorithm in DBN algorithm, collect the hospital data by investigating its related factors, clean and process the collected data, and sample and model the processed data multiple times. It is shown that a single DBN classifier is better than support vector machine and logistic regression algorithm. The model established by the integrated deep confidence network has a significant improvement in classification accuracy compared to a single DBN classifier, and solves the unstable classification effect of a single DBN classifier.

Sign in / Sign up

Export Citation Format

Share Document