Denoising and feature extraction of weld seam profiles by stacked denoising autoencoder

2021 ◽  
Author(s):  
Ran Li ◽  
Hongming Gao
Keyword(s):  
Feature Extraction ◽  
Weld Seam ◽  
Denoising Autoencoder

Low-level structure feature extraction for image processing via stacked sparse denoising autoencoder

2017 ◽  
Vol 243 ◽  
pp. 12-20 ◽  
Author(s):  
Zunlin Fan ◽  
Duyan Bi ◽  
Linyuan He ◽  
Ma Shiping ◽  
Shan Gao ◽  
...  
Keyword(s):  
Image Processing ◽  
Feature Extraction ◽  
Level Structure ◽  
Denoising Autoencoder ◽  
Low Level ◽  
Structure Feature

scholarly journals Denoising Autoencoder-Based Feature Extraction to Robust SSVEP-Based BCIs

Sensors ◽  
2021 ◽  
Vol 21 (15) ◽  
pp. 5019
Author(s):  
Yeou-Jiunn Chen ◽  
Pei-Chung Chen ◽  
Shih-Chung Chen ◽  
Chung-Min Wu
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Deep Neural Network ◽  
Evoked Potential ◽  
Noise Suppression ◽  
Denoising Autoencoder ◽  
Practical Applications ◽  
Computer Interfaces ◽  
Recognition Result ◽  
Lateral Sclerosis

For subjects with amyotrophic lateral sclerosis (ALS), the verbal and nonverbal communication is greatly impaired. Steady state visually evoked potential (SSVEP)-based brain computer interfaces (BCIs) is one of successful alternative augmentative communications to help subjects with ALS communicate with others or devices. For practical applications, the performance of SSVEP-based BCIs is severely reduced by the effects of noises. Therefore, developing robust SSVEP-based BCIs is very important to help subjects communicate with others or devices. In this study, a noise suppression-based feature extraction and deep neural network are proposed to develop a robust SSVEP-based BCI. To suppress the effects of noises, a denoising autoencoder is proposed to extract the denoising features. To obtain an acceptable recognition result for practical applications, the deep neural network is used to find the decision results of SSVEP-based BCIs. The experimental results showed that the proposed approaches can effectively suppress the effects of noises and the performance of SSVEP-based BCIs can be greatly improved. Besides, the deep neural network outperforms other approaches. Therefore, the proposed robust SSVEP-based BCI is very useful for practical applications.

scholarly journals Multiscale Feature Extraction and its Application in the Weld Seam Quality Prediction for Plasma Arc Welding

2021 ◽  
Author(s):  
Hao Dong ◽  
Cai Yan ◽  
Zihan Li ◽  
Xueming Hua
Keyword(s):  
Feature Extraction ◽  
Arc Welding ◽  
Weld Seam ◽  
Welding Process ◽  
Welding Current ◽  
Support Vector ◽  
Quality Prediction ◽  
Plasma Arc Welding ◽  
Plasma Arc ◽  
Feature Extraction Method

Abstract Weld seam quality prediction is important for intelligent robot welding. Current models with single scale feature extraction methods meet difficulty when facing complex physical instability in welding process. In this paper, a novel feature extraction method based on sliding multiscale windows is proposed to improve model accuracy and calculation speed. A group of windows with different width are established to extract multiscale information of complex objective. Windows slide throughout process and be synchronized on the timeline for feature correlation. Based on the feature vector extracted from multiscale-windows, Support vector machine (SVM) with radial basis function (RBF) kernel is used after signal denoising and dimension reduction by Primary components analysis (PCA). The best window width is determined by model training. The proposed method is used to predict seam quality for Plasma arc welding (PAW) in the field of shipbuilding. The results show that the model with multiscale feature extraction is helpful to improve prediction precision and recall ratio.

Thin Laser Weld Seam Feature Extraction on Industrial Production Line

2011 ◽  
Vol 143-144 ◽  
pp. 194-198
Author(s):  
Li Wei Wang ◽  
He Xu Sun ◽  
Hai Yong Chen
Keyword(s):  
Image Processing ◽  
Feature Extraction ◽  
Weld Seam ◽  
Structured Light ◽  
Ccd Camera ◽  
Image Plane ◽  
Processing Method ◽  
Laser Weld ◽  
Robust Method ◽  
Excellent Performance

The study proposes a robust method to extract the line structured light stripes in industrial environments. The line structured light stripe is projected onto the laser weld seam to be measured by a projector, and deformation of the stripe is detected by a CCD camera with industrial microscope lens. An image processing method that can efficiently locate the deformation of the stripe in the image plane is presented. Finally, the method is applied to two kinds of laser weld seam specimen applications, and excellent performance is shown by some experimental results.

scholarly journals Deep Denoising Autoencoding Method for Feature Extraction and Recognition of Vehicle Adhesion Status

2018 ◽  
Vol 2018 ◽  
pp. 1-8 ◽  
Author(s):  
Jing He ◽  
Linfan Liu ◽  
Changfan Zhang ◽  
Kaihui Zhao ◽  
Jian Sun ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Support Vector ◽  
Recognition Method ◽  
Denoising Autoencoder ◽  
Sparse Autoencoder ◽  
Learning Machine ◽  
Network Weight ◽  
Weight Learning ◽  
Status Data ◽  
Status Recognition

Feature extraction and classification for deep learning are studied to recognize the problem of vehicle adhesion status. Data concentration acquired by automobile sensors contains considerable noise. Thus, a sparse autoencoder (stacked denoising autoencoder) is introduced to achieve network weight learning, restore original pure signal data by use of overlapping convergence strategy, and construct multiclassification support vector machine (SVM) for classification. The sensors are adopted in different road environments to acquire data signals and recognize the adhesion status online. Results show that the proposed method can achieve higher accuracies than those of the adhesion status recognition method based on SVM and extreme learning machine.

Point cloud 3D parent surface reconstruction and weld seam feature extraction for robotic grinding path planning

2020 ◽  
Vol 107 (1-2) ◽  
pp. 827-841
Author(s):  
Xiangfei Wang ◽  
Xiaoqiang Zhang ◽  
Xukai Ren ◽  
Lufeng Li ◽  
Hengjian Feng ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Path Planning ◽  
Surface Reconstruction ◽  
Point Cloud ◽  
Weld Seam ◽  
Robotic Grinding

scholarly journals A Robust Wi-Fi Fingerprint Positioning Algorithm Using Stacked Denoising Autoencoder and Multi-Layer Perceptron

2019 ◽  
Vol 11 (11) ◽  
pp. 1293 ◽  
Author(s):  
Rongrong Wang ◽  
Zhaohui Li ◽  
Haiyong Luo ◽  
Fang Zhao ◽  
Wenhua Shao ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Regression Model ◽  
Extraction Method ◽  
Large Time ◽  
Training Dataset ◽  
Time Interval ◽  
Multi Layer Perceptron ◽  
Denoising Autoencoder ◽  
Feature Extraction Method ◽  
Testing Dataset

With the increasing demand for location-based services, Wi-Fi-based indoor positioning technology has attracted much attention in recent years because of its ubiquitous deployment and low cost. Considering that Wi-Fi signals fluctuate greatly with time, extracting robust features of Wi-Fi signals is the key point to maintaining good positioning accuracy. To handle the dynamic fluctuation with time and sparsity of Wi-Fi signals, we propose an SDAE (Stacked Denoising Autoencoder)-based feature extraction method, which can obtain a robust and time-independent Wi-Fi fingerprint by learning the reconstruction distribution from a raw Wi-Fi signal and an artificial-noise-added Wi-Fi signal. We also leverage the strong representation ability of MLP (Multi-Layer Perceptron) to build a regression model, which maps the extracted features to the corresponding location. To fully evaluate the performance of our proposed algorithm, three datasets are applied, which represent three different scenarios, namely, spacious area with time interval, no time interval, and complex area with large time interval. The experimental results confirm the validity of our proposed SDAE-based feature extraction method, which can accurately reflect Wi-Fi signals in corresponding locations. Compared with other regression models, our proposed regression model can better map the extracted features to the target position. The average positioning error of our proposed algorithm is 4.24 m when there is a 52-day interval between training dataset and testing dataset. That confirms that the proposed algorithm outperforms other state-of-the-art positioning algorithms when there is a large time interval between training dataset and testing dataset.

Sign in / Sign up

Export Citation Format

Share Document