Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729354
Mostafa Farrokhi Afsharyan, M. Hoseinzade
Brain Computer Interfaces (BCI) translating brain wave signals into practical commands to operate external devices by which augment human capabilities. However, many issues face the development of BCIs such as how to extract commands from EEGs due to the low signal-to-noise ratio (SNR) of EEG signals. This paper investigates an EEG-driven hardware-in-loop (HIL) experimental robot for BCI stimulation system individualized design and validation. Based on power spectrum data collected in real-time by the two TGAM electrodes, we developed a novel BCI stimulation system that allows us to adjust robot navigation. By using the SVM model, the EEG signals are preprocessed and converted into mental commands (e.g. forward, left …) to navigate the simulated robot. The average accuracy of the robot movement was 62.6%, which obtained Cohen's Kappa coefficient are significantly better than chance (κ = 0.50). Our results showed that the robot control can be achieved with reduced accuracy under the respective experimental conditions in a simulation environment.
{"title":"A Hardware in Loop Simulation Robot Control by Weareable Electroencephalography (EEG)-Based Brain Computer Interface (BCI)","authors":"Mostafa Farrokhi Afsharyan, M. Hoseinzade","doi":"10.1109/ICSPIS54653.2021.9729354","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729354","url":null,"abstract":"Brain Computer Interfaces (BCI) translating brain wave signals into practical commands to operate external devices by which augment human capabilities. However, many issues face the development of BCIs such as how to extract commands from EEGs due to the low signal-to-noise ratio (SNR) of EEG signals. This paper investigates an EEG-driven hardware-in-loop (HIL) experimental robot for BCI stimulation system individualized design and validation. Based on power spectrum data collected in real-time by the two TGAM electrodes, we developed a novel BCI stimulation system that allows us to adjust robot navigation. By using the SVM model, the EEG signals are preprocessed and converted into mental commands (e.g. forward, left …) to navigate the simulated robot. The average accuracy of the robot movement was 62.6%, which obtained Cohen's Kappa coefficient are significantly better than chance (κ = 0.50). Our results showed that the robot control can be achieved with reduced accuracy under the respective experimental conditions in a simulation environment.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129115982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729360
S. Fadaei
Content-based image retrieval (CBIR) is one of the most applicable image processing techniques which includes two main steps: feature extraction and retrieval. A feature vector related to visual contents of image is extracted from the image in the feature extraction step. Three set features color, texture and shape are extracted from image in typical CBIR systems. Dominant color descriptor (DCD) is a method based on color information of the image. There are many color spaces to represent an image, so DCD can be implemented in any of these color spaces. In this paper color spaces RGB, CMY, HSV, CIE Lab, CIE Luv and HMMD are considered and effect of them in DCD features is investigated. Also, the CBIR precision is affected by the number of partitions in DCD method which is analyzed in this paper. Simulation results on Corel-1k dataset show that the HSV color space achieves better precision comparing the other color spaces.
{"title":"Comparision of color spaces in DCD-based content-based image retrieval systems","authors":"S. Fadaei","doi":"10.1109/ICSPIS54653.2021.9729360","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729360","url":null,"abstract":"Content-based image retrieval (CBIR) is one of the most applicable image processing techniques which includes two main steps: feature extraction and retrieval. A feature vector related to visual contents of image is extracted from the image in the feature extraction step. Three set features color, texture and shape are extracted from image in typical CBIR systems. Dominant color descriptor (DCD) is a method based on color information of the image. There are many color spaces to represent an image, so DCD can be implemented in any of these color spaces. In this paper color spaces RGB, CMY, HSV, CIE Lab, CIE Luv and HMMD are considered and effect of them in DCD features is investigated. Also, the CBIR precision is affected by the number of partitions in DCD method which is analyzed in this paper. Simulation results on Corel-1k dataset show that the HSV color space achieves better precision comparing the other color spaces.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122451081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729370
Asma Bahrani, Babak Majidi, M. Eshghi
In recent years, advancement of the industry and increased human activities created significant pollution in the marine environment and coastal regions of the Persian Gulf. These pollutions cause various diseases and serious damages to the human health and animal species. Early identification of various pollutions helps the coastal management to organize their resources and rapidly respond to the problems. Due to the large scale of the coastal regions, manual investigation of the pollutions is a very time-consuming task. Unmanned robots can be used as autonomous agents for rapid large-scale detection and classification of pollutions in the coastal regions. In this paper, an artificial intelligence-based vision system for autonomous marine pollution detection is proposed. A combination of computer vision and machine learning methods are used for autonomous detection of various pollutions in the coastal and marine environment. In this study, 3000 images of Persian Gulf coastal pollutions is collected and used for training an artificial intelligence system for coastal conservation. The experimental results shows that the proposed framework has a 98% accuracy for identifying and classifying coastal and marine pollutions. The proposed system can be used as the vision system of an autonomous coastal conservation robot and increase the speed of coastal conservation and management significantly.
{"title":"Autonomous oil spill and pollution detection for large-scale conservation in marine eco-cyber-physical systems","authors":"Asma Bahrani, Babak Majidi, M. Eshghi","doi":"10.1109/ICSPIS54653.2021.9729370","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729370","url":null,"abstract":"In recent years, advancement of the industry and increased human activities created significant pollution in the marine environment and coastal regions of the Persian Gulf. These pollutions cause various diseases and serious damages to the human health and animal species. Early identification of various pollutions helps the coastal management to organize their resources and rapidly respond to the problems. Due to the large scale of the coastal regions, manual investigation of the pollutions is a very time-consuming task. Unmanned robots can be used as autonomous agents for rapid large-scale detection and classification of pollutions in the coastal regions. In this paper, an artificial intelligence-based vision system for autonomous marine pollution detection is proposed. A combination of computer vision and machine learning methods are used for autonomous detection of various pollutions in the coastal and marine environment. In this study, 3000 images of Persian Gulf coastal pollutions is collected and used for training an artificial intelligence system for coastal conservation. The experimental results shows that the proposed framework has a 98% accuracy for identifying and classifying coastal and marine pollutions. The proposed system can be used as the vision system of an autonomous coastal conservation robot and increase the speed of coastal conservation and management significantly.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130653954","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729382
Alireza Taheri Dehkordi, Hani Ghasemi, M. J. V. Zoej
Estimation of Suspended Sediment Concentration (SSC), regarded as a crucial component of hydrological and ecological processes, can provide a better understanding of water quality. This study aims to use Sentinel-2 (S2) level-2A (L2A) images with less than 1% cloud coverage and supervised machine learning-based regression models to estimate SSC along the Missouri River. The model gets the reflectance values of different spectral bands and predicts the corresponding SSC value for each water pixel. Time-series data of three different ground measuring stations and surface reflectance values of the closest pixel to each station are used to train and validate the model. Two popular regression models, Support Vector Regression (SVR) and Random Forests (RF), are trained, validated, and compared online in the Google Earth Engine (GEE) processing platform by using 68 satellite images, without the need to be downloaded. The results demonstrated that the RF model with a root mean square error (RMSE) and mean absolute error (MAE) of 59.521 and 46.493 mg/L outperforms the SVR model. Moreover, the RF model resulted in a higher correlation between the real and predicted SSC values (R2 = 0.79 and Pearson's r = 0.877). Also, the two short wave infra-red (SWIR) and red bands play more considerable roles in SSC estimation using S2L2A images than other bands.
{"title":"Machine Learning-Based Estimation of Suspended Sediment Concentration along Missouri River using Remote Sensing Imageries in Google Earth Engine","authors":"Alireza Taheri Dehkordi, Hani Ghasemi, M. J. V. Zoej","doi":"10.1109/ICSPIS54653.2021.9729382","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729382","url":null,"abstract":"Estimation of Suspended Sediment Concentration (SSC), regarded as a crucial component of hydrological and ecological processes, can provide a better understanding of water quality. This study aims to use Sentinel-2 (S2) level-2A (L2A) images with less than 1% cloud coverage and supervised machine learning-based regression models to estimate SSC along the Missouri River. The model gets the reflectance values of different spectral bands and predicts the corresponding SSC value for each water pixel. Time-series data of three different ground measuring stations and surface reflectance values of the closest pixel to each station are used to train and validate the model. Two popular regression models, Support Vector Regression (SVR) and Random Forests (RF), are trained, validated, and compared online in the Google Earth Engine (GEE) processing platform by using 68 satellite images, without the need to be downloaded. The results demonstrated that the RF model with a root mean square error (RMSE) and mean absolute error (MAE) of 59.521 and 46.493 mg/L outperforms the SVR model. Moreover, the RF model resulted in a higher correlation between the real and predicted SSC values (R2 = 0.79 and Pearson's r = 0.877). Also, the two short wave infra-red (SWIR) and red bands play more considerable roles in SSC estimation using S2L2A images than other bands.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133253156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729368
Hossein Khaleghy, I. Izadi
Industrial alarm systems have recently progressed considerably in terms of network complexity and the number of alarms. The increase in complexity and number of alarms presents challenges in these systems that decrease system efficiency and cause distrust of the operator, which might result in widespread damages. One contributing factor in alarm inefficiency is the correlated alarms. These alarms do not contain new information and only confuse the operator. This paper tries to present a novel method for detecting correlated alarms based on artificial intelligence methods to help the operator. The proposed method is based on graph embedding and alarm clustering, resulting in the detection of correlated alarms. To evaluate the proposed method, a case study is conducted on the well-known Tennessee-Eastman process.
{"title":"Detection of Correlated Alarms Using Graph Embedding","authors":"Hossein Khaleghy, I. Izadi","doi":"10.1109/ICSPIS54653.2021.9729368","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729368","url":null,"abstract":"Industrial alarm systems have recently progressed considerably in terms of network complexity and the number of alarms. The increase in complexity and number of alarms presents challenges in these systems that decrease system efficiency and cause distrust of the operator, which might result in widespread damages. One contributing factor in alarm inefficiency is the correlated alarms. These alarms do not contain new information and only confuse the operator. This paper tries to present a novel method for detecting correlated alarms based on artificial intelligence methods to help the operator. The proposed method is based on graph embedding and alarm clustering, resulting in the detection of correlated alarms. To evaluate the proposed method, a case study is conducted on the well-known Tennessee-Eastman process.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130085225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729341
Fateme Amjadipour, H. Ghassemian, M. Imani
Nowadays, one of the most challenging issues in the field of remote sensing and satellite imagery is production of cadastral maps. Synthetic aperture radar (SAR) images are among the most widely used satellite images in the last decade. Due to radar nature of these images, the buildings in SAR images face with two problems: shadow and layover. Morphological mathematics are efficient tools for detection of buildings with providing a contextual profile containing shape and geometrical characteristics of the objects in radar images. By using the suggested method, two characteristics of shadow and brightness are detected separately. Then, the construction areas are extracted by using a fuzzy fusion approach. In this method, various parameters such as size and direction of the structural element and the weighting factor of the shadow, bright area, and the recursive parameter have to be determined independently. To this end, an iterative method using MSE is suggested. The experimental results show a detection rate of 94.3% achieved by the proposed method.
{"title":"Estimation of Free Parameters of Morphological Profiles for Building Extraction Using SAR Images","authors":"Fateme Amjadipour, H. Ghassemian, M. Imani","doi":"10.1109/ICSPIS54653.2021.9729341","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729341","url":null,"abstract":"Nowadays, one of the most challenging issues in the field of remote sensing and satellite imagery is production of cadastral maps. Synthetic aperture radar (SAR) images are among the most widely used satellite images in the last decade. Due to radar nature of these images, the buildings in SAR images face with two problems: shadow and layover. Morphological mathematics are efficient tools for detection of buildings with providing a contextual profile containing shape and geometrical characteristics of the objects in radar images. By using the suggested method, two characteristics of shadow and brightness are detected separately. Then, the construction areas are extracted by using a fuzzy fusion approach. In this method, various parameters such as size and direction of the structural element and the weighting factor of the shadow, bright area, and the recursive parameter have to be determined independently. To this end, an iterative method using MSE is suggested. The experimental results show a detection rate of 94.3% achieved by the proposed method.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131095584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729355
Zahra Hemmati, G. Mirjalily, Zahra Mohtajollah
The centralized structure of software defined networks makes them vulnerable to distributed denial of service attacks. Given that these attacks can easily destroy the computational and communicational resources of controller and switches, they make the network fail in a short time. Hence, it is vital to protect the controller. Utilizing the unique features of software defined networks, this paper propounds an effective method to detect distributed denial of services attacks. For this purpose, entropy was used to detect attacks. Furthermore, this method utilizes a dynamic threshold instead of a static one to distinguish between normal and attack traffic. The dynamic threshold heightens the accuracy of attack detection in the proposed algorithm to 98% on average while the accuracy in the benchmark algorithm using entropy and the static threshold is 96%.
{"title":"Entropy-based DDoS Attack Detection in SDN using Dynamic Threshold","authors":"Zahra Hemmati, G. Mirjalily, Zahra Mohtajollah","doi":"10.1109/ICSPIS54653.2021.9729355","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729355","url":null,"abstract":"The centralized structure of software defined networks makes them vulnerable to distributed denial of service attacks. Given that these attacks can easily destroy the computational and communicational resources of controller and switches, they make the network fail in a short time. Hence, it is vital to protect the controller. Utilizing the unique features of software defined networks, this paper propounds an effective method to detect distributed denial of services attacks. For this purpose, entropy was used to detect attacks. Furthermore, this method utilizes a dynamic threshold instead of a static one to distinguish between normal and attack traffic. The dynamic threshold heightens the accuracy of attack detection in the proposed algorithm to 98% on average while the accuracy in the benchmark algorithm using entropy and the static threshold is 96%.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"147 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122506819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729390
Mohammad Dehghani, Diyana Tehrany Dehkordy, M. Bahrani
Regarding the development of the web and increasing user interaction, different users' opinions about different phenomena have been observed. In recent years, the detection of Abusive language in online content used by users has become a necessity. Twitter is a platform in which users can share text messages. On Twitter, different people express their opinion on different topics with different kinds of literature, some of which are accompanied by Abusive words. On the one hand, Abusive comments can be derogatory and harmful to those who share content. On the other hand, filtering these comments in languages other than English is difficult and time-consuming. Most social media platforms are still looking for more efficient ways to filter comments because the manual method is expensive, slow, and risky. Automating helps better identify and filter Abusive comments and increase user safety. In the present article, a deep learning method is presented to detect users' Abusive words in Persian tweets. Due to the lack of appropriate data in Persian, we created a database of 33338 Persian tweets, of which 10% contained Abusive words and 90% were non-Abusive. Perhaps the easiest way is to use a fixed list and filter comments. So, a list of 648 Abusive words in Persian was prepared and used to test the database (accuracy of 76%). Finally, a deep neural network is implemented to detect Abusive words using the Bert language model, and it had the best performance with an accuracy of 97.7%.
{"title":"Abusive words Detection in Persian tweets using machine learning and deep learning techniques","authors":"Mohammad Dehghani, Diyana Tehrany Dehkordy, M. Bahrani","doi":"10.1109/ICSPIS54653.2021.9729390","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729390","url":null,"abstract":"Regarding the development of the web and increasing user interaction, different users' opinions about different phenomena have been observed. In recent years, the detection of Abusive language in online content used by users has become a necessity. Twitter is a platform in which users can share text messages. On Twitter, different people express their opinion on different topics with different kinds of literature, some of which are accompanied by Abusive words. On the one hand, Abusive comments can be derogatory and harmful to those who share content. On the other hand, filtering these comments in languages other than English is difficult and time-consuming. Most social media platforms are still looking for more efficient ways to filter comments because the manual method is expensive, slow, and risky. Automating helps better identify and filter Abusive comments and increase user safety. In the present article, a deep learning method is presented to detect users' Abusive words in Persian tweets. Due to the lack of appropriate data in Persian, we created a database of 33338 Persian tweets, of which 10% contained Abusive words and 90% were non-Abusive. Perhaps the easiest way is to use a fixed list and filter comments. So, a list of 648 Abusive words in Persian was prepared and used to test the database (accuracy of 76%). Finally, a deep neural network is implemented to detect Abusive words using the Bert language model, and it had the best performance with an accuracy of 97.7%.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123708331","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-29DOI: 10.1109/ICSPIS54653.2021.9729340
Mohamad Ebrahim Aghili, H. Ghassemian, M. Imani
Hyperspectral image (HSI) classification is one of the most important applications among all types of classification fields. Proper classification of spectral data leads to discovery of important land covers. In recent years, many methods have been introduced to increase the HSI classification accuracy. Methods based on neural networks show superior results compared to other methods. Among them, the two-dimensional convolutional neural networks (2D-CNNs) inspired by the human eye retina have achieved higher accuracy in classification. In most cases, HSI classifiers use only spectral features. In this paper, the spectral-spatial feature fusion and HSI classification using 2D-CNN are focused. For this purpose, the first 2D-convolutional layer of CNN is substituted by two combined 2D-Gabor-Shapelet filter banks. This layer extracts contextual information and provides valuable joint spectral-spatial features. The experimental results on real HSI (including the urban and agricultural areas and their mixture) show that the proposed method improves the overall classification performance. Compared to several famous HSI classification based on neural networks, the proposed method has higher speed and classification accuracy.
{"title":"A New Spectral-Spatial Network for Feature Fusion and Classification of Hyperspectral Images","authors":"Mohamad Ebrahim Aghili, H. Ghassemian, M. Imani","doi":"10.1109/ICSPIS54653.2021.9729340","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729340","url":null,"abstract":"Hyperspectral image (HSI) classification is one of the most important applications among all types of classification fields. Proper classification of spectral data leads to discovery of important land covers. In recent years, many methods have been introduced to increase the HSI classification accuracy. Methods based on neural networks show superior results compared to other methods. Among them, the two-dimensional convolutional neural networks (2D-CNNs) inspired by the human eye retina have achieved higher accuracy in classification. In most cases, HSI classifiers use only spectral features. In this paper, the spectral-spatial feature fusion and HSI classification using 2D-CNN are focused. For this purpose, the first 2D-convolutional layer of CNN is substituted by two combined 2D-Gabor-Shapelet filter banks. This layer extracts contextual information and provides valuable joint spectral-spatial features. The experimental results on real HSI (including the urban and agricultural areas and their mixture) show that the proposed method improves the overall classification performance. Compared to several famous HSI classification based on neural networks, the proposed method has higher speed and classification accuracy.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125591659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-09DOI: 10.1109/ICSPIS54653.2021.9729356
Ali Mobaien, Arman Kheirati Roonizi, R. Boostani
In this work, we present a powerful notch filter for power-line interference (PLI) cancelation from biomedical signals. This filter has a unit gain and a zero-phase response. Moreover, the filter can be implemented adaptively to adjust its bandwidth based on the signal-to-noise ratio. To realize this filter, a dynamic model is defined for PLI based on its sinusoid property. Then, a constrained least square error estimation is used to emerge the PLI based on the observations while the constraint is the PLI dynamic. At last, the estimated PLI is subtracted from recordings. The proposed filter is assessed using synthetic data and real biomedical recordings in different noise levels. The results demonstrate this filter as a very powerful and effective means for canceling the PLI out.
{"title":"A powerful notch filter for PLI cancelation","authors":"Ali Mobaien, Arman Kheirati Roonizi, R. Boostani","doi":"10.1109/ICSPIS54653.2021.9729356","DOIUrl":"https://doi.org/10.1109/ICSPIS54653.2021.9729356","url":null,"abstract":"In this work, we present a powerful notch filter for power-line interference (PLI) cancelation from biomedical signals. This filter has a unit gain and a zero-phase response. Moreover, the filter can be implemented adaptively to adjust its bandwidth based on the signal-to-noise ratio. To realize this filter, a dynamic model is defined for PLI based on its sinusoid property. Then, a constrained least square error estimation is used to emerge the PLI based on the observations while the constraint is the PLI dynamic. At last, the estimated PLI is subtracted from recordings. The proposed filter is assessed using synthetic data and real biomedical recordings in different noise levels. The results demonstrate this filter as a very powerful and effective means for canceling the PLI out.","PeriodicalId":286966,"journal":{"name":"2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124947476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：