Pub Date : 2020-11-25DOI: 10.1109/ICSPIS51252.2020.9340135
M. Mathews, M. AnzarS., R. K. Krishnan, A. Panthakkan
Automated techniques for retinal vessel segmentation is an active research area for the past three decades. Features associated with retinal blood vessels like morphology, area, diameter, tortuosity are important to assess the onset and progression of many eye-related and cardiovascular diseases. For retinal vessel segmentation, we propose two deep neural networks: U-net with EfficientNet as the backbone and EfficientNet encoder with LinkNet decoder. Gamma adjustment and contrast limited histogram equalization is the pre-processing stages adopted. EfficientNetB3 with U-net provide significant improvement. Results are evaluated on benchmark fundus image datasets like DRIVE [1], STARE [2], HRF [3], and CHASE_DB1 [4]. The proposed architecture obtained 96.35% accuracy, 86.35% sensitivity, 97.67% specificity, and an F1 score of 0.8465 on the DRIVE dataset.
{"title":"EfficientNet for retinal blood vessel segmentation","authors":"M. Mathews, M. AnzarS., R. K. Krishnan, A. Panthakkan","doi":"10.1109/ICSPIS51252.2020.9340135","DOIUrl":"https://doi.org/10.1109/ICSPIS51252.2020.9340135","url":null,"abstract":"Automated techniques for retinal vessel segmentation is an active research area for the past three decades. Features associated with retinal blood vessels like morphology, area, diameter, tortuosity are important to assess the onset and progression of many eye-related and cardiovascular diseases. For retinal vessel segmentation, we propose two deep neural networks: U-net with EfficientNet as the backbone and EfficientNet encoder with LinkNet decoder. Gamma adjustment and contrast limited histogram equalization is the pre-processing stages adopted. EfficientNetB3 with U-net provide significant improvement. Results are evaluated on benchmark fundus image datasets like DRIVE [1], STARE [2], HRF [3], and CHASE_DB1 [4]. The proposed architecture obtained 96.35% accuracy, 86.35% sensitivity, 97.67% specificity, and an F1 score of 0.8465 on the DRIVE dataset.","PeriodicalId":373750,"journal":{"name":"2020 3rd International Conference on Signal Processing and Information Security (ICSPIS)","volume":"80 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125886835","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 : 2020-11-25DOI: 10.1109/ICSPIS51252.2020.9340127
S. Sani
Sleep apnea is a serious medical condition that influence millions of adults and children in the United States. It is unknown whether or not sleep apnea can cause any abnormalities in cardiovascular variability in patients with no cardiovascular disease. In this study, we investigated new cardiovascular variables which correspond with obstructive sleep apnea. The power spectrum of the electrocardiogram signal (ECG) was studied in healthy aged-matched adults $boldsymbol{(mathrm{n}=15)}$ and patients diagnosed with moderate obstructive sleep apnea $boldsymbol{(mathrm{n}=15)}$. The power spectral analysis was divided into three main frequency bands: very low frequency $boldsymbol{(0.0033sim 0.04mathrm{Hz})}$; low frequency $boldsymbol{(0.04sim 0.15 mathrm{Hz})}$; and high frequency $boldsymbol{(0.15sim 0.4mathrm{Hz})}$. Frequency domain analysis revealed that the very low frequency component of the ECG signal was increased in obstructive sleep apnea patients during episodes of obstructive sleep apnea, whereas other frequencies (low and high frequency) showed no significant changes. This new cardiovascular variable can be used as a new important feature in any comprehensive obstructive sleep apnea prediction model (to increase the accuracy of the model) or in any other noninvasive approaches for diagnosing or monitoring obstructive sleep apnea at home.
{"title":"The Effect of Obstructive Sleep Apnea on the Cardiovascular Variability","authors":"S. Sani","doi":"10.1109/ICSPIS51252.2020.9340127","DOIUrl":"https://doi.org/10.1109/ICSPIS51252.2020.9340127","url":null,"abstract":"Sleep apnea is a serious medical condition that influence millions of adults and children in the United States. It is unknown whether or not sleep apnea can cause any abnormalities in cardiovascular variability in patients with no cardiovascular disease. In this study, we investigated new cardiovascular variables which correspond with obstructive sleep apnea. The power spectrum of the electrocardiogram signal (ECG) was studied in healthy aged-matched adults $boldsymbol{(mathrm{n}=15)}$ and patients diagnosed with moderate obstructive sleep apnea $boldsymbol{(mathrm{n}=15)}$. The power spectral analysis was divided into three main frequency bands: very low frequency $boldsymbol{(0.0033sim 0.04mathrm{Hz})}$; low frequency $boldsymbol{(0.04sim 0.15 mathrm{Hz})}$; and high frequency $boldsymbol{(0.15sim 0.4mathrm{Hz})}$. Frequency domain analysis revealed that the very low frequency component of the ECG signal was increased in obstructive sleep apnea patients during episodes of obstructive sleep apnea, whereas other frequencies (low and high frequency) showed no significant changes. This new cardiovascular variable can be used as a new important feature in any comprehensive obstructive sleep apnea prediction model (to increase the accuracy of the model) or in any other noninvasive approaches for diagnosing or monitoring obstructive sleep apnea at home.","PeriodicalId":373750,"journal":{"name":"2020 3rd International Conference on Signal Processing and Information Security (ICSPIS)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114178169","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 : 2020-11-25DOI: 10.1109/ICSPIS51252.2020.9340138
M. Rai, Hussain Al-Ahmad, O. Gouda, Dina Jamal, M. A. Talib, Q. Nasir
In the last few years, the easy access to images and videos shared online have been continuously increased. The generative adversarial networks using deep learning leads to create very realistic deepfake videos by playing with the digital content of images and videos. The spread of such deepfake videos on social media networks urged the international community to consider seriously its danger and accordingly encouraged the researchers around the world to develop powerful deepfake detection methods. Many approaches are available in the recent literature. In this paper, the proposed approach is based on exploiting the residual noise which is the difference between original image and its denoised version. The study of residual noise has shown effectiveness in deep-fake detection with regards to its distinctive and discriminative features which can be effectively captured by convolutional neural networks with transfer learning. The performance of our approach is evaluated on two datasets: low-resolution video sequences of the FaceForensics++ and high-resolution videos from Kaggle Deepfake Detection challenge (DFDC). The obtained results show relevant accuracy in comparison with other competitive methods.
{"title":"Fighting Deepfake by Residual Noise Using Convolutional Neural Networks","authors":"M. Rai, Hussain Al-Ahmad, O. Gouda, Dina Jamal, M. A. Talib, Q. Nasir","doi":"10.1109/ICSPIS51252.2020.9340138","DOIUrl":"https://doi.org/10.1109/ICSPIS51252.2020.9340138","url":null,"abstract":"In the last few years, the easy access to images and videos shared online have been continuously increased. The generative adversarial networks using deep learning leads to create very realistic deepfake videos by playing with the digital content of images and videos. The spread of such deepfake videos on social media networks urged the international community to consider seriously its danger and accordingly encouraged the researchers around the world to develop powerful deepfake detection methods. Many approaches are available in the recent literature. In this paper, the proposed approach is based on exploiting the residual noise which is the difference between original image and its denoised version. The study of residual noise has shown effectiveness in deep-fake detection with regards to its distinctive and discriminative features which can be effectively captured by convolutional neural networks with transfer learning. The performance of our approach is evaluated on two datasets: low-resolution video sequences of the FaceForensics++ and high-resolution videos from Kaggle Deepfake Detection challenge (DFDC). The obtained results show relevant accuracy in comparison with other competitive methods.","PeriodicalId":373750,"journal":{"name":"2020 3rd International Conference on Signal Processing and Information Security (ICSPIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130125636","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 : 2020-11-25DOI: 10.1109/ICSPIS51252.2020.9340136
K. Makdi, Frederick T. Sheldon, A. A. Hussein
Contemporarily, cloud computing is becoming the most preferred choice for IT firms because it offers flexibility and pay-per-use services. Nonetheless, privacy and security issues are significant challenges in the successful deployment of cloud computing attributed to its distributed and open architectures that are exposed to intrusions. The open and distributed structures of cloud computing are increasingly appealing to potential cybercriminals. Conventional intrusion detection systems are largely ineffective in the cloud computing environment because of their openness. This paper examines the deployment of novel intrusion detection systems involving a trust-based adaptive security model for intrusion detection through deep learning.
{"title":"Trusted Security Model for IDS Using Deep Learning","authors":"K. Makdi, Frederick T. Sheldon, A. A. Hussein","doi":"10.1109/ICSPIS51252.2020.9340136","DOIUrl":"https://doi.org/10.1109/ICSPIS51252.2020.9340136","url":null,"abstract":"Contemporarily, cloud computing is becoming the most preferred choice for IT firms because it offers flexibility and pay-per-use services. Nonetheless, privacy and security issues are significant challenges in the successful deployment of cloud computing attributed to its distributed and open architectures that are exposed to intrusions. The open and distributed structures of cloud computing are increasingly appealing to potential cybercriminals. Conventional intrusion detection systems are largely ineffective in the cloud computing environment because of their openness. This paper examines the deployment of novel intrusion detection systems involving a trust-based adaptive security model for intrusion detection through deep learning.","PeriodicalId":373750,"journal":{"name":"2020 3rd International Conference on Signal Processing and Information Security (ICSPIS)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128584572","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 : 2020-11-25DOI: 10.1109/ICSPIS51252.2020.9340128
Hadeel Mohammed Jawad, Samir Tout
Cybersecurity experts predict that cybersecurity is going to be the new cold war. Arab countries are exposed to cyber-attacks that are aimed at stealing personal data and trade secrets. Furthermore, statistics show that women in such countries are exposed to different forms of cyber violence. This paper introduces a new mobile app in the Arabic language to educate Arab-speaking people in the Middle East and North Africa (MENA) about cybersecurity and to increase their awareness of information assurance and cybercrimes. The app was developed for Android and iOS devices and it includes multiple-choice information assurance questions, terms, and articles. Examples of the term definitions are Two-Factor Authentication, Ethical Hacking, and Honeypot. The app data could be increased in the next update of the app.
{"title":"Introducing a Mobile App to Increase Cybersecurity Awareness in MENA","authors":"Hadeel Mohammed Jawad, Samir Tout","doi":"10.1109/ICSPIS51252.2020.9340128","DOIUrl":"https://doi.org/10.1109/ICSPIS51252.2020.9340128","url":null,"abstract":"Cybersecurity experts predict that cybersecurity is going to be the new cold war. Arab countries are exposed to cyber-attacks that are aimed at stealing personal data and trade secrets. Furthermore, statistics show that women in such countries are exposed to different forms of cyber violence. This paper introduces a new mobile app in the Arabic language to educate Arab-speaking people in the Middle East and North Africa (MENA) about cybersecurity and to increase their awareness of information assurance and cybercrimes. The app was developed for Android and iOS devices and it includes multiple-choice information assurance questions, terms, and articles. Examples of the term definitions are Two-Factor Authentication, Ethical Hacking, and Honeypot. The app data could be increased in the next update of the app.","PeriodicalId":373750,"journal":{"name":"2020 3rd International Conference on Signal Processing and Information Security (ICSPIS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116425305","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 : 2020-11-25DOI: 10.1109/ICSPIS51252.2020.9340151
José de Jesús Rugeles Uribe, E. Guillén
Large-scale cyber attacks using IoT devices have increased in recent years. One of the strategies to deal with this problem is the use of penetration test techniques. The aim of this study was to develop a vulnerability assessment for an IoT M2M node that uses GSM technology. A test scenario was designed consisting of a GMS network created using USRP N210 and OpenBTS radios in a multicell configuration. An IoT-GSM node was designed from a sim8001 radio module, used in several M2M devices. The IoT node stores the measurements of the radio bases operational parameters that make up the GSM network. An algorithm for controlling an attacking base radio was designed using the OpenBTS API, which allows the deployment of a “man in the middle” attack. The entire test deployment can be carried out remotely. Analysis of the results of the measurements obtained lets us understand the attack's behavior in detail and determine the IoT-GSM node's vulnerability. The results obtained show the potential of SDR and OpenBTS technology as penetration test tools to analyze vulnerabilities of IoT systems.
{"title":"Vulnerability Assessment for IoT Nodes Using OpenBTS and Software Defined Radios","authors":"José de Jesús Rugeles Uribe, E. Guillén","doi":"10.1109/ICSPIS51252.2020.9340151","DOIUrl":"https://doi.org/10.1109/ICSPIS51252.2020.9340151","url":null,"abstract":"Large-scale cyber attacks using IoT devices have increased in recent years. One of the strategies to deal with this problem is the use of penetration test techniques. The aim of this study was to develop a vulnerability assessment for an IoT M2M node that uses GSM technology. A test scenario was designed consisting of a GMS network created using USRP N210 and OpenBTS radios in a multicell configuration. An IoT-GSM node was designed from a sim8001 radio module, used in several M2M devices. The IoT node stores the measurements of the radio bases operational parameters that make up the GSM network. An algorithm for controlling an attacking base radio was designed using the OpenBTS API, which allows the deployment of a “man in the middle” attack. The entire test deployment can be carried out remotely. Analysis of the results of the measurements obtained lets us understand the attack's behavior in detail and determine the IoT-GSM node's vulnerability. The results obtained show the potential of SDR and OpenBTS technology as penetration test tools to analyze vulnerabilities of IoT systems.","PeriodicalId":373750,"journal":{"name":"2020 3rd International Conference on Signal Processing and Information Security (ICSPIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130066574","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 : 2020-11-25DOI: 10.1109/icspis51252.2020.9340137
R. El-Khazali, S. Momani, I. Batiha
this work presents a new discrete-time fractional-order phase-locked loop (DTFoPLL). The new model is developed by introducing a 1st-order z-transfer function that approximates a dynamic fractional-order discrete-time integrator that depends on its fractional order. Such structure is used to realize both the discrete-time fractional-order filter (DTFoF) and a fractional-order digital controlled-oscillator (FoDCO) of the PLL. The flexibility of the design follows from choosing low-order PLLs (between $boldsymbol{0.1leqalphaleq 0.2)}$ that provides wider bandwidth and requires lower gains than its integer-order counter parts. The main points of this work are illustrated via numerical simulation.
{"title":"A New Discrete-Time Model of Fractional-Order PLL","authors":"R. El-Khazali, S. Momani, I. Batiha","doi":"10.1109/icspis51252.2020.9340137","DOIUrl":"https://doi.org/10.1109/icspis51252.2020.9340137","url":null,"abstract":"this work presents a new discrete-time fractional-order phase-locked loop (DTFoPLL). The new model is developed by introducing a 1st-order z-transfer function that approximates a dynamic fractional-order discrete-time integrator that depends on its fractional order. Such structure is used to realize both the discrete-time fractional-order filter (DTFoF) and a fractional-order digital controlled-oscillator (FoDCO) of the PLL. The flexibility of the design follows from choosing low-order PLLs (between $boldsymbol{0.1leqalphaleq 0.2)}$ that provides wider bandwidth and requires lower gains than its integer-order counter parts. The main points of this work are illustrated via numerical simulation.","PeriodicalId":373750,"journal":{"name":"2020 3rd International Conference on Signal Processing and Information Security (ICSPIS)","volume":"482 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116526590","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：