Pub Date : 2017-10-01DOI: 10.1109/CCST.2017.8167851
Alvaro Duque de Quevedo, Fernando Ibanez Urzaiz, J. Gismero-Menoyo, A. Asensio-López
This paper presents the first experimental results of an ubiquitous frequency modulated continuous wave (FMCW) radar system demonstrator, working in the 8.75 GHz (X band). This system, which main application is in the area of surface surveillance radars, has been developed within the framework of the project numbered as TEC2014-53815-R (RAD-DAR), and is currently under its field-testing phase. This article presents the demonstrator and describes its main blocks. Subsequently, it focuses on signal acquisition and digitization, to later explain the offline coherent signal processing carried out with the captured data. After this, the document introduces the chosen scenario for the first field tests, and shows the first results of the offline processing, illustrated with Range-Doppler matrices and detection figures. Finally, the paper introduces future work lines, including the development of a radar data processor taking advantage of the achieved strong range-speed association.
{"title":"X-band ubiquitous radar system: First experimental results","authors":"Alvaro Duque de Quevedo, Fernando Ibanez Urzaiz, J. Gismero-Menoyo, A. Asensio-López","doi":"10.1109/CCST.2017.8167851","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167851","url":null,"abstract":"This paper presents the first experimental results of an ubiquitous frequency modulated continuous wave (FMCW) radar system demonstrator, working in the 8.75 GHz (X band). This system, which main application is in the area of surface surveillance radars, has been developed within the framework of the project numbered as TEC2014-53815-R (RAD-DAR), and is currently under its field-testing phase. This article presents the demonstrator and describes its main blocks. Subsequently, it focuses on signal acquisition and digitization, to later explain the offline coherent signal processing carried out with the captured data. After this, the document introduces the chosen scenario for the first field tests, and shows the first results of the offline processing, illustrated with Range-Doppler matrices and detection figures. Finally, the paper introduces future work lines, including the development of a radar data processor taking advantage of the achieved strong range-speed association.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125795846","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167797
R. Sánchez-Reillo, Ines Ortega-Fernandez, Wendy Ponce-Hernandez, Helga C. Quiros-Sandoval
Biometric R&D has to deal with personal data. From the Universal Declaration of Human Rights, privacy of a human being shall be protected, and this is addressed in different forms in each region of the world. In the case of the European Union, Data Protection Directives, Laws and Regulation have been established, and interpreted in different ways by each European Member State. Such a diversity has pushed the European Union to generate an improved regulation that will be mandatory in 2018. Biometric R&D shall not only comply with the current Directive, but also has to adapt its work to the new Regulation. This work is intended to describe the situation and provide a recommended procedure when having to acquire personal data.
{"title":"How to implement EU data protection regulation for R&D on personal data","authors":"R. Sánchez-Reillo, Ines Ortega-Fernandez, Wendy Ponce-Hernandez, Helga C. Quiros-Sandoval","doi":"10.1109/CCST.2017.8167797","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167797","url":null,"abstract":"Biometric R&D has to deal with personal data. From the Universal Declaration of Human Rights, privacy of a human being shall be protected, and this is addressed in different forms in each region of the world. In the case of the European Union, Data Protection Directives, Laws and Regulation have been established, and interpreted in different ways by each European Member State. Such a diversity has pushed the European Union to generate an improved regulation that will be mandatory in 2018. Biometric R&D shall not only comply with the current Directive, but also has to adapt its work to the new Regulation. This work is intended to describe the situation and provide a recommended procedure when having to acquire personal data.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128494828","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167838
Patrick A. H. Bours, Sukanya Brahmanpally
Keystroke Dynamics can be used as an unobtrusive method to enhance password authentication, by checking the typing rhythm of the user. Fixed passwords will give an attacker the possibility to try to learn to mimic the typing behaviour of a victim. In this paper we will investigate the performance of a keystroke dynamic (KD) system when the users have to type given (English) words. Under the assumption that it is easy to type words in your native language and difficult in a foreign language will we also test the performance of such a challenge-based KD system when the challenges are not common English words, but words in the native language of the user. We collected data from participants with 6 different native language backgrounds and had them type random 8–12 character words in each of the 6 languages. The participants also typed random English words and random French words. English was assumed to be a language familiar to all participants, while French was not a native language to any participant and most likely most participants were not fluent in French. Analysis showed that using language dependent words gave a better performance of the challenge-based KD compared to an all English challenge-based system. When using words in a native language, then the performance of the participants with their mother-tongue equal to that native language had a similar performance compared to the all English challenge-based system, but the non-native speakers had an FMR that was significantly lower than the native language speakers. We found that native Telugu speakers had an FMR of less than 1% when writing Spanish or Slovak words. We also found that duration features were best to recognize genuine users, but latency features performed best to recognize non-native impostor users.
{"title":"Language dependent challenge-based keystroke dynamics","authors":"Patrick A. H. Bours, Sukanya Brahmanpally","doi":"10.1109/CCST.2017.8167838","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167838","url":null,"abstract":"Keystroke Dynamics can be used as an unobtrusive method to enhance password authentication, by checking the typing rhythm of the user. Fixed passwords will give an attacker the possibility to try to learn to mimic the typing behaviour of a victim. In this paper we will investigate the performance of a keystroke dynamic (KD) system when the users have to type given (English) words. Under the assumption that it is easy to type words in your native language and difficult in a foreign language will we also test the performance of such a challenge-based KD system when the challenges are not common English words, but words in the native language of the user. We collected data from participants with 6 different native language backgrounds and had them type random 8–12 character words in each of the 6 languages. The participants also typed random English words and random French words. English was assumed to be a language familiar to all participants, while French was not a native language to any participant and most likely most participants were not fluent in French. Analysis showed that using language dependent words gave a better performance of the challenge-based KD compared to an all English challenge-based system. When using words in a native language, then the performance of the participants with their mother-tongue equal to that native language had a similar performance compared to the all English challenge-based system, but the non-native speakers had an FMR that was significantly lower than the native language speakers. We found that native Telugu speakers had an FMR of less than 1% when writing Spanish or Slovak words. We also found that duration features were best to recognize genuine users, but latency features performed best to recognize non-native impostor users.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132166158","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167820
L. Vergara, A. Salazar, J. Belda, G. Safont, S. Moral, S. Iglesias
In this paper, several methods based on signal processing on graphs are proposed to improve the performance of credit card fraud detection. The proposed methods consist of a variant of the classic iterative amplitude adjusted Fourier transform (IAAFT) and two methods that we have called iterative surrogate signals on graph algorithms (ISSG). The objective is to generate surrogate samples from the original scarce fraud samples to improve the training of the detectors by lowering the variance of the estimate. A reliable augmentation of the target scarce population of frauds is important considering issues such as labeling cost; algorithm testing; data confidentiality; and constantly changing of patterns in the data streaming source. We have approached several scenarios with different legitimate and non-legitimate transaction ratios showing the feasibility of improving detection capabilities evaluated by means of receiver operating characteristic (ROC) curves and several key performance indicators (KPI) commonly used in financial business.
{"title":"Signal processing on graphs for improving automatic credit card fraud detection","authors":"L. Vergara, A. Salazar, J. Belda, G. Safont, S. Moral, S. Iglesias","doi":"10.1109/CCST.2017.8167820","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167820","url":null,"abstract":"In this paper, several methods based on signal processing on graphs are proposed to improve the performance of credit card fraud detection. The proposed methods consist of a variant of the classic iterative amplitude adjusted Fourier transform (IAAFT) and two methods that we have called iterative surrogate signals on graph algorithms (ISSG). The objective is to generate surrogate samples from the original scarce fraud samples to improve the training of the detectors by lowering the variance of the estimate. A reliable augmentation of the target scarce population of frauds is important considering issues such as labeling cost; algorithm testing; data confidentiality; and constantly changing of patterns in the data streaming source. We have approached several scenarios with different legitimate and non-legitimate transaction ratios showing the feasibility of improving detection capabilities evaluated by means of receiver operating characteristic (ROC) curves and several key performance indicators (KPI) commonly used in financial business.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129279626","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167815
D. Morocho, A. Morales, Julian Fierrez, J. Ortega-Garcia
The present work analyzes performance, abilities and contributions of the human being (layman) in semi-automatic signature recognition systems. During the last decade the performance of Automatic Signature Verification systems have been improved based on new machine learning techniques and better knowledge about intraclass and interclass variability of signers. However, there is still room for improvements and some real world applications demands lower error rates. This work analyzes collaborative tools such as crowdsourcing and human-assisted schemes developed to improve Automatic Signature Verification systems. The performance of humans in semi-automatic recognition tasks is directly related to the information provided during the comparisons. How humans can help automatic systems goes from direct forgery detection to semiautomatic attribute labeling. In this work, we present recent advances, analyzing their performance according to the same experimental protocol. The results suggest the potential of comparative attributes as a way to improve Automatic Signature Verification systems.
{"title":"Humans in the loop: Study of semi-automatic signature recognition based on attributes","authors":"D. Morocho, A. Morales, Julian Fierrez, J. Ortega-Garcia","doi":"10.1109/CCST.2017.8167815","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167815","url":null,"abstract":"The present work analyzes performance, abilities and contributions of the human being (layman) in semi-automatic signature recognition systems. During the last decade the performance of Automatic Signature Verification systems have been improved based on new machine learning techniques and better knowledge about intraclass and interclass variability of signers. However, there is still room for improvements and some real world applications demands lower error rates. This work analyzes collaborative tools such as crowdsourcing and human-assisted schemes developed to improve Automatic Signature Verification systems. The performance of humans in semi-automatic recognition tasks is directly related to the information provided during the comparisons. How humans can help automatic systems goes from direct forgery detection to semiautomatic attribute labeling. In this work, we present recent advances, analyzing their performance according to the same experimental protocol. The results suggest the potential of comparative attributes as a way to improve Automatic Signature Verification systems.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126728586","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167805
Belén Ríos-Sánchez, Miguel Viana-Matesanz, C. S. Ávila
In this work, the Curvelet transform is proposed as a fairly new feature extraction method for palmprint recognition. Particularly, a multiscale analysis has been performed at four levels, assessing and combining the features extracted at each level in order to find those which better represent the palmprint. Feature matching has been conducted by means of Euclidean distance and Support Vector Machines (SVMs), and comparative results are provided. In addition, a multimodal approach involving the extracted palmprint features and hand geometry features has also been evaluated, obtaining an improvement of the results in relation to monomodal biometrics. Evaluations have been carried out following an evaluation protocol based on the definitions suggested by the ISO/IDE 19795 norm that allows for a fair comparison between the different methods. To this end, images coming from two different contact-less databases, which cover different capturing conditions, have been employed.
{"title":"Curvelets for contact-less hand biometrics under varied environmental conditions","authors":"Belén Ríos-Sánchez, Miguel Viana-Matesanz, C. S. Ávila","doi":"10.1109/CCST.2017.8167805","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167805","url":null,"abstract":"In this work, the Curvelet transform is proposed as a fairly new feature extraction method for palmprint recognition. Particularly, a multiscale analysis has been performed at four levels, assessing and combining the features extracted at each level in order to find those which better represent the palmprint. Feature matching has been conducted by means of Euclidean distance and Support Vector Machines (SVMs), and comparative results are provided. In addition, a multimodal approach involving the extracted palmprint features and hand geometry features has also been evaluated, obtaining an improvement of the results in relation to monomodal biometrics. Evaluations have been carried out following an evaluation protocol based on the definitions suggested by the ISO/IDE 19795 norm that allows for a fair comparison between the different methods. To this end, images coming from two different contact-less databases, which cover different capturing conditions, have been employed.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122166040","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167846
Arash Habibi Lashkari, A. Seo, G. Gil, A. Ghorbani
Online advertisements (ads) have taken over the web, nowedays most websites contain some sort of ads. While ads produce revenue for the server maintainer or to businesses, they have become intrusive and dangerous as ever. The ads use more bandwidth, show inappropriate content, and spread malware such as adware and ransomware. Although there are many products to block ads, also known as ad blockers, most depend on static filter lists that must be managed manually and frequently updated. When malicious advertisers can produce millions of new URLs within minutes, this is not the most effective method against ads. In this paper we propose our own ad blocker, CIC-AB, which uses machine learning techniques to detect new and unknown ads without needing to update a filter list. The proposed ad blocker has been developed as an extension for the common browsers (e.g. Firefox and Chrome). It classifies URLs, both HTTP and HTTPS, as: non-ad, normal-ad and malicious-ad. The analysis showed the average precision, recall and False Positive rate of CIC-AB for five classifiers namely; Naive Bayes (NB), Support Vector Machine (SVM), K-Nearest Neighbour (KNN), Random Forest (RF) and Decision Tree (DT) is 97.16%, 94.96% and 3.38% respectively.
{"title":"CIC-AB: Online ad blocker for browsers","authors":"Arash Habibi Lashkari, A. Seo, G. Gil, A. Ghorbani","doi":"10.1109/CCST.2017.8167846","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167846","url":null,"abstract":"Online advertisements (ads) have taken over the web, nowedays most websites contain some sort of ads. While ads produce revenue for the server maintainer or to businesses, they have become intrusive and dangerous as ever. The ads use more bandwidth, show inappropriate content, and spread malware such as adware and ransomware. Although there are many products to block ads, also known as ad blockers, most depend on static filter lists that must be managed manually and frequently updated. When malicious advertisers can produce millions of new URLs within minutes, this is not the most effective method against ads. In this paper we propose our own ad blocker, CIC-AB, which uses machine learning techniques to detect new and unknown ads without needing to update a filter list. The proposed ad blocker has been developed as an extension for the common browsers (e.g. Firefox and Chrome). It classifies URLs, both HTTP and HTTPS, as: non-ad, normal-ad and malicious-ad. The analysis showed the average precision, recall and False Positive rate of CIC-AB for five classifiers namely; Naive Bayes (NB), Support Vector Machine (SVM), K-Nearest Neighbour (KNN), Random Forest (RF) and Decision Tree (DT) is 97.16%, 94.96% and 3.38% respectively.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"928 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116193270","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167853
Francisco-Javier Romero-Paisano, F. Pérez-Martínez, Susan Martínez-Cordero, J. Calvo-Gallego
The article presents an accurate system for surveillance purposes of multiple targets in wide areas. The surveillance system is based on a GB-ROSAR (Ground-Based-Rotating Synthetic Aperture Radar). It is a fixed radar with the ability of obtaining images by synthesizing the radiating aperture by a circular movement of the antenna. The prototype makes use of FM-CW (Frequency-Modulated Continuous-Wave) signals, operating in millimeter wave bands. The system has been developed with low cost devices. The objective of this work is to show, through simulations and experimental results, the real possibilities of this type of systems in the setting of precise surveillance of several targets simultaneously over wide areas. This study contains the design and construction of the proposed surveillance system, the implemented signal processing, the obtained results and its interpretation, and the conclusions about this work.
{"title":"Accurate multi-target surveillance system over wide areas","authors":"Francisco-Javier Romero-Paisano, F. Pérez-Martínez, Susan Martínez-Cordero, J. Calvo-Gallego","doi":"10.1109/CCST.2017.8167853","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167853","url":null,"abstract":"The article presents an accurate system for surveillance purposes of multiple targets in wide areas. The surveillance system is based on a GB-ROSAR (Ground-Based-Rotating Synthetic Aperture Radar). It is a fixed radar with the ability of obtaining images by synthesizing the radiating aperture by a circular movement of the antenna. The prototype makes use of FM-CW (Frequency-Modulated Continuous-Wave) signals, operating in millimeter wave bands. The system has been developed with low cost devices. The objective of this work is to show, through simulations and experimental results, the real possibilities of this type of systems in the setting of precise surveillance of several targets simultaneously over wide areas. This study contains the design and construction of the proposed surveillance system, the implemented signal processing, the obtained results and its interpretation, and the conclusions about this work.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128123975","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167835
Pedro Moura, Paulo A. P. Fazendeiro, Pedro Marques, A. Ferreira
Within the necessary security requirements, access control measures are essential to provide adequate means to protect data from unauthorized accesses. However, current and traditional solutions are commonly based on predefined access policies and roles and are therefore inflexible by assuming uniform access control decisions through people's different type of devices, environments and situational conditions, and across enterprises, location and time. We live in an age of the mobile paradigm of anytime/anywhere access as the smartphone is the most ubiquitous device that people now hold. In this new age, access control models need to determine adaptable access decisions based on multiple factors aggregated at the moment of request and not just perform a predefined comparison of attributes. This paper presents a new access control model: SoTRAACE — Socio-Technical Risk-Adaptable Access Control Model. This model aggregates attributes from various domains to help performing a risk assessment that is balanced against the operational needs at the moment of each request, so to provide the most accurate and secure access decision. As a proof of concept, SoTRAACE is used to model and compare two different use case scenarios in the healthcare sector.
{"title":"SoTRAACE — Socio-technical risk-adaptable access control model","authors":"Pedro Moura, Paulo A. P. Fazendeiro, Pedro Marques, A. Ferreira","doi":"10.1109/CCST.2017.8167835","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167835","url":null,"abstract":"Within the necessary security requirements, access control measures are essential to provide adequate means to protect data from unauthorized accesses. However, current and traditional solutions are commonly based on predefined access policies and roles and are therefore inflexible by assuming uniform access control decisions through people's different type of devices, environments and situational conditions, and across enterprises, location and time. We live in an age of the mobile paradigm of anytime/anywhere access as the smartphone is the most ubiquitous device that people now hold. In this new age, access control models need to determine adaptable access decisions based on multiple factors aggregated at the moment of request and not just perform a predefined comparison of attributes. This paper presents a new access control model: SoTRAACE — Socio-Technical Risk-Adaptable Access Control Model. This model aggregates attributes from various domains to help performing a risk assessment that is balanced against the operational needs at the moment of each request, so to provide the most accurate and secure access decision. As a proof of concept, SoTRAACE is used to model and compare two different use case scenarios in the healthcare sector.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131724768","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 : 2017-10-01DOI: 10.1109/CCST.2017.8167818
Rafael Martin Nieto, Jesus Molina Merchan, Álvaro García-Martín, J. Sanchez
There is a large demand in the area of video-surveillance, especially in people detection, which has caused a large increase in the number of researches and resources in this field. As training images and annotations are not always available, it is important to consider the cost involved in creating the detector models. For example, for elderly people detection, the detector must have into account different positions such as standing, sitting, in a wheelchair, etc. Therefore, this work has the main objective of reducing the amount of resources needed to generate the detection model, saving the cost of having to record new sequences and generate the associated annotations for a detector training. To achieve this, three synthetic image datasets have been created in order to train three different models, evaluating which model is optimal and finally analyzing its feasibility by comparing it with a people detector for wheelchair users trained with real images. Other people detection scenarios in which this technique could be applied are, for example, people riding horses or motorbikes, or people carrying supermarket carts. The synthetic datasets have been generated by combining images of standing people with wheelchair images, combining image patches, and segmenting sections of people (trunk, legs, etc.) to add them to the wheelchair image. As expected, the obtained results have a reduction of efficiency (between 21 and 25%) in exchange for the enormous saving in human annotation and resources to record real images.
{"title":"Generation and evaluation of synthetic models for training people detectors","authors":"Rafael Martin Nieto, Jesus Molina Merchan, Álvaro García-Martín, J. Sanchez","doi":"10.1109/CCST.2017.8167818","DOIUrl":"https://doi.org/10.1109/CCST.2017.8167818","url":null,"abstract":"There is a large demand in the area of video-surveillance, especially in people detection, which has caused a large increase in the number of researches and resources in this field. As training images and annotations are not always available, it is important to consider the cost involved in creating the detector models. For example, for elderly people detection, the detector must have into account different positions such as standing, sitting, in a wheelchair, etc. Therefore, this work has the main objective of reducing the amount of resources needed to generate the detection model, saving the cost of having to record new sequences and generate the associated annotations for a detector training. To achieve this, three synthetic image datasets have been created in order to train three different models, evaluating which model is optimal and finally analyzing its feasibility by comparing it with a people detector for wheelchair users trained with real images. Other people detection scenarios in which this technique could be applied are, for example, people riding horses or motorbikes, or people carrying supermarket carts. The synthetic datasets have been generated by combining images of standing people with wheelchair images, combining image patches, and segmenting sections of people (trunk, legs, etc.) to add them to the wheelchair image. As expected, the obtained results have a reduction of efficiency (between 21 and 25%) in exchange for the enormous saving in human annotation and resources to record real images.","PeriodicalId":371622,"journal":{"name":"2017 International Carnahan Conference on Security Technology (ICCST)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131897732","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：