Pub Date : 2018-09-01DOI: 10.1109/ICCP.2018.8516600
Calin Timbus, Vlad-Cristian Miclea, C. Lemnaru
We present a method for detecting and classifying traffic signs based on two deep neural network architectures. A Fully Convolutional Network (FCN) - based semantic segmentation model is modified to extract traffic sign regions of interest. These regions are further passed to a Convolutional Neural Network (CNN) for traffic sign classification. We propose a novel CNN architecture for the classification step. In evaluating our approach, we contrast the efficiency and the robustness of the deep learning image segmentation approach with classical image processing filters traditionally applied for traffic sign detection. We also show the effectiveness of our CNN-based recognition method by integrating it in our system.
{"title":"Semantic segmentation-based traffic sign detection and recognition using deep learning techniques","authors":"Calin Timbus, Vlad-Cristian Miclea, C. Lemnaru","doi":"10.1109/ICCP.2018.8516600","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516600","url":null,"abstract":"We present a method for detecting and classifying traffic signs based on two deep neural network architectures. A Fully Convolutional Network (FCN) - based semantic segmentation model is modified to extract traffic sign regions of interest. These regions are further passed to a Convolutional Neural Network (CNN) for traffic sign classification. We propose a novel CNN architecture for the classification step. In evaluating our approach, we contrast the efficiency and the robustness of the deep learning image segmentation approach with classical image processing filters traditionally applied for traffic sign detection. We also show the effectiveness of our CNN-based recognition method by integrating it in our system.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126147750","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 : 2018-09-01DOI: 10.1109/iccp.2018.8516588
{"title":"ICCP 2018 Keynote Lecture","authors":"","doi":"10.1109/iccp.2018.8516588","DOIUrl":"https://doi.org/10.1109/iccp.2018.8516588","url":null,"abstract":"","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124957034","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 : 2018-09-01DOI: 10.1109/ICCP.2018.8516594
T. Stefanut, V. Bâcu, C. Nandra, Denisa Balasz, D. Gorgan, Ovidiu Vaduvescu
In the past two decades an increasing interest in discovering Near Earth Objects has been noted in the astronomical community. Dedicated surveys have been operated for data acquisition and processing, resulting in the present discovery of over 18.000 objects that are closer than 30 million miles of Earth. Nevertheless, recent events have shown that there still are many undiscovered asteroids that can be on collision course to Earth. This article presents an original NEO detection algorithm developed in the NEARBY research object, that has been integrated into an automated MOPS processing pipeline aimed at identifying moving space objects based on the blink method. Proposed solution can be considered an approach of Big Data processing and analysis, implementing visual analytics techniques for rapid human data validation.
{"title":"NEARBY Platform: Algorithm for Automated Asteroids Detection in Astronomical Images","authors":"T. Stefanut, V. Bâcu, C. Nandra, Denisa Balasz, D. Gorgan, Ovidiu Vaduvescu","doi":"10.1109/ICCP.2018.8516594","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516594","url":null,"abstract":"In the past two decades an increasing interest in discovering Near Earth Objects has been noted in the astronomical community. Dedicated surveys have been operated for data acquisition and processing, resulting in the present discovery of over 18.000 objects that are closer than 30 million miles of Earth. Nevertheless, recent events have shown that there still are many undiscovered asteroids that can be on collision course to Earth. This article presents an original NEO detection algorithm developed in the NEARBY research object, that has been integrated into an automated MOPS processing pipeline aimed at identifying moving space objects based on the blink method. Proposed solution can be considered an approach of Big Data processing and analysis, implementing visual analytics techniques for rapid human data validation.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"202 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114427883","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 : 2018-09-01DOI: 10.1109/ICCP.2018.8516584
Narjess Dali, Sadok Bouamama
Constraint Satisfaction Problems (CSPs) are among the easiest and more used formalisms to model real-world-constrained problems (transport, planning, scheduling, Indeed, the Genetic Algorithm (GA) is one of the optimization methods used to solve CSPs. This meta-heuristic finds a good solution in a reasonable time. However, it could be inefficient when dealing with very large-scale problems, in particular CSPs. Therefore, the High Performance Computing (HPC) is recommended, as an additional way, to accelerate the research. This paper introduces two parallel genetic algorithm-based approaches using GPU for solving Maximal Constraint Satisfaction Problems (Max-CSPs). The first approach is based on one parallelism level, while the second approach is based on two parallelism levels. The experimental results presented in this work, prove how efficient our proposed approaches are.
{"title":"New parallel Genetic Algorithms on GPU for solving Max-CSPs","authors":"Narjess Dali, Sadok Bouamama","doi":"10.1109/ICCP.2018.8516584","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516584","url":null,"abstract":"Constraint Satisfaction Problems (CSPs) are among the easiest and more used formalisms to model real-world-constrained problems (transport, planning, scheduling, Indeed, the Genetic Algorithm (GA) is one of the optimization methods used to solve CSPs. This meta-heuristic finds a good solution in a reasonable time. However, it could be inefficient when dealing with very large-scale problems, in particular CSPs. Therefore, the High Performance Computing (HPC) is recommended, as an additional way, to accelerate the research. This paper introduces two parallel genetic algorithm-based approaches using GPU for solving Maximal Constraint Satisfaction Problems (Max-CSPs). The first approach is based on one parallelism level, while the second approach is based on two parallelism levels. The experimental results presented in this work, prove how efficient our proposed approaches are.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"149 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125242033","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 : 2018-09-01DOI: 10.1109/ICCP.2018.8516605
Andreea Sicaru, A. Morar, A. Moldoveanu, F. Moldoveanu, M. Mocanu
Digital technology plays an important role in everyday activities and facilitates the lives of everyone. This raises the problem of designing technology accessible for everyone, including people with disabilities such as visual impairments. Assistive technology for visually impaired people requires specific approaches in its design due to the fact that the main focus is not on the visual aspects. The game industry has concentrated most of its resources on developing games for the regular community, leaving people with visual impairments only a few options. This paper presents the status of the applications and games for the blind, identifies the main challenges and the perspectives for development.
{"title":"Computer Gaming for the Visually Impaired -- Status and Perspectives","authors":"Andreea Sicaru, A. Morar, A. Moldoveanu, F. Moldoveanu, M. Mocanu","doi":"10.1109/ICCP.2018.8516605","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516605","url":null,"abstract":"Digital technology plays an important role in everyday activities and facilitates the lives of everyone. This raises the problem of designing technology accessible for everyone, including people with disabilities such as visual impairments. Assistive technology for visually impaired people requires specific approaches in its design due to the fact that the main focus is not on the visual aspects. The game industry has concentrated most of its resources on developing games for the regular community, leaving people with visual impairments only a few options. This paper presents the status of the applications and games for the blind, identifies the main challenges and the perspectives for development.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130551956","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 : 2018-09-01DOI: 10.1109/ICCP.2018.8516635
S. Kaymak, Ali Serener
Age-related macular degeneration (AMD) is an eye disease that damages the retina, causing vision loss. Diabetic macular edema (DME) is also a form of vision loss for diabetic people. It is therefore crucial to detect AMD and DME in the early stages for the timely treatment of the eye and the prevention of any vision impairment. Automatic detection of DME and AMD on optical coherence tomography (OCT) images are presented in this paper. The method used is based on training a deep learning algorithm to classify them into healthy, dry AMD, wet AMD and DME categories. This method outperforms a transfer learning based method proposed recently in the literature for classification of OCT images into AMD and DME categories.
{"title":"Automated Age-Related Macular Degeneration and Diabetic Macular Edema Detection on OCT Images using Deep Learning","authors":"S. Kaymak, Ali Serener","doi":"10.1109/ICCP.2018.8516635","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516635","url":null,"abstract":"Age-related macular degeneration (AMD) is an eye disease that damages the retina, causing vision loss. Diabetic macular edema (DME) is also a form of vision loss for diabetic people. It is therefore crucial to detect AMD and DME in the early stages for the timely treatment of the eye and the prevention of any vision impairment. Automatic detection of DME and AMD on optical coherence tomography (OCT) images are presented in this paper. The method used is based on training a deep learning algorithm to classify them into healthy, dry AMD, wet AMD and DME categories. This method outperforms a transfer learning based method proposed recently in the literature for classification of OCT images into AMD and DME categories.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121235397","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 : 2018-09-01DOI: 10.1109/ICCP.2018.8516631
F. Vancea, S. Nedevschi
Vehicle taillight detection is an important topic in the fields of collision avoidance systems and autonomous vehicles. By analyzing the changes in the taillights of vehicles, the intention of the driver can be understood, which can prevent possible accidents. This paper presents a convolutional neural network architecture capable of segmenting taillight pixels by detecting vehicles and uses already computed features to segment taillights. The network is composed of a Faster RCNN that detects vehicles and classify them based their orientation relative to the camera and a subnetwork that is responsible for segmenting taillight pixels from vehicles that have their rear facing the camera. Multiple Faster RCNN configurations were trained and evaluated. This work also presents a way of adapting the ERFNet semantic segmentation architecture for the purpose of taillight extraction, object detection and classification. The networks were trained and evaluated using the KITTI object detection dataset.
{"title":"Semantic information based vehicle relative orientation and taillight detection","authors":"F. Vancea, S. Nedevschi","doi":"10.1109/ICCP.2018.8516631","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516631","url":null,"abstract":"Vehicle taillight detection is an important topic in the fields of collision avoidance systems and autonomous vehicles. By analyzing the changes in the taillights of vehicles, the intention of the driver can be understood, which can prevent possible accidents. This paper presents a convolutional neural network architecture capable of segmenting taillight pixels by detecting vehicles and uses already computed features to segment taillights. The network is composed of a Faster RCNN that detects vehicles and classify them based their orientation relative to the camera and a subnetwork that is responsible for segmenting taillight pixels from vehicles that have their rear facing the camera. Multiple Faster RCNN configurations were trained and evaluated. This work also presents a way of adapting the ERFNet semantic segmentation architecture for the purpose of taillight extraction, object detection and classification. The networks were trained and evaluated using the KITTI object detection dataset.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"201 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121312352","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 : 2018-09-01DOI: 10.1109/ICCP.2018.8516616
Adrian Groza, Pınar Öztürk, R. R. Slavescu, A. Marginean, R. Prasath
Climate experts have agreed that global warming is at least partially caused by certain human activities. Different from experts, this agreement has not reached all the public arena yet. That is, people have different views and therefore argue about climate change issues. We are interested in analysing people's arguments on global warming. The large number of conveyed arguments need somehow to be aggregated in order to have a top level view on what people believe. To build such collective opinion, we use subjective logic. Based on subjective reasoning we are able to assess the expectance that a debate topic be accepted by a given community or arguers. We collected arguments on climate change from three debates sites: Debatepedia, For and Against and Debate.org. We can analyse the differences between such communities. We use the consensus operator in subjective logic to aggregate similar opinions from distinct debate communities. Moreover, various debate topics can refer to the same issue but with different phases. We apply the affinity propagation algorithm to cluster the debates. Our approach for analysing people arguments can be applied in different domains, other than the one exemplified here, that is climate change.
{"title":"Analysing Climate Change Arguments Using Subjective Logic","authors":"Adrian Groza, Pınar Öztürk, R. R. Slavescu, A. Marginean, R. Prasath","doi":"10.1109/ICCP.2018.8516616","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516616","url":null,"abstract":"Climate experts have agreed that global warming is at least partially caused by certain human activities. Different from experts, this agreement has not reached all the public arena yet. That is, people have different views and therefore argue about climate change issues. We are interested in analysing people's arguments on global warming. The large number of conveyed arguments need somehow to be aggregated in order to have a top level view on what people believe. To build such collective opinion, we use subjective logic. Based on subjective reasoning we are able to assess the expectance that a debate topic be accepted by a given community or arguers. We collected arguments on climate change from three debates sites: Debatepedia, For and Against and Debate.org. We can analyse the differences between such communities. We use the consensus operator in subjective logic to aggregate similar opinions from distinct debate communities. Moreover, various debate topics can refer to the same issue but with different phases. We apply the affinity propagation algorithm to cluster the debates. Our approach for analysing people arguments can be applied in different domains, other than the one exemplified here, that is climate change.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127521236","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 : 2018-09-01DOI: 10.1109/ICCP.2018.8516644
Andrei Zamfira, H. Ciocarlie
Many detection techniques have been proposed until now that struggle to keep up with the inherent complexity of applications, networks and protocols, resulting also in the growing rate of attacks that exploit them. Security frameworks that are created using an ontological approach are the next-gen systems of defense that have some advantages over the conventional techniques because they can capture the context of information and are capable to filter these contents depending on some certain factors. This paper proposes a method of creating an ontology that can be used for improving detection capabilities of attacks at all application levels. The ontology serves as a data model and knowledge base of the cyberoperations domain that conceptualizes and stores various types of data needed in the process of detecting an aware situation, such as information about attacks (types), OSI stack levels to which are targeted (software, network, hardware), countermeasure methods, resources necessary, knowledge required etc. The quality of the proposed model was assessed using a methodology known as OntoClean, that is a comprehensive suite of metrics for ontology evaluation that can comprise up to 15 criteria, as will be discussed during this paper. The ontology was tested in attack detection using a prototype web application firewall. In the evaluation process we used the famous dataset Kyoto2006+ proposed by the University of Kyoto in this scope. The results yielded for attacks detection by our proposed system were compared to other existing security solutions, like ModSecurit and Snort. In the conclusion section are stated the future directions of this research towards constructing reliable systems for cyber-security.
{"title":"Developing An Ontology Of Cyber-Operations In Networks Of Computers","authors":"Andrei Zamfira, H. Ciocarlie","doi":"10.1109/ICCP.2018.8516644","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516644","url":null,"abstract":"Many detection techniques have been proposed until now that struggle to keep up with the inherent complexity of applications, networks and protocols, resulting also in the growing rate of attacks that exploit them. Security frameworks that are created using an ontological approach are the next-gen systems of defense that have some advantages over the conventional techniques because they can capture the context of information and are capable to filter these contents depending on some certain factors. This paper proposes a method of creating an ontology that can be used for improving detection capabilities of attacks at all application levels. The ontology serves as a data model and knowledge base of the cyberoperations domain that conceptualizes and stores various types of data needed in the process of detecting an aware situation, such as information about attacks (types), OSI stack levels to which are targeted (software, network, hardware), countermeasure methods, resources necessary, knowledge required etc. The quality of the proposed model was assessed using a methodology known as OntoClean, that is a comprehensive suite of metrics for ontology evaluation that can comprise up to 15 criteria, as will be discussed during this paper. The ontology was tested in attack detection using a prototype web application firewall. In the evaluation process we used the famous dataset Kyoto2006+ proposed by the University of Kyoto in this scope. The results yielded for attacks detection by our proposed system were compared to other existing security solutions, like ModSecurit and Snort. In the conclusion section are stated the future directions of this research towards constructing reliable systems for cyber-security.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126742454","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 : 2018-09-01DOI: 10.1109/ICCP.2018.8516593
Sergiu Redeca, Adrian Groza
We focus here on designing agents for games with incomplete information, such that the Stratego game. We develop two playing agents that use probabilities and forward reasoning with multiple-ply. We also proposed various evaluation functions for a given position and we analyse the importance of the starting configuration.
{"title":"Designing agents for the Stratego game","authors":"Sergiu Redeca, Adrian Groza","doi":"10.1109/ICCP.2018.8516593","DOIUrl":"https://doi.org/10.1109/ICCP.2018.8516593","url":null,"abstract":"We focus here on designing agents for games with incomplete information, such that the Stratego game. We develop two playing agents that use probabilities and forward reasoning with multiple-ply. We also proposed various evaluation functions for a given position and we analyse the importance of the starting configuration.","PeriodicalId":259007,"journal":{"name":"2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP)","volume":"172 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121043682","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：