Pub Date : 2019-02-01DOI: 10.1109/KBEI.2019.8734980
H. Khastavaneh, H. Ebrahimpour-Komleh
In this study we consider multi-view capabilities of convolutional neural networks as one of the best methods of representation learning. Multi-view learning as a machine learning technique deals with the task of learning from multiple distinct views or multiple distinct feature sets. Moreover, multi-view feature learning attempts to abstract and summarize distinct feature sets for further machine learning and pattern recognition tasks. In contrast to traditional multi-view learning methods, convolutional neural networks are able to generate representations from unstructured raw data; these features are very essential for real world applications. It is concluded that CNNs are inherently multi-view representation learning methods able to handle both natural and artificial views.
{"title":"On Multi-view Interpretation of Convolutional Neural Networks","authors":"H. Khastavaneh, H. Ebrahimpour-Komleh","doi":"10.1109/KBEI.2019.8734980","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8734980","url":null,"abstract":"In this study we consider multi-view capabilities of convolutional neural networks as one of the best methods of representation learning. Multi-view learning as a machine learning technique deals with the task of learning from multiple distinct views or multiple distinct feature sets. Moreover, multi-view feature learning attempts to abstract and summarize distinct feature sets for further machine learning and pattern recognition tasks. In contrast to traditional multi-view learning methods, convolutional neural networks are able to generate representations from unstructured raw data; these features are very essential for real world applications. It is concluded that CNNs are inherently multi-view representation learning methods able to handle both natural and artificial views.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"237 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123168088","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8734976
Ramin Ebrahim Nakhli, Hadi Moradi, Mohammad Sadeghi
with ever-increasing data on the internet, finding the desired content has become harder and that is why recommender systems’ role is very important in business. As a specific example, media service providers, such as Netflix, can improve their service by recommending desirable content to each user. Most of the previous studies used explicit feedback of users, through likes and dislikes, to recommend items to their customers. However, in many cases, there is not much explicit feedback about items which cripples typical recommender systems to operate efficiently and provide accurate recommendation. In this paper, a percentage of view approach is proposed to find relevant movies for customers. To prove the effectiveness of the approach, first, it is shown that this feature can be a good indicator of users’ like and dislike. Then the best approach is determined and used in a recommender system for Namava, a media service provider. Then the performance of this recommender system is compared to a random recommender system and the effectiveness of the approach is shown.
{"title":"Movie Recommender System Based on Percentage of View","authors":"Ramin Ebrahim Nakhli, Hadi Moradi, Mohammad Sadeghi","doi":"10.1109/KBEI.2019.8734976","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8734976","url":null,"abstract":"with ever-increasing data on the internet, finding the desired content has become harder and that is why recommender systems’ role is very important in business. As a specific example, media service providers, such as Netflix, can improve their service by recommending desirable content to each user. Most of the previous studies used explicit feedback of users, through likes and dislikes, to recommend items to their customers. However, in many cases, there is not much explicit feedback about items which cripples typical recommender systems to operate efficiently and provide accurate recommendation. In this paper, a percentage of view approach is proposed to find relevant movies for customers. To prove the effectiveness of the approach, first, it is shown that this feature can be a good indicator of users’ like and dislike. Then the best approach is determined and used in a recommender system for Namava, a media service provider. Then the performance of this recommender system is compared to a random recommender system and the effectiveness of the approach is shown.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122682819","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8734911
M. M. AlyanNezhadi, Hossein Dabbaghan, Samira Moghani, M. Forghani
Over the past years, forgery paintings of famous artists have been sold as original. In order to spot the fake painting, the experts make the decision based on personal experience and with the help of examining some characteristics of the painting and painter. Applying the image processing methods to artwork can reduce the need of the expert and provide quick and reliable results to recognize the originality of the artwork. In this paper, the proposed method is able to identify the painter of artwork using image processing and data mining techniques. The method consists of two typical main stages, feature extraction, and classification. In the feature extraction, 11 statistical features are extracted from each image. These features have been selected in such a way that maximize the distinction of painters. In the second step, the painters are identified by hierarchical classification. In order to evaluate the performance of proposed method, it has applied to a collection of 348 paintings from eight Iranian artists. The method has been able to identify the artwork painter with the accuracy about 84.21%.
{"title":"A Painting Artist Recognition System Based on Image Processing and Hierarchical SVM","authors":"M. M. AlyanNezhadi, Hossein Dabbaghan, Samira Moghani, M. Forghani","doi":"10.1109/KBEI.2019.8734911","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8734911","url":null,"abstract":"Over the past years, forgery paintings of famous artists have been sold as original. In order to spot the fake painting, the experts make the decision based on personal experience and with the help of examining some characteristics of the painting and painter. Applying the image processing methods to artwork can reduce the need of the expert and provide quick and reliable results to recognize the originality of the artwork. In this paper, the proposed method is able to identify the painter of artwork using image processing and data mining techniques. The method consists of two typical main stages, feature extraction, and classification. In the feature extraction, 11 statistical features are extracted from each image. These features have been selected in such a way that maximize the distinction of painters. In the second step, the painters are identified by hierarchical classification. In order to evaluate the performance of proposed method, it has applied to a collection of 348 paintings from eight Iranian artists. The method has been able to identify the artwork painter with the accuracy about 84.21%.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123625305","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8735081
M. Kashani, S. Gorgin, S. V. Shojaedini
Each social network can be modeled as a graph G = [V, E] in which V is a vertex representing a person in this social network, and E is an edge representing the existence of a relationship between two individuals. The social infrastructure with size m is known as the Km group in the social network G with m individual. In other words, in a Km, each person knows other individuals and is in touch with them. The present study aims at developing a method for optimizing interpersonal communication in the social network using a simple cellular automaton algorithm. The experimental results obtained from both simulated social network and two real social networks were analyzed. The findings revealed that the proposed method has the potential to considerably reduce not only the number of colors assigned but also the running time of the program.
{"title":"Improvement of the Recognition of Relationships in Social Networks Using Complementary Graph Coloring Based on Cellular Automata","authors":"M. Kashani, S. Gorgin, S. V. Shojaedini","doi":"10.1109/KBEI.2019.8735081","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8735081","url":null,"abstract":"Each social network can be modeled as a graph G = [V, E] in which V is a vertex representing a person in this social network, and E is an edge representing the existence of a relationship between two individuals. The social infrastructure with size m is known as the Km group in the social network G with m individual. In other words, in a Km, each person knows other individuals and is in touch with them. The present study aims at developing a method for optimizing interpersonal communication in the social network using a simple cellular automaton algorithm. The experimental results obtained from both simulated social network and two real social networks were analyzed. The findings revealed that the proposed method has the potential to considerably reduce not only the number of colors assigned but also the running time of the program.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129788484","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8735039
M. Karamirad, C. Ghobadi, J. Nourinia, Bahman Mohamamdi
A single layer sinusoidal-cross type frequency selective surface (FSS) polarizer is presented that evaluated through transmission matrix method (TMM). It is demonstrated that the linear to circular polarization (CP) gets converted around 1 THz by the polarizer. The proposed polarizer is compact with a dimension of 0.4λ0 × 0.4λ0 × 0.08λ0 at the operating frequency. The simulation results show that the proposed polarizer operates within 0.95 THz to 1.1 THz. The polarizer has a 3 dB axial ratio (AR) fractional bandwidth of 20.31%. Moreover, the insertion loss is 1.2 dB at 1 THz. Results reveal that the AR of the presented structure is independent to angle of incident wave. Due to the strong circular conversion, compactness, and low sensibility to incidence angles, the presented linear to circular polarization convertor has sufficient potential in terahertz integrated components and communications.
{"title":"A THz Polarization Conversion Based on Sinusoidal Cross Pattern with Wide Angular Response","authors":"M. Karamirad, C. Ghobadi, J. Nourinia, Bahman Mohamamdi","doi":"10.1109/KBEI.2019.8735039","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8735039","url":null,"abstract":"A single layer sinusoidal-cross type frequency selective surface (FSS) polarizer is presented that evaluated through transmission matrix method (TMM). It is demonstrated that the linear to circular polarization (CP) gets converted around 1 THz by the polarizer. The proposed polarizer is compact with a dimension of 0.4λ0 × 0.4λ0 × 0.08λ0 at the operating frequency. The simulation results show that the proposed polarizer operates within 0.95 THz to 1.1 THz. The polarizer has a 3 dB axial ratio (AR) fractional bandwidth of 20.31%. Moreover, the insertion loss is 1.2 dB at 1 THz. Results reveal that the AR of the presented structure is independent to angle of incident wave. Due to the strong circular conversion, compactness, and low sensibility to incidence angles, the presented linear to circular polarization convertor has sufficient potential in terahertz integrated components and communications.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130068720","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8734945
Omid Alavi, S. Ahmadi, Abbas Hooshmand Viki
Over the past decades, power electronic circuits have been widely used in various equipment and industry sectors. Due to this widespread use, there is a lot of interest in understanding how to improve the lifespan of these circuits. Therefore, power modules have been evaluated as the most failure-prone component in these converters, and what is the most important in these components is the thermal management. In this paper, a simple method for determining the junction temperature of an IGBT module, taking into account the effects of operating temperature on the physical characteristics and behavior of the materials used within the module, is proposed. A resistance-capacitance (RC) thermal model is usually used to estimate the temperature, but the main drawback of the conventional model is that the values of the RC parameters are considered constant. In fact, thermal conductivity and specific heat capacity will change with temperature fluctuations. The proposed approach was implemented in the MATLAB environment by using a recursive loop, which modifies the RC values according to the prior temperature. The results showed that in the case study, the estimated junction temperature with variable RCs is 11.25% higher than the model with the fixed RC values.
{"title":"A Simplified Approach for Estimating Junction Temperature of IGBT Modules Considering Variable Thermal RCs","authors":"Omid Alavi, S. Ahmadi, Abbas Hooshmand Viki","doi":"10.1109/KBEI.2019.8734945","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8734945","url":null,"abstract":"Over the past decades, power electronic circuits have been widely used in various equipment and industry sectors. Due to this widespread use, there is a lot of interest in understanding how to improve the lifespan of these circuits. Therefore, power modules have been evaluated as the most failure-prone component in these converters, and what is the most important in these components is the thermal management. In this paper, a simple method for determining the junction temperature of an IGBT module, taking into account the effects of operating temperature on the physical characteristics and behavior of the materials used within the module, is proposed. A resistance-capacitance (RC) thermal model is usually used to estimate the temperature, but the main drawback of the conventional model is that the values of the RC parameters are considered constant. In fact, thermal conductivity and specific heat capacity will change with temperature fluctuations. The proposed approach was implemented in the MATLAB environment by using a recursive loop, which modifies the RC values according to the prior temperature. The results showed that in the case study, the estimated junction temperature with variable RCs is 11.25% higher than the model with the fixed RC values.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125377484","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8734955
Seyyed Mohammad Reza Hashemi, H. Hassanpour, Ehsan Kozegar
Today, face detection is of paramount importance due to its wide range of benefits and applications. Results of studies in this field have led to obtaining methods for face detection. In each of these methods, two factors of accuracy and runtime are important. The necessity of using accurate and quick methods has led to more studies in this regard. In face images, the individual’s face is not always directed toward the camera. Therefore, one of the important challenges in face detection is the correct detection when the angle of the image to the camera is not in the same direction and there are differences in the degree measure of the image angle, compared to the main image. The main issue in face detection, which has always been considered an important concept, is achieving the desired result based on the needs of security departments and scientific organizations. The present study aimed to evaluate the face detection methods in angular positions. In addition, one of the objectives of the research was assessing the advantages and disadvantages of each technique. In the end, the evaluated methods were compared on databases of ORL, ORL, and ORL in terms of accuracy of face detection. According to the results, the PCA+LDA+GABOR was the most efficient method due to special attention to viewing angles in different images that can result in high detection accuracy, compared to other methods.
{"title":"An Overview of Face Detection Methods in Angular Positions","authors":"Seyyed Mohammad Reza Hashemi, H. Hassanpour, Ehsan Kozegar","doi":"10.1109/KBEI.2019.8734955","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8734955","url":null,"abstract":"Today, face detection is of paramount importance due to its wide range of benefits and applications. Results of studies in this field have led to obtaining methods for face detection. In each of these methods, two factors of accuracy and runtime are important. The necessity of using accurate and quick methods has led to more studies in this regard. In face images, the individual’s face is not always directed toward the camera. Therefore, one of the important challenges in face detection is the correct detection when the angle of the image to the camera is not in the same direction and there are differences in the degree measure of the image angle, compared to the main image. The main issue in face detection, which has always been considered an important concept, is achieving the desired result based on the needs of security departments and scientific organizations. The present study aimed to evaluate the face detection methods in angular positions. In addition, one of the objectives of the research was assessing the advantages and disadvantages of each technique. In the end, the evaluated methods were compared on databases of ORL, ORL, and ORL in terms of accuracy of face detection. According to the results, the PCA+LDA+GABOR was the most efficient method due to special attention to viewing angles in different images that can result in high detection accuracy, compared to other methods.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117089658","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8734909
M. Bakhshipour, E. Rokrok, F. Namdari, M. Sedaghat
In this paper, a multi objective optimization method based on the swarm robotics search and rescue (SRSR) algorithm is provided to simultaneously determination of optimal capacity and location of DG units and capacitor banks along with network reconfiguration in the distribution system. Also the load uncertainty is considered in the proposed investigation based on triangular fuzzy method. The objective function considers six performance indexes including active losses, reactive losses, voltage deviation, voltage stability index, section loadability and index of balancing current of sections. Each of these indicators is weighed based on their importance and their composition forms in the objective function of the optimization problem. The proposed method has been studied and evaluated on IEEE 33 bus system, also the convergence and efficiency of the proposed SRSR approach is compared with two well-known genetic (GA) and cuckoo search (CS) algorithms. The obtained simulation results indicate the capability of the presented SRSR algorithm in determining the optimal capacity and location of the DG sources and capacitor banks in different operating conditions, and illuminate the improvement of system performance indexes, especially the reduction of active losses in the network.
{"title":"Optimal DG and capacitor allocation along with network reconfiguration using Swarm robotics search & rescue algorithm","authors":"M. Bakhshipour, E. Rokrok, F. Namdari, M. Sedaghat","doi":"10.1109/KBEI.2019.8734909","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8734909","url":null,"abstract":"In this paper, a multi objective optimization method based on the swarm robotics search and rescue (SRSR) algorithm is provided to simultaneously determination of optimal capacity and location of DG units and capacitor banks along with network reconfiguration in the distribution system. Also the load uncertainty is considered in the proposed investigation based on triangular fuzzy method. The objective function considers six performance indexes including active losses, reactive losses, voltage deviation, voltage stability index, section loadability and index of balancing current of sections. Each of these indicators is weighed based on their importance and their composition forms in the objective function of the optimization problem. The proposed method has been studied and evaluated on IEEE 33 bus system, also the convergence and efficiency of the proposed SRSR approach is compared with two well-known genetic (GA) and cuckoo search (CS) algorithms. The obtained simulation results indicate the capability of the presented SRSR algorithm in determining the optimal capacity and location of the DG sources and capacitor banks in different operating conditions, and illuminate the improvement of system performance indexes, especially the reduction of active losses in the network.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121348602","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8735083
A. Ghaznavi, S. Almodarresi
Multi-agent system control based on consensus algorithm is an important solution to coordinate distributed energy resources within micro and nanogrids. A significant challenge in the application of consensus algorithm is delay among agents particularly in networks with hubs. In this paper, a novel hierarchical consensus algorithm based on the idea of virtual reduction of hub agents is proposed. As a result, significant gain in consensus speed is obtained which offers a potential time margin for applying techniques against delay, communications failure and other risks to information exchange among agents. The algorithm application is studied in the coordination of distributed photovoltaic resources of several nanogrids as a possible scale-free network with hub via simulation as well.
{"title":"Robust Distributed Resource Aggregation Control in Smart Grid based on a Novel Hierarchical Consensus","authors":"A. Ghaznavi, S. Almodarresi","doi":"10.1109/KBEI.2019.8735083","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8735083","url":null,"abstract":"Multi-agent system control based on consensus algorithm is an important solution to coordinate distributed energy resources within micro and nanogrids. A significant challenge in the application of consensus algorithm is delay among agents particularly in networks with hubs. In this paper, a novel hierarchical consensus algorithm based on the idea of virtual reduction of hub agents is proposed. As a result, significant gain in consensus speed is obtained which offers a potential time margin for applying techniques against delay, communications failure and other risks to information exchange among agents. The algorithm application is studied in the coordination of distributed photovoltaic resources of several nanogrids as a possible scale-free network with hub via simulation as well.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121507095","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 : 2019-02-01DOI: 10.1109/KBEI.2019.8735018
Mahta Hassan pour Zonoozi, Donya Afshar Jahanshahi, Ali Mollaahmadi Dehaqi
Nowadays using authentication systems for twins’ recognition is very important and practical. Uni-biometric systems face some problems like unavailability, noise, the possibility of misleading and lack of uniqueness. So it’s better to use a multi-biometric system. Score level fusion has high performance and low time complexity among the existing levels of biometric fusion and it has been always the most interesting level for researchers. Since a complete physiological twins’ datasets have not been accessible, we have collected a multi-biometric dataset consist of fingerprint, face and iris biometrics of twins.
{"title":"Twins’ biometric fusion and introducing a new dataset","authors":"Mahta Hassan pour Zonoozi, Donya Afshar Jahanshahi, Ali Mollaahmadi Dehaqi","doi":"10.1109/KBEI.2019.8735018","DOIUrl":"https://doi.org/10.1109/KBEI.2019.8735018","url":null,"abstract":"Nowadays using authentication systems for twins’ recognition is very important and practical. Uni-biometric systems face some problems like unavailability, noise, the possibility of misleading and lack of uniqueness. So it’s better to use a multi-biometric system. Score level fusion has high performance and low time complexity among the existing levels of biometric fusion and it has been always the most interesting level for researchers. Since a complete physiological twins’ datasets have not been accessible, we have collected a multi-biometric dataset consist of fingerprint, face and iris biometrics of twins.","PeriodicalId":339990,"journal":{"name":"2019 5th Conference on Knowledge Based Engineering and Innovation (KBEI)","volume":"404 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127407957","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：