Pub Date : 2015-03-03DOI: 10.1109/AISP.2015.7123519
Hossein Yeganeh Markid, Behrouz Zamani Dadaneh, M. Moghaddam
Feature selection is the process of choosing a subset of relevant as well as irredundant features from a bigger set. In other words, it removes redundant and irrelevant features from original set. In this paper, a new algorithm which is called bidirectional ant colony optimization feature selection (BDACOFS) based on ant colony optimization (ACO) algorithm and inspired from ACOFS (a recently proposed feature selection method) is presented. In the proposed algorithm, problem is modeled by a circular graph in which every node has only two arcs to its subsequent node. One of arcs represents selecting and another implies deselecting the next node. In addition, heuristic desirability of every node's selection is calculated according to two factors; one is related to discrimination ability of features and second one is related to mutual information among features. The proposed algorithm has been tested against some well-known datasets and its performance has been compared to some well-known algorithms. The result indicates that proposed algorithm by adding mutual statistical information to its heuristic desirability could remove more redundant features than original ACOFS. Meanwhile it keeps classification accuracy as highly as the original ACOFS.
{"title":"Bidirectional ant colony optimization for feature selection","authors":"Hossein Yeganeh Markid, Behrouz Zamani Dadaneh, M. Moghaddam","doi":"10.1109/AISP.2015.7123519","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123519","url":null,"abstract":"Feature selection is the process of choosing a subset of relevant as well as irredundant features from a bigger set. In other words, it removes redundant and irrelevant features from original set. In this paper, a new algorithm which is called bidirectional ant colony optimization feature selection (BDACOFS) based on ant colony optimization (ACO) algorithm and inspired from ACOFS (a recently proposed feature selection method) is presented. In the proposed algorithm, problem is modeled by a circular graph in which every node has only two arcs to its subsequent node. One of arcs represents selecting and another implies deselecting the next node. In addition, heuristic desirability of every node's selection is calculated according to two factors; one is related to discrimination ability of features and second one is related to mutual information among features. The proposed algorithm has been tested against some well-known datasets and its performance has been compared to some well-known algorithms. The result indicates that proposed algorithm by adding mutual statistical information to its heuristic desirability could remove more redundant features than original ACOFS. Meanwhile it keeps classification accuracy as highly as the original ACOFS.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122873695","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123529
Amir Bidokhti, S. Ghaemmaghami
With rapid proliferation of affordable video capturing devices and state-of-the-art video editing software tools, it is now easier than ever to manipulate video contents. In this paper a passive method for copy/move video forgery detection in MPEG videos is proposed. The method first divides each video frame into suspicious and apparently innocent parts. Subsequently, an optical flow coefficient is computed from each part. Forgeries are located when an unusual trend in the optical flow coefficient of the suspicious object is detected. Experiments on a set of forged and original sequences validate the justifications made by the proposed method.
{"title":"Detection of regional copy/move forgery in MPEG videos using optical flow","authors":"Amir Bidokhti, S. Ghaemmaghami","doi":"10.1109/AISP.2015.7123529","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123529","url":null,"abstract":"With rapid proliferation of affordable video capturing devices and state-of-the-art video editing software tools, it is now easier than ever to manipulate video contents. In this paper a passive method for copy/move video forgery detection in MPEG videos is proposed. The method first divides each video frame into suspicious and apparently innocent parts. Subsequently, an optical flow coefficient is computed from each part. Forgeries are located when an unusual trend in the optical flow coefficient of the suspicious object is detected. Experiments on a set of forged and original sequences validate the justifications made by the proposed method.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123561118","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123510
Reza Behinfaraz, M. Badamchizadeh
This paper proposes a new method for synchronization of two different fractional-order chaotic systems. By using fractional calculus properties and some result of the stability theorem of fractional-order systems, we suggest a new method to achieve the synchronization in such cases. The analytical conditions for synchronization of these different fractional-order systems are derived by utilizing Laplace transform. For transforming our problem into a general synchronization between fractional-order chaotic systems with equal orders, we used fractional operators in the controller, and nonlinear feedback control is suggested by using of the active control method concepts. We present an example that illustrate the performance and application of proposed method.
{"title":"New approach to synchronization of two different fractional-order chaotic systems","authors":"Reza Behinfaraz, M. Badamchizadeh","doi":"10.1109/AISP.2015.7123510","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123510","url":null,"abstract":"This paper proposes a new method for synchronization of two different fractional-order chaotic systems. By using fractional calculus properties and some result of the stability theorem of fractional-order systems, we suggest a new method to achieve the synchronization in such cases. The analytical conditions for synchronization of these different fractional-order systems are derived by utilizing Laplace transform. For transforming our problem into a general synchronization between fractional-order chaotic systems with equal orders, we used fractional operators in the controller, and nonlinear feedback control is suggested by using of the active control method concepts. We present an example that illustrate the performance and application of proposed method.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122505434","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123536
A. Noroozi, R. Malekzadeh
One of the ways to enhance the information retrieval performance is query expansion (QE) which means adding some terms to the query in order to reduce mismatch between information needs and retrieved documents. In this way “Query Drift” occurring for ambiguous queries is a common problem. Special case of this problem is “Outweighting” that usually occurs for long queries, that is, some augmented words strongly related to an individual query words but not to the all. In this paper we propose a new method for QE to reduce the effects of disambiguated query terms and decrease query drifting. In proposed method for word outweighting elimination, query terms are grouped based on their semantic relationships. For each group, candidates are fetched from WordNet that relates to the all of words group. Then by using recursive structure of Hopfield network words with the most relationship with other words are selected. Moreover, the Term Semantic Network has used to overcome some of the shortcomings of WordNet. Evaluation results on CACM and CERC test collections show that the proposed method is effective and improve 4% and 12% of Mean Average Precision respectively.
{"title":"Integration of recursive structure of hopfield and ontologies for query expansion","authors":"A. Noroozi, R. Malekzadeh","doi":"10.1109/AISP.2015.7123536","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123536","url":null,"abstract":"One of the ways to enhance the information retrieval performance is query expansion (QE) which means adding some terms to the query in order to reduce mismatch between information needs and retrieved documents. In this way “Query Drift” occurring for ambiguous queries is a common problem. Special case of this problem is “Outweighting” that usually occurs for long queries, that is, some augmented words strongly related to an individual query words but not to the all. In this paper we propose a new method for QE to reduce the effects of disambiguated query terms and decrease query drifting. In proposed method for word outweighting elimination, query terms are grouped based on their semantic relationships. For each group, candidates are fetched from WordNet that relates to the all of words group. Then by using recursive structure of Hopfield network words with the most relationship with other words are selected. Moreover, the Term Semantic Network has used to overcome some of the shortcomings of WordNet. Evaluation results on CACM and CERC test collections show that the proposed method is effective and improve 4% and 12% of Mean Average Precision respectively.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"84 9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131069156","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123511
S. M. Tabatabaei, Abdollah Chalechale
Local binary pattern (LBP) is a simple and computationally efficient texture descriptor which has been attracting many attentions since its introduction; due to the extensive research done in this regard, diverse variants of LBP have been introduced in recent years. While original form of this operator encodes structures like spots, edges, and corners in form of a binary code, a more recent type of LBP called high order directional derivative LBP (DLBP) reveals some alternative structures such as convexities and concavities. Even though these structures are important features in the images, another significant consideration is the relationship between them. For instance, there is a high probability that an edge structure be present near another one. In this paper, we have introduced a novel texture descriptor named HJDLBP (high order joint DLBP) which is able to encode relationships between micro patterns in addition to the prevalent structures. To evaluate the proposed descriptor, we have considered two renowned JAFFE and YALE facial image databases and then exploited the proposed texture descriptor for face recognition issues. The experiments are implemented in software in the following manner: as a first step, the face part of each image is segmented from its background using Viola and Jones algorithm. Afterward, the micro patterns and relationships between them are extracted from rectangularly partitioned face images; and their histograms are constructed as well. Finally, a group of SVMs are trained for classification. We have compared obtained results using the new operator with the results attained when conventional LBP and high order DLBP are applied for feature extraction from image blocks. The comparative results show the efficacy of the proposed operator as a texture descriptor.
局部二值模式(LBP)是一种简单、计算效率高的纹理描述符，自提出以来一直受到人们的关注;由于在这方面进行了广泛的研究，近年来出现了多种LBP变体。虽然该算子的原始形式以二进制编码的形式编码点、边和角等结构，但最近一种称为高阶方向导数LBP (DLBP)的LBP揭示了一些替代结构，如凸和凹。尽管这些结构是图像中的重要特征，但另一个重要的考虑是它们之间的关系。例如，一个边缘结构很有可能出现在另一个边缘结构附近。本文提出了一种新的纹理描述符HJDLBP (high order joint DLBP)，该描述符除了可以编码流行结构外，还可以编码微图案之间的关系。为了评估所提出的描述符，我们考虑了两个著名的JAFFE和YALE面部图像数据库，然后利用所提出的纹理描述符进行人脸识别问题。实验在软件中实现的方式如下:第一步，使用Viola和Jones算法将每张图像的人脸部分从背景中分割出来。然后，从矩形分割的人脸图像中提取微模式及其相互关系;它们的直方图也被构造出来了。最后，训练一组支持向量机进行分类。我们将使用新算子获得的结果与使用传统LBP和高阶DLBP从图像块中提取特征的结果进行了比较。对比结果表明了该算子作为纹理描述符的有效性。
{"title":"HJDLBP: A novel texture descriptor and its application in face recognition","authors":"S. M. Tabatabaei, Abdollah Chalechale","doi":"10.1109/AISP.2015.7123511","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123511","url":null,"abstract":"Local binary pattern (LBP) is a simple and computationally efficient texture descriptor which has been attracting many attentions since its introduction; due to the extensive research done in this regard, diverse variants of LBP have been introduced in recent years. While original form of this operator encodes structures like spots, edges, and corners in form of a binary code, a more recent type of LBP called high order directional derivative LBP (DLBP) reveals some alternative structures such as convexities and concavities. Even though these structures are important features in the images, another significant consideration is the relationship between them. For instance, there is a high probability that an edge structure be present near another one. In this paper, we have introduced a novel texture descriptor named HJDLBP (high order joint DLBP) which is able to encode relationships between micro patterns in addition to the prevalent structures. To evaluate the proposed descriptor, we have considered two renowned JAFFE and YALE facial image databases and then exploited the proposed texture descriptor for face recognition issues. The experiments are implemented in software in the following manner: as a first step, the face part of each image is segmented from its background using Viola and Jones algorithm. Afterward, the micro patterns and relationships between them are extracted from rectangularly partitioned face images; and their histograms are constructed as well. Finally, a group of SVMs are trained for classification. We have compared obtained results using the new operator with the results attained when conventional LBP and high order DLBP are applied for feature extraction from image blocks. The comparative results show the efficacy of the proposed operator as a texture descriptor.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130037272","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123502
A. Golkar, S. Jafari, M. Golkar, Seyed Mohammad Sadegh Dashti, S. M. Fakhrahmad
In this paper, the role of nouns in reducing the conceptual density of contexts has been examined. A new method is proposed to identify and prune nouns with negative impact on conceptual density of contexts. In the proposed method, a fitness function is offered; a fitness degree is assigned to unambiguous nouns sense within the context. Using the mean fitness degree of unambiguous nouns' sense, a threshold is produced for that context. This threshold is then used as a measure to prune the sense of nouns with lower fitness degree that reduces the conceptual density of the context. Finally, by implementing this method on the contexts produced by conceptual density method, all contexts will be optimized significantly; this significantly increases the accuracy of disambiguation.
{"title":"Improve word sense disambiguation by proposing a pruning method for optimizing conceptual density's contexts","authors":"A. Golkar, S. Jafari, M. Golkar, Seyed Mohammad Sadegh Dashti, S. M. Fakhrahmad","doi":"10.1109/AISP.2015.7123502","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123502","url":null,"abstract":"In this paper, the role of nouns in reducing the conceptual density of contexts has been examined. A new method is proposed to identify and prune nouns with negative impact on conceptual density of contexts. In the proposed method, a fitness function is offered; a fitness degree is assigned to unambiguous nouns sense within the context. Using the mean fitness degree of unambiguous nouns' sense, a threshold is produced for that context. This threshold is then used as a measure to prune the sense of nouns with lower fitness degree that reduces the conceptual density of the context. Finally, by implementing this method on the contexts produced by conceptual density method, all contexts will be optimized significantly; this significantly increases the accuracy of disambiguation.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130284045","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123482
Fahimeh Sadat Saleh, R. Azmi
Skin lesion segmentation is one of the most important steps in automated early skin cancer detection, since the accuracy of the following steps significantly depends on it. In this paper, a two-stage approach based on Mean Shift and spectral graph partitioning algorithms is proposed. This method effectively extracts lesion borders. Moreover, a distinctive advantage of this approach is extracting the region of interest levels that is not addressed in pervious state of the art methods. In the first stage, the image is segmented to regions using Mean Shift algorithm. In the second stage, a graph-based representation is used to demonstrate the structure of the extracted regions and their relationships. Afterwards a clustering process is applied, considering the neighborhood system and analyzing the color and texture distance between regions. The proposed method is applied to 170 dermoscopic images and evaluated with two different metrics. This evaluation has performed by means of the segmentation results provided by an experienced dermatologist as the ground truth. Experiments demonstrate that in this method, challenging features of skin lesions are handled as might be expected when compared to five state of the art methods.
{"title":"Automatic multiple regions segmentation of dermoscopy images","authors":"Fahimeh Sadat Saleh, R. Azmi","doi":"10.1109/AISP.2015.7123482","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123482","url":null,"abstract":"Skin lesion segmentation is one of the most important steps in automated early skin cancer detection, since the accuracy of the following steps significantly depends on it. In this paper, a two-stage approach based on Mean Shift and spectral graph partitioning algorithms is proposed. This method effectively extracts lesion borders. Moreover, a distinctive advantage of this approach is extracting the region of interest levels that is not addressed in pervious state of the art methods. In the first stage, the image is segmented to regions using Mean Shift algorithm. In the second stage, a graph-based representation is used to demonstrate the structure of the extracted regions and their relationships. Afterwards a clustering process is applied, considering the neighborhood system and analyzing the color and texture distance between regions. The proposed method is applied to 170 dermoscopic images and evaluated with two different metrics. This evaluation has performed by means of the segmentation results provided by an experienced dermatologist as the ground truth. Experiments demonstrate that in this method, challenging features of skin lesions are handled as might be expected when compared to five state of the art methods.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124839046","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123513
Razieh Falahian, M. M. Dastjerdi, S. Gharibzadeh
The recent meteoric significant developments in the biological and medical sciences have been the culmination of substantial efforts devoted to precisely modeling the behavior of biological systems and their responses to various stimuli. The complicated interactions within varied components of biological systems as well as with their environments make them extremely complex nonlinear systems. The results of several contemporary relevant investigations have manifested their chaotic behavioral patterns. With the aim of modeling this specific behavior of bio-systems, we employ a particular multilayer feed-forward neural network. The distinctive feature of our modeling method, which makes it dominant within the modeling techniques, is training the select neural network with the chaotic map extracted from the under-study time series. Our results, which are briefly represented in this paper, confirm that the specified neural network does possess the potentiality to model the chaotic dynamics of biological systems., even in the presence of noise. In pursuance of evaluating our model, we assess and model the chaotic response of the brain to the flicker light through some recorded electroretinogram data.
{"title":"Authentic modeling of complex dynamics of biological systems by the manipulation of artificial intelligence","authors":"Razieh Falahian, M. M. Dastjerdi, S. Gharibzadeh","doi":"10.1109/AISP.2015.7123513","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123513","url":null,"abstract":"The recent meteoric significant developments in the biological and medical sciences have been the culmination of substantial efforts devoted to precisely modeling the behavior of biological systems and their responses to various stimuli. The complicated interactions within varied components of biological systems as well as with their environments make them extremely complex nonlinear systems. The results of several contemporary relevant investigations have manifested their chaotic behavioral patterns. With the aim of modeling this specific behavior of bio-systems, we employ a particular multilayer feed-forward neural network. The distinctive feature of our modeling method, which makes it dominant within the modeling techniques, is training the select neural network with the chaotic map extracted from the under-study time series. Our results, which are briefly represented in this paper, confirm that the specified neural network does possess the potentiality to model the chaotic dynamics of biological systems., even in the presence of noise. In pursuance of evaluating our model, we assess and model the chaotic response of the brain to the flicker light through some recorded electroretinogram data.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116786203","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123485
Ali Akbar Nasiri, M. Fathy
In this paper a novel approach is proposed to detect reference point for fingerprint images. Reference point extraction is a key component in automatic fingerprint identification and recognition systems. A new method was proposed for fingerprint reference point extraction, based on field flow curve and clustering. High curvature points in the flow curves are used in our reference point detection. Because we use flow curve instead of ridge for reference point detection, our method is robust to noise and has a good result on fingerprint image with low quality. Also our method has the ability to detect a reference point for an arch class fingerprint which is hard for other methods to detect it. The experiments are conducted on FVC2002-DB2a and FVC2004 to measure the performance of our reference point detection. Experimental results show that our algorithm is robust and it has better results than other approaches.
{"title":"An effective algorithm for fingerprint reference point detection based on filed flow curves","authors":"Ali Akbar Nasiri, M. Fathy","doi":"10.1109/AISP.2015.7123485","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123485","url":null,"abstract":"In this paper a novel approach is proposed to detect reference point for fingerprint images. Reference point extraction is a key component in automatic fingerprint identification and recognition systems. A new method was proposed for fingerprint reference point extraction, based on field flow curve and clustering. High curvature points in the flow curves are used in our reference point detection. Because we use flow curve instead of ridge for reference point detection, our method is robust to noise and has a good result on fingerprint image with low quality. Also our method has the ability to detect a reference point for an arch class fingerprint which is hard for other methods to detect it. The experiments are conducted on FVC2002-DB2a and FVC2004 to measure the performance of our reference point detection. Experimental results show that our algorithm is robust and it has better results than other approaches.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115694810","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 : 2015-03-03DOI: 10.1109/AISP.2015.7123489
Maryam Karimi, M. Dehghan, Seyyed Majid Nourhoseini
Delay sensitive applications need to overcome the service problems in dynamic environments with respect to both the multimedia source data (e.g., variable bit-rate) and the wireless channels (e.g., fading channel). This paper considers the problem of point to point transmission of scalable video coding over a fading channel. We formulate the rate adaptation challenge of WLAN multimedia networks as a Markov Decision Process and resolve this problem online based on reinforcement learning. The buffer state, channel state, and video state were considered as a joint state of system to maximize the average Quality of Service under delay constraints. To improve the convergence speed of learning, system's underlying dynamics were partitioned into a priori known and a priori unknown components. The proposed learning algorithm exploits known information about the system, so that less information needs to be learned compared with that in conventional reinforcement learning algorithms.
{"title":"A systematic framework for dynamically optimizing delay-sensitive wireless transmission","authors":"Maryam Karimi, M. Dehghan, Seyyed Majid Nourhoseini","doi":"10.1109/AISP.2015.7123489","DOIUrl":"https://doi.org/10.1109/AISP.2015.7123489","url":null,"abstract":"Delay sensitive applications need to overcome the service problems in dynamic environments with respect to both the multimedia source data (e.g., variable bit-rate) and the wireless channels (e.g., fading channel). This paper considers the problem of point to point transmission of scalable video coding over a fading channel. We formulate the rate adaptation challenge of WLAN multimedia networks as a Markov Decision Process and resolve this problem online based on reinforcement learning. The buffer state, channel state, and video state were considered as a joint state of system to maximize the average Quality of Service under delay constraints. To improve the convergence speed of learning, system's underlying dynamics were partitioned into a priori known and a priori unknown components. The proposed learning algorithm exploits known information about the system, so that less information needs to be learned compared with that in conventional reinforcement learning algorithms.","PeriodicalId":405857,"journal":{"name":"2015 The International Symposium on Artificial Intelligence and Signal Processing (AISP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130898465","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：