Pub Date : 2016-08-01DOI: 10.1109/FSKD.2016.7603493
Zujie Hu, Yifei Wei, Xiaojuan Wang, Mei Song
The explosive increase of wireless traffic volume and mobile terminal result in a great deal of greenhouse gas emission and energy consumption in telecommunications industry. In the recent years, it is absorbed in operators' wide attention that the base station supplied with renewable energy is introduced for cellular networks. In order to make the best of renewable energy in green base station, we formulate the problem of maximum usage ratio of renewable energy in the way of minimum cost by introducing green relay stations (GRS). In the paper, we propose a new scheme which will contribute to reduce the total energy consumption and improve the utilization of renewable energy. The proposed strategy is GRS assistant cell zooming scheme applied in LTE network using green and traditional macro base station. Simulation results and numerical analysis show that the proposed scheme can enormously improve the utilization of renewable energy and reduce the total energy consumption of traditional base station.
{"title":"Green relay station assisted cell zooming scheme for cellular networks","authors":"Zujie Hu, Yifei Wei, Xiaojuan Wang, Mei Song","doi":"10.1109/FSKD.2016.7603493","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603493","url":null,"abstract":"The explosive increase of wireless traffic volume and mobile terminal result in a great deal of greenhouse gas emission and energy consumption in telecommunications industry. In the recent years, it is absorbed in operators' wide attention that the base station supplied with renewable energy is introduced for cellular networks. In order to make the best of renewable energy in green base station, we formulate the problem of maximum usage ratio of renewable energy in the way of minimum cost by introducing green relay stations (GRS). In the paper, we propose a new scheme which will contribute to reduce the total energy consumption and improve the utilization of renewable energy. The proposed strategy is GRS assistant cell zooming scheme applied in LTE network using green and traditional macro base station. Simulation results and numerical analysis show that the proposed scheme can enormously improve the utilization of renewable energy and reduce the total energy consumption of traditional base station.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129780817","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603357
Peiguo Fu, Xiaohui Hu
Anomaly detection is a hot research field in the area of machine learning and data mining. The current outlier mining approaches which are based on the distance or the nearest neighbor are resulted in too long operation time results when using for the high-dimensional and massive data. Many improvements have been proposed to improve the results of the algorithms, but not yet satisfy the demand of the increasing data, the detection is ineffective. So, this paper presents a biased sampling-based of density anomaly detection algorithm. Firstly, In order to avoid complex kernel function estimation and integration, we divide the data set as grids and use the number of data points in the grid as an approximate density. In order to achieve the purpose of reducing the complexity of calculating the divided cluster, we use the hash table method to map the grid to the hash table unit while calculate the number of data points. After that we roll-up the neighbor grids which has the similar density in local and then calculate the approximate density of the combined data clusters. Next we use the probability-based biased sampling method to detect the data required detection to have a subset; then we use the method based on the density of local outlier detection to calculate the abnormal factor of each object in the subset. Because of using the biased sampling data, the abnormal factor both local outlier factor and global outlier factor; after we have the abnormal factor of each object in the subset, the higher the score of the point is, the higher the degree of outliers. The experiments on various artificial and real-life data sets confirm that, compared with the previous related methods, our method has better accuracy, scalability, and more efficient computation.
{"title":"Biased-sampling of density-based local outlier detection algorithm","authors":"Peiguo Fu, Xiaohui Hu","doi":"10.1109/FSKD.2016.7603357","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603357","url":null,"abstract":"Anomaly detection is a hot research field in the area of machine learning and data mining. The current outlier mining approaches which are based on the distance or the nearest neighbor are resulted in too long operation time results when using for the high-dimensional and massive data. Many improvements have been proposed to improve the results of the algorithms, but not yet satisfy the demand of the increasing data, the detection is ineffective. So, this paper presents a biased sampling-based of density anomaly detection algorithm. Firstly, In order to avoid complex kernel function estimation and integration, we divide the data set as grids and use the number of data points in the grid as an approximate density. In order to achieve the purpose of reducing the complexity of calculating the divided cluster, we use the hash table method to map the grid to the hash table unit while calculate the number of data points. After that we roll-up the neighbor grids which has the similar density in local and then calculate the approximate density of the combined data clusters. Next we use the probability-based biased sampling method to detect the data required detection to have a subset; then we use the method based on the density of local outlier detection to calculate the abnormal factor of each object in the subset. Because of using the biased sampling data, the abnormal factor both local outlier factor and global outlier factor; after we have the abnormal factor of each object in the subset, the higher the score of the point is, the higher the degree of outliers. The experiments on various artificial and real-life data sets confirm that, compared with the previous related methods, our method has better accuracy, scalability, and more efficient computation.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"2017 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129023623","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603274
Q. Li, Y. Mou, Junxia Guan, Qinghua Su, Beiping Wu, Haiming Wu
For solving numerical integral problems, a composite Simpson method based on Differential Evolution algorithm (S-DE) is proposed. The proposed method can be viewed as a piecewise integration method. It firstly uses the differential evolution algorithm (DE) to find the optimal segmentation points on the integral interval of an integrand. The approximate integral value of the integrand is then calculated by a composite Simpson method. The comparative analyses of numerical experiment results show the advantages of S-DE on a class of integral problems.
{"title":"Composite simpson method based on differential evolution algorithm for numerical integral","authors":"Q. Li, Y. Mou, Junxia Guan, Qinghua Su, Beiping Wu, Haiming Wu","doi":"10.1109/FSKD.2016.7603274","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603274","url":null,"abstract":"For solving numerical integral problems, a composite Simpson method based on Differential Evolution algorithm (S-DE) is proposed. The proposed method can be viewed as a piecewise integration method. It firstly uses the differential evolution algorithm (DE) to find the optimal segmentation points on the integral interval of an integrand. The approximate integral value of the integrand is then calculated by a composite Simpson method. The comparative analyses of numerical experiment results show the advantages of S-DE on a class of integral problems.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129348039","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603171
Lilue Fan, Aijia Ouyang
For its low efficiency in solving constrained optimization problems, the particle swarm optimization (PSO) is combined with immune algorithm (IA) in this paper. At the same time, an adaptive penalty function formula is designed to propose a hybrid immune PSO (HIPSO) algorithm for finding solution in constrained optimization problems. Through tests of 13 benchmark functions and three engineering optimization examples, it is clear that the performance of the HIPSO algorithm is equal to that of the HPSO algorithm. Whats more, the IA algorithm is not only better than IA algorithm and the PSO algorithm, but also co-evolutionary algorithm and other six kinds of algorithms.
{"title":"Hybrid immune PSO algorithm for engineering optimization problems","authors":"Lilue Fan, Aijia Ouyang","doi":"10.1109/FSKD.2016.7603171","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603171","url":null,"abstract":"For its low efficiency in solving constrained optimization problems, the particle swarm optimization (PSO) is combined with immune algorithm (IA) in this paper. At the same time, an adaptive penalty function formula is designed to propose a hybrid immune PSO (HIPSO) algorithm for finding solution in constrained optimization problems. Through tests of 13 benchmark functions and three engineering optimization examples, it is clear that the performance of the HIPSO algorithm is equal to that of the HPSO algorithm. Whats more, the IA algorithm is not only better than IA algorithm and the PSO algorithm, but also co-evolutionary algorithm and other six kinds of algorithms.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129437400","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603465
Tianchun Xu, Xiaohui Chen, Guo Wei, Weidong Wang
People count is an important indicator in video surveillance. Due to the overlapping objects and cluttered background, counting people accurately in actual crowded scene remains a non-trivial problem. Existing regression-based methods either learn a single model mapping the global feature to people count, or estimate localized count by training a large number of regressors. In this paper, we present an intermediate approach using the accumulated HOG feature. Our approach is able to capture the spatial difference of crowd structure and does not need to train a large number of regressors. Contrast to the low-level features existing regression-based methods generally use, the accumulated HOG feature is more robust. Extensive evaluations have been done on five benchmark datasets in the field of crowd counting, which demonstrate the robustness and effectiveness of our approach. In particular, the processing speed is fast enough to be applied to practical applications.
{"title":"Crowd counting using accumulated HOG","authors":"Tianchun Xu, Xiaohui Chen, Guo Wei, Weidong Wang","doi":"10.1109/FSKD.2016.7603465","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603465","url":null,"abstract":"People count is an important indicator in video surveillance. Due to the overlapping objects and cluttered background, counting people accurately in actual crowded scene remains a non-trivial problem. Existing regression-based methods either learn a single model mapping the global feature to people count, or estimate localized count by training a large number of regressors. In this paper, we present an intermediate approach using the accumulated HOG feature. Our approach is able to capture the spatial difference of crowd structure and does not need to train a large number of regressors. Contrast to the low-level features existing regression-based methods generally use, the accumulated HOG feature is more robust. Extensive evaluations have been done on five benchmark datasets in the field of crowd counting, which demonstrate the robustness and effectiveness of our approach. In particular, the processing speed is fast enough to be applied to practical applications.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"193 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126707183","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603192
Wei Wang, Lei Gao, Yu-Xiao Zhu, Hui Gao
Individuals can get the behavioral information from mass medias, as they always face various kinds of mass medias. Unfortunately, a systematical study about the effects of mass medias on social contagions is still lacking. To this end, we propose a novel non-Markovian social contagion model, in which individuals obtain the behavioral information not only from their neighbors but also from the mass medias. Through extensive numerical simulations, we find that the mass medias promote the adoption of behavior, and decrease the critical behavioral information transmission probability. In addition, we also note that the heterogeneity of degree distribution promotes the behavior adoption.
{"title":"Effects of mass medias on the dynamics of social contagions","authors":"Wei Wang, Lei Gao, Yu-Xiao Zhu, Hui Gao","doi":"10.1109/FSKD.2016.7603192","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603192","url":null,"abstract":"Individuals can get the behavioral information from mass medias, as they always face various kinds of mass medias. Unfortunately, a systematical study about the effects of mass medias on social contagions is still lacking. To this end, we propose a novel non-Markovian social contagion model, in which individuals obtain the behavioral information not only from their neighbors but also from the mass medias. Through extensive numerical simulations, we find that the mass medias promote the adoption of behavior, and decrease the critical behavioral information transmission probability. In addition, we also note that the heterogeneity of degree distribution promotes the behavior adoption.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129292667","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603452
S. Tsuruoka, Satoru Kimura, Kenji Hayakawa, H. Takase, H. Kawanaka
In this article, we aim to provide additional information related to students' understandings by analyzing their behaviors, especially typing their answers in short descriptive-quizzes. We collected typing processes for actual quizzes, asked self-confidence in the quizzes, and discuss their relationship between them. By some simple experiments, we find that typing that lacked confidence would cause suspension of typing process, and unpracticed students also cause suspensions, but they are shorter than the former case first case (less than 10 sec).
{"title":"Support teachers for quiz in large class — Analysis of typing processes for descriptive answers","authors":"S. Tsuruoka, Satoru Kimura, Kenji Hayakawa, H. Takase, H. Kawanaka","doi":"10.1109/FSKD.2016.7603452","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603452","url":null,"abstract":"In this article, we aim to provide additional information related to students' understandings by analyzing their behaviors, especially typing their answers in short descriptive-quizzes. We collected typing processes for actual quizzes, asked self-confidence in the quizzes, and discuss their relationship between them. By some simple experiments, we find that typing that lacked confidence would cause suspension of typing process, and unpracticed students also cause suspensions, but they are shorter than the former case first case (less than 10 sec).","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124188651","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603181
Yuandan Li, Shiwen Zhang, Zhiyong Li
Non-dominated sorting multi-objective optimization algorithms can constantly lead to the population of Pareto front optimal. However, the non-dominated sorting strategy lacks high capability to explore the Pareto front in the evolutionary subsequent process. We introduce a mixed strategy game model into evolutionary algorithms in this paper. Based on this strategy, we propose a novel multi-objective evolutionary algorithm (MSG-MOEA). A player adopts a strategy against the rest of the players with a certain probability in their respective strategy space instead of some specific strategy. According to the results of the game earning, the player constantly updates this probability to maximize the interest of his own objective. Through the players' constant pursuit of the maximal interest, a kind of tension could be brought to the population, which would push forward the population to the Pareto front. The proposed approach has been used some test functions and metrics for validation which are taken from the standard multi-objective optimization evolutionary literature. The experiment results have been compared against the NSGAII algorithm, which is one of the most highly competitive EMO algorithms. Algorithm analysis and simulation results show that the proposed algorithm performs well in solving complex multi-objective optimization problems.
{"title":"A multi-objective evolutionary algorithm based on mixed game strategy","authors":"Yuandan Li, Shiwen Zhang, Zhiyong Li","doi":"10.1109/FSKD.2016.7603181","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603181","url":null,"abstract":"Non-dominated sorting multi-objective optimization algorithms can constantly lead to the population of Pareto front optimal. However, the non-dominated sorting strategy lacks high capability to explore the Pareto front in the evolutionary subsequent process. We introduce a mixed strategy game model into evolutionary algorithms in this paper. Based on this strategy, we propose a novel multi-objective evolutionary algorithm (MSG-MOEA). A player adopts a strategy against the rest of the players with a certain probability in their respective strategy space instead of some specific strategy. According to the results of the game earning, the player constantly updates this probability to maximize the interest of his own objective. Through the players' constant pursuit of the maximal interest, a kind of tension could be brought to the population, which would push forward the population to the Pareto front. The proposed approach has been used some test functions and metrics for validation which are taken from the standard multi-objective optimization evolutionary literature. The experiment results have been compared against the NSGAII algorithm, which is one of the most highly competitive EMO algorithms. Algorithm analysis and simulation results show that the proposed algorithm performs well in solving complex multi-objective optimization problems.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"187 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123386449","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603231
Changjiang Shi, Qian Wang, Wencang Zhao
The polygonal representation method is put forward in the paper. The method is based on recursion and boundary division which describes the shape of the incomplete and overlapped weed seeds. The method extracts the contour shape features as local features using the scale space method. The local features are irrelevant to the position and orientation, at the same time, meet the scale, rotation and translation invariance. The incomplete and overlapped weed seeds are identified using the self-organization delayed neural network. The adjacent corner features are analyzed, compared and identified by spatial adjacency relationship among the angle characteristics. At last, the method was proved feasible the experiment by in recognizing the incomplete and overlapped weed seeds.
{"title":"Recognition of incomplete and overlapped weed seeds based on self-organization delayed neural network","authors":"Changjiang Shi, Qian Wang, Wencang Zhao","doi":"10.1109/FSKD.2016.7603231","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603231","url":null,"abstract":"The polygonal representation method is put forward in the paper. The method is based on recursion and boundary division which describes the shape of the incomplete and overlapped weed seeds. The method extracts the contour shape features as local features using the scale space method. The local features are irrelevant to the position and orientation, at the same time, meet the scale, rotation and translation invariance. The incomplete and overlapped weed seeds are identified using the self-organization delayed neural network. The adjacent corner features are analyzed, compared and identified by spatial adjacency relationship among the angle characteristics. At last, the method was proved feasible the experiment by in recognizing the incomplete and overlapped weed seeds.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123469506","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 : 2016-08-01DOI: 10.1109/FSKD.2016.7603244
Yongxing Jia, Chuanzhen Rong, Y. Wang, Ying Zhu, Yu Yang
Image fusion is the technology that combines more than one images into an image, to lay the foundation for further image processing tasks. The paper proposed a novel image fusion framework based on improved adaptive PCNN. PCNN is evolved from mammal's visual cortex neuron model, and characterized by its pulse synchronization and acquisition of the neurons. It has been proved that it is very suitable for the field of image processing, and it has been successfully applied in the field of image fusion. The two source images were input into two parallel PCNN networks, and the gray value of the image was used as the external stimuli of PCNN; At the same time, an improved Sum-modified-laplacian was selected as the image focus evaluation fuction, and linking strength of the corresponding neuron of the PCNN was calculated. The ignition map could be obtained after PCNN ignition, and the clearer part of the images were selected to generate the fused image by comparing the ignition map. In the end, the fused image was generated by pixel by pixel window-based consistency verification, and the final fusion result was obtained. Experimental results show that the proposed method is superior to the traditional image fusion methods in terms of subjective and objective evaluation criteria.
{"title":"A multi-focus image fusion algorithm using modified adaptive PCNN model","authors":"Yongxing Jia, Chuanzhen Rong, Y. Wang, Ying Zhu, Yu Yang","doi":"10.1109/FSKD.2016.7603244","DOIUrl":"https://doi.org/10.1109/FSKD.2016.7603244","url":null,"abstract":"Image fusion is the technology that combines more than one images into an image, to lay the foundation for further image processing tasks. The paper proposed a novel image fusion framework based on improved adaptive PCNN. PCNN is evolved from mammal's visual cortex neuron model, and characterized by its pulse synchronization and acquisition of the neurons. It has been proved that it is very suitable for the field of image processing, and it has been successfully applied in the field of image fusion. The two source images were input into two parallel PCNN networks, and the gray value of the image was used as the external stimuli of PCNN; At the same time, an improved Sum-modified-laplacian was selected as the image focus evaluation fuction, and linking strength of the corresponding neuron of the PCNN was calculated. The ignition map could be obtained after PCNN ignition, and the clearer part of the images were selected to generate the fused image by comparing the ignition map. In the end, the fused image was generated by pixel by pixel window-based consistency verification, and the final fusion result was obtained. Experimental results show that the proposed method is superior to the traditional image fusion methods in terms of subjective and objective evaluation criteria.","PeriodicalId":373155,"journal":{"name":"2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121675235","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：