It is a typical multi-objective optimization problem for the scientific decision of bidding to seek cooperating partner in virtual enterprise. With the optimization model proposed, partner selection is solved by the improved genetic algorithm. In the evolution process, individual survive rate is dynamic according to queue of individuals 'fitness values before roulette wheel selection, avoiding premature convergence. Crossover and mutation operators are accordingly adaptive to fitness value and iterative degree, which endows individuals with self- adaptability with the variation of the environment. Finally, the example demonstrates the validity of the adaptive genetic algorithm.
{"title":"Multi-objective Optimization in Partner Selection","authors":"Xuesen Ma, Jianghong Han, Zhengfeng Hou, Zhenchun Wei","doi":"10.1109/ICNC.2007.485","DOIUrl":"https://doi.org/10.1109/ICNC.2007.485","url":null,"abstract":"It is a typical multi-objective optimization problem for the scientific decision of bidding to seek cooperating partner in virtual enterprise. With the optimization model proposed, partner selection is solved by the improved genetic algorithm. In the evolution process, individual survive rate is dynamic according to queue of individuals 'fitness values before roulette wheel selection, avoiding premature convergence. Crossover and mutation operators are accordingly adaptive to fitness value and iterative degree, which endows individuals with self- adaptability with the variation of the environment. Finally, the example demonstrates the validity of the adaptive genetic algorithm.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"296 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126689494","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}
The oilfield remaining oil distribution forecast is called world-level difficult problems by oil domain specialists in the world. The source of low forecast correctness are only consider objective evidences or subjective evidence, so the forecast results still exist limitation, it result in low accuracy, reliability and so on to identify the classification characteristics and to compute quantitative parameters. So, how to fuse all objective evidences and subjective evidences is a key problem to research remaining oil distribution. A new model is proposed, it integrated BP neural networks combination models and two-level D-S evidence reasoning models, the exact classification results are implemented about many remaining oil distribution characteristics. The classification output reliability of each BP network and the reasoning result reliability of each domain fuzzy expert system are regarded as basic probability assignment of input evidence in D-S evidence reasoning model. The model has applied successfully in Daqing Oilfield of China.
{"title":"Neural Evidence Integration Model and Its Application","authors":"Shouzhi Wei, N. Jin, Hui Liu","doi":"10.1109/ICNC.2007.494","DOIUrl":"https://doi.org/10.1109/ICNC.2007.494","url":null,"abstract":"The oilfield remaining oil distribution forecast is called world-level difficult problems by oil domain specialists in the world. The source of low forecast correctness are only consider objective evidences or subjective evidence, so the forecast results still exist limitation, it result in low accuracy, reliability and so on to identify the classification characteristics and to compute quantitative parameters. So, how to fuse all objective evidences and subjective evidences is a key problem to research remaining oil distribution. A new model is proposed, it integrated BP neural networks combination models and two-level D-S evidence reasoning models, the exact classification results are implemented about many remaining oil distribution characteristics. The classification output reliability of each BP network and the reasoning result reliability of each domain fuzzy expert system are regarded as basic probability assignment of input evidence in D-S evidence reasoning model. The model has applied successfully in Daqing Oilfield of China.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130621480","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}
Technology applying neural network to detection of high resolution radar is advised. Firstly, it analyses the principle of detection technique of high resolution radar and the conception of "distance corridor". Secondly, it introduces the basal principle of BP algorithm. The principle and structure of three layers BP neural network is analysed. Thirdly, the detection technology and algorithm of high resolution radar based on BP neural network is researched. The result of research of target detection technology based on BP neural network possesses superperformance, which is in view of four distance corridor and ten distance corridor, which are the neural network that possesses four input neurons and ten input neurons. At last, the research is carried out by Matlab, the result are visible and understandable.
{"title":"Study of Detection Technique Simulation of High Resolution Radar Based on BP Neural Network","authors":"Hou Xuan, He Mingyi","doi":"10.1109/ICNC.2007.684","DOIUrl":"https://doi.org/10.1109/ICNC.2007.684","url":null,"abstract":"Technology applying neural network to detection of high resolution radar is advised. Firstly, it analyses the principle of detection technique of high resolution radar and the conception of \"distance corridor\". Secondly, it introduces the basal principle of BP algorithm. The principle and structure of three layers BP neural network is analysed. Thirdly, the detection technology and algorithm of high resolution radar based on BP neural network is researched. The result of research of target detection technology based on BP neural network possesses superperformance, which is in view of four distance corridor and ten distance corridor, which are the neural network that possesses four input neurons and ten input neurons. At last, the research is carried out by Matlab, the result are visible and understandable.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123835062","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}
In this paper, the authors propose a new procedure for copyright protection by using a bio-inspired wavelet based data hiding approach. The proposed method takes advantage of human visual system (HVS) characteristics to provide better watermarked image quality. It also exploits visual secret sharing (VSS) technique to guarantee the security of the procedure. Performance improvement with respect to the existing algorithms is obtained by genetic algorithm (GA) optimization. The experimental results show that the proposed algorithm yields a watermark that is invisible to human eyes and is robust to various intentional and unintentional attacks. The experimental results are also compared with the results of some previous works.
{"title":"A Bio-Inspired Content Adaptive Approach for Multiresolution-Based Image Watermarking","authors":"E. Vahedi, C. Lucas, R. Zoroofi, M. Shiva","doi":"10.1109/ICNC.2007.4","DOIUrl":"https://doi.org/10.1109/ICNC.2007.4","url":null,"abstract":"In this paper, the authors propose a new procedure for copyright protection by using a bio-inspired wavelet based data hiding approach. The proposed method takes advantage of human visual system (HVS) characteristics to provide better watermarked image quality. It also exploits visual secret sharing (VSS) technique to guarantee the security of the procedure. Performance improvement with respect to the existing algorithms is obtained by genetic algorithm (GA) optimization. The experimental results show that the proposed algorithm yields a watermark that is invisible to human eyes and is robust to various intentional and unintentional attacks. The experimental results are also compared with the results of some previous works.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123893258","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}
In order to distinguish faces of various angles during face recognition, an algorithm of the combination of approximate dynamic programming (ADP) which is called action dependent heuristic dynamic programming (ADHDP) and particle swarm optimization (PSO) is presented and used, that is to say, ADP is applied for dynamically changing the values of the PSO parameters. During the process of face recognition, the discrete cosine transformation (DCT) is first introduced to reduce negative effects. Then K-L transformation can be used to compress images and decrease data dimensions. According to principal component analysis (PCA), main parts of vectors are extracted for data representation. Finally, radial basis function (RBF) neural network is enrolled to recognize various faces. The training of RBF neural network is exploited by ADP-PSO. In terms of ORL face database, the experimental result gives a clear view of its highly accurate efficiency.
{"title":"An Effective Hybrid ADP-PSO Strategy for Optimization and Its Application to Face Recognition","authors":"Yongzhong Lu","doi":"10.1109/ICNC.2007.188","DOIUrl":"https://doi.org/10.1109/ICNC.2007.188","url":null,"abstract":"In order to distinguish faces of various angles during face recognition, an algorithm of the combination of approximate dynamic programming (ADP) which is called action dependent heuristic dynamic programming (ADHDP) and particle swarm optimization (PSO) is presented and used, that is to say, ADP is applied for dynamically changing the values of the PSO parameters. During the process of face recognition, the discrete cosine transformation (DCT) is first introduced to reduce negative effects. Then K-L transformation can be used to compress images and decrease data dimensions. According to principal component analysis (PCA), main parts of vectors are extracted for data representation. Finally, radial basis function (RBF) neural network is enrolled to recognize various faces. The training of RBF neural network is exploited by ADP-PSO. In terms of ORL face database, the experimental result gives a clear view of its highly accurate efficiency.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123191079","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}
A new technique for facial expression recognition is proposed, which uses active appearance model (AAM) to extract facial feature points and combines useful local shape features to form a classifier. To enhance performance of AAM, we use Adaboost to locate eye position to initialize AAM. After extraction of facial feature points, we analyze local facial changes and use some simple features to form an effective classifier. At last, we demonstrate our approach by experiments.
{"title":"Facial Expression Recognition using AAM and Local Facial Features","authors":"Fangqi Tang, Benzai Deng","doi":"10.1109/ICNC.2007.373","DOIUrl":"https://doi.org/10.1109/ICNC.2007.373","url":null,"abstract":"A new technique for facial expression recognition is proposed, which uses active appearance model (AAM) to extract facial feature points and combines useful local shape features to form a classifier. To enhance performance of AAM, we use Adaboost to locate eye position to initialize AAM. After extraction of facial feature points, we analyze local facial changes and use some simple features to form an effective classifier. At last, we demonstrate our approach by experiments.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"112 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123661987","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}
A multi-subpopulation accelerating genetic algorithm based on attractors(MAGA) is proposed to cope with the drawback of genetic algorithms. MAGA views the excellent individuals as attractors and generates local small populations in the neighbor of them to maintain the diversity of the population. In the course of searching, MAGA constantly shrinks the searching neighbor and uses the accelerating operators to speed up the evolution of MAGA. The convergence analysis shows MAGA can converge to global optimization under some circumstances. Finally, MAGA's efficiency is validated through optimization of two benchmark functions.
{"title":"A Multi-subpopulation Accelerating Genetic Algorithm Based on Attractors (MAGA): Performance in Function Optimization","authors":"Zhiyi Lin, Yuanxiang Li","doi":"10.1109/ICNC.2007.73","DOIUrl":"https://doi.org/10.1109/ICNC.2007.73","url":null,"abstract":"A multi-subpopulation accelerating genetic algorithm based on attractors(MAGA) is proposed to cope with the drawback of genetic algorithms. MAGA views the excellent individuals as attractors and generates local small populations in the neighbor of them to maintain the diversity of the population. In the course of searching, MAGA constantly shrinks the searching neighbor and uses the accelerating operators to speed up the evolution of MAGA. The convergence analysis shows MAGA can converge to global optimization under some circumstances. Finally, MAGA's efficiency is validated through optimization of two benchmark functions.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121422353","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}
This paper proposes an improved face recognition method based on the combination of Principal Component Analysis and Neural Networks. This method adopts Principal Component Analysis (PCA) to abstract principal eigenvectors of the image in order to get best feature description, hence to reduce the number of inputs of neural networks. After this, these image data of reduced dimensions are input into a feed forward neural network to be trained. The weights of neural networks are optimized using Particle Swarm Optimization (PSO) algorithm. Then this well-trained network is tested using samples from standard human face database. The results show that this method gains higher recognition rate in contrast with some other methods.
{"title":"Human Face Recognition Based on Principal Component Analysis and Particle Swarm Optimization-BP Neural Network ","authors":"Lei Du, Zhenhong Jia, Liang Xue","doi":"10.1109/ICNC.2007.418","DOIUrl":"https://doi.org/10.1109/ICNC.2007.418","url":null,"abstract":"This paper proposes an improved face recognition method based on the combination of Principal Component Analysis and Neural Networks. This method adopts Principal Component Analysis (PCA) to abstract principal eigenvectors of the image in order to get best feature description, hence to reduce the number of inputs of neural networks. After this, these image data of reduced dimensions are input into a feed forward neural network to be trained. The weights of neural networks are optimized using Particle Swarm Optimization (PSO) algorithm. Then this well-trained network is tested using samples from standard human face database. The results show that this method gains higher recognition rate in contrast with some other methods.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114200201","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}
In biomathematics, singularly perturbed predator-prey systems are of common occurrence. A singularly perturbed problem with nonlinear predator-prey reaction diffusion system in 2 dimension is studied. The system changes rapidly near initial time layer. Traditional numerical method failed to simulate the system. Numerical simulation of this kind of system is rare so far, this motives us to consider novel simulation technique. Firstly stretched variable is introduced so that the analytic solution is decomposed into the reduced solution and the initial layer correction solution. Secondly, the nonlinearization process of the reduced problem system is proposed. Thirdly, two numerical method, stretched variable method and Shishkin- type method, are constructed. Finally, simulation example is studied to demonstrate that both stretched variable method and Shishkin-type method are efficient computational method. Shishkin-type method is more practical in use for this kind of complicated system.
{"title":"Numerical Simulation Technique for Nonlinear Singularly Perturbed Predator-Prey Reaction Diffusion System in Biomathematics","authors":"X. Cai, Zhongdi Cen","doi":"10.1109/ICNC.2007.507","DOIUrl":"https://doi.org/10.1109/ICNC.2007.507","url":null,"abstract":"In biomathematics, singularly perturbed predator-prey systems are of common occurrence. A singularly perturbed problem with nonlinear predator-prey reaction diffusion system in 2 dimension is studied. The system changes rapidly near initial time layer. Traditional numerical method failed to simulate the system. Numerical simulation of this kind of system is rare so far, this motives us to consider novel simulation technique. Firstly stretched variable is introduced so that the analytic solution is decomposed into the reduced solution and the initial layer correction solution. Secondly, the nonlinearization process of the reduced problem system is proposed. Thirdly, two numerical method, stretched variable method and Shishkin- type method, are constructed. Finally, simulation example is studied to demonstrate that both stretched variable method and Shishkin-type method are efficient computational method. Shishkin-type method is more practical in use for this kind of complicated system.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116199912","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}
Risk management project is an important aspect of general project risk element transmission theory. Traditional network planning technology encountered great obstacles in project risk management issues, and often unable to accurately forecast the risk, resulting great loss of costs. To address the cost-time optimization problem considering the risk elements, this paper established a network planning risk element model, which divides risk elements into discrete model and continuous model to be discussed separately. In the discrete model costs and risk element matrix is introduced to get the corresponding programming model; In Continuous model the idea of machine learning model is used to minimum the desired risk. Based on this model, by using genetic algorithm's efficient and rapid global search capability, this paper improves the genetic algorithm developed by Feng and others, increases the risk elements and eventually gets the cost-time curve considering risk elements. This effectively solves the network planning cost optimization problem.
{"title":"The Research of Network Planning Risk Element Transmission Theory Based on Genetic Algorithm","authors":"Cunbin Li, Kecheng Wang","doi":"10.1109/ICNC.2007.739","DOIUrl":"https://doi.org/10.1109/ICNC.2007.739","url":null,"abstract":"Risk management project is an important aspect of general project risk element transmission theory. Traditional network planning technology encountered great obstacles in project risk management issues, and often unable to accurately forecast the risk, resulting great loss of costs. To address the cost-time optimization problem considering the risk elements, this paper established a network planning risk element model, which divides risk elements into discrete model and continuous model to be discussed separately. In the discrete model costs and risk element matrix is introduced to get the corresponding programming model; In Continuous model the idea of machine learning model is used to minimum the desired risk. Based on this model, by using genetic algorithm's efficient and rapid global search capability, this paper improves the genetic algorithm developed by Feng and others, increases the risk elements and eventually gets the cost-time curve considering risk elements. This effectively solves the network planning cost optimization problem.","PeriodicalId":250881,"journal":{"name":"Third International Conference on Natural Computation (ICNC 2007)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116202903","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：