{"title":"Person/vehicle classification based on deep belief networks","authors":"Ning Sun, G. Han, K. Du, Jixin Liu, Xiaofei Li","doi":"10.1109/ICNC.2014.6975819","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975819","url":null,"abstract":"In this paper, we investigated the deep learning model for object classification. Robust classification networks were trained based on Deep Belief Networks (DBN) combined with several object representations included image pixel value, feature histogram by Histogram of Oriented Gradients (HOG) operator and eigen-features to distinguish four categories: pedestrian, biker, vehicle and others in the real scene. In addition, an image dataset called NUPTERC, in which the sample images collected from real surveillance video and Internet, was built to test the proposed methods. Experiments based on NUPTERC dataset demonstrated that the proposed deep learning architecture could achieve superior person vehicle classification performance under illumination changes, large pose variations and different resolution.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127529097","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}
{"title":"Discriminating gender on Chinese microblog: A study of online behaviour, writing style and preferred vocabulary","authors":"Li Li, Maosong Sun, Zhiyuan Liu","doi":"10.1109/ICNC.2014.6975942","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975942","url":null,"abstract":"As user attributes are useful for applications such as personalized recommendation, adverting and so on, user attribute predication on Twitter has attracted intensive attentions in recent years. Although Chinese micro-blogging services are different from Twitter on various aspects such as language, user behaviours and so on, few efforts have been made on Chinese micro-blogging services. In this paper, we propose a gender prediction model for Chinese microblog which exploits features including online behaviour, writing style, and preferred vocabulary. Experimental results on Sina Weibo, which is one of the most popular micro-blogging services in China, show that our model achieves the state-of-the-art accuracy 94.3%. We also find significant distinctions between male and female microblog users on online behaviour, writing style and preferred vocabulary, which would be helpful for improving personalized applications.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125431376","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}
Xuyang Wang, Ta Li, Yeming Xiao, Jielin Pan, Yonghong Yan
{"title":"Improved mandarin spoken term detection by using deep neural network for keyword verification","authors":"Xuyang Wang, Ta Li, Yeming Xiao, Jielin Pan, Yonghong Yan","doi":"10.1109/ICNC.2014.6975825","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975825","url":null,"abstract":"In this paper, we propose to use Deep Neural Network (DNN), which has been proved to be the state-of-the-art technique in speech recognition, to re-estimate the confidence of keyword hypotheses in the verification stage of spoken term detection. The speech recognition system based on DNN outperforms that based on conventional Gaussian Mixture Model (GMM) but suffers from the increased decoding time. When the speed of decoding or indexing is critical, it seems to be a trade-off between the performance and the speed to utilize DNN in keyword verification. Inspired by the utilization and acceleration of DNN in the decoding stage, we explored an efficient method to replace GMM by DNN in the verification stage. 5% relative reduction of equal error rate (EER) is achieved and the improvement of recall in the high precision region is especially significant, which is essential to practical tasks. Meanwhile, the search time decreases more than 50% compared to the time derived from the verification on DNN without any refinements.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126906910","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}
{"title":"Mobility model based handover algorithm in LTE-Advanced","authors":"Jie Xu, Yanan Zhao, Xiaorong Zhu","doi":"10.1109/ICNC.2014.6975840","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975840","url":null,"abstract":"In this paper, we propose a new adaptive pre-handover algorithm according to the characteristics of heterogeneous networks and users' demands for seamless connection. Firstly, the movement of user is predicted based on Gauss-Markov mobility model. Then, the handover triggering time is decided by the value of reference signal receiving quality (RSRQ) in the predicted position. Finally, a reasonable pre-handover decision based on users' movement speed and service type in using is made to reduce unnecessary handovers (UHO) and lower handover failure (HOF) rate. Simulation results show that compared with traditional received signal strength (RSS) based handover algorithm, the proposed algorithm makes handover decision more accurate and reasonable. It is successful in decreasing HOF rate and improving system throughput.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125766965","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}
{"title":"An improved shuffled frog leaping algorithm for robot path planning","authors":"J. Ni, Xiahong Yin, Junfeng Chen, Xinyun Li","doi":"10.1109/ICNC.2014.6975893","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975893","url":null,"abstract":"Path planning is one of the most significant and challenging subjects in robot control field. In this paper, a path planning method based on an improved shuffled frog leaping algorithm is proposed. In the proposed approach, a novel updating mechanism based on the median strategy is used to avoid local optimal solution problem in the general shuffled frog leaping algorithm. Furthermore, the fitness function is modified to make the path generated by the shuffled frog leaping algorithm smoother. In each iteration, the globally best frog is obtained and its position is used to lead the movement of the robot. Finally, some simulation experiments are carried out. The experimental results show the feasibility and effectiveness of the proposed algorithm in path planning for mobile robots.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"213 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123690219","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}
{"title":"Cellular automata model for bacterial information sharing mechanism","authors":"Jixin Liu, Xiaofei Li, G. Han, Ning Sun, K. Du","doi":"10.1109/ICNC.2014.6975861","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975861","url":null,"abstract":"Although a bacterium looks like a simple structure microbe, their family is one of the oldest organisms and ubiquitous in every habitat on Earth. The main reason is the bacterial (or prokaryotes') global gene information sharing mechanism termed as bacterial gene sharing (BGS). Based on the research of BGS and Schrodinger's life entropy theory, we presents a novel artificial life system with cellular automata, BGS-CA model, to simulate the bacterial gene information sharing. Some experiments are used to verify the viability and effectiveness of this model.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121660435","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}
{"title":"ANN-based diagnosis of boiler four-tube leakage faults under different loads and operating modes","authors":"Liangyu Ma, Ting Liu, Lei Cheng, Ningshu Wang","doi":"10.1109/ICNC.2014.6975815","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975815","url":null,"abstract":"Four-tube leakage faults are among the most common faults in a large-scale power plant boiler unit, which may result in abnormal boiler shutdown, economic loss and even endanger the safety of operating personnel. Therefore, It is of great significance to grasp the rules of four-tube leakage faults and to recognize the fault type and location in real time with advanced fault diagnosis approach. With the help of a full-scope simulator, detailed fault simulation tests are carried out for the four-tube leakage faults of a 600MW supercritical boiler unit under different coordinated control modes. An intelligent fault diagnosis method, which combines artificial neural network (ANN) with symptom zoom technology, is applied to realize online fault diagnosis of four-tube leakage faults of varied severity at multiple load points and different operating modes. Fault diagnosis simulation tests show that this method can recognize the four-tube leakage faults correctly with certain engineering practicability.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131946077","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}
{"title":"A monotone interpolation method for the interpolation of ore size cumulative distribution in mineral processing industry","authors":"Zheng Wang, Huangang Wang, Wenli Xu, Junwu Zhou","doi":"10.1109/ICNC.2014.6975948","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975948","url":null,"abstract":"Interpolation is a mathematical method to obtain new data points within the range of several known data points. The basic idea of interpolation is to construct a continuous function which passes through all the known data points so that new data points can be obtained from this function. In many practical problems, the interpolation function should be monotone due to physical meanings. However, nearly all of the classic interpolation methods, such as cubic spline interpolation, polynomial interpolation, etc. cannot ensure monotonicity. These methods therefore cannot be used to solve monotone interpolation problems. For this reason, monotone interpolation problems are presented separately from normal interpolation problems and corresponding monotone interpolation methods are required. In this paper, an interpolation framework named generalized mean Hermite interpolation framework which contains several renowned interpolation methods is developed and a monotone interpolation method based on this framework is proposed. It can be proved that among all the methods within the framework, the proposed method is the best one. The method is proposed for the interpolation problem of ore size cumulative distribution which is concerned in the simulation of mineral processing industry. It is an important problem because poor interpolation may lead to inaccurate simulation. However, despite of its importance, it is poorly done in nearly all of the existing mineral processing simulation softwares. Even JKSimMet, the most famous and widely used mineral processing simulation software in the world, cannot do a satisfactory job at this problem. Considering this, the proposed method is applied. Experiments show that its performance is much better than JKSimMet.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121100956","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}
{"title":"A problem solving strategy based on a case study from SARS epidemic","authors":"Simon Wu, H. Wee","doi":"10.1109/ICNC.2014.6975829","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975829","url":null,"abstract":"In the first half of the year 2003, the whole world was seriously panicked by SARS, the Severe Acute Respiratory Syndromes. Taiwan was also heavily stricken. While the public in Taiwan was run over by the fear of this unprecedented disaster, the key factors contributing to the ineffective control of the epidemic was identified to be a series of fault decisions made by the panic-stricken official. Since March 31, 2013, a cumulative total of 400 H7N9 influenza infections, including 125 deaths, have been confirmed in China and Hong Kong. Thus far, a cumulative total of 601 suspected H7N9 cases and 156 deaths worldwide have been reported to Taiwan CDC (Centers for Disease Control). The infection in Taiwan has been confirmed in four imported cases, including one death. In order to provide a problem solving strategy for well managing this potential crisis by government, this study tries to use the concept of Case Based Reasoning (CBR) to apply a case study of SARS epidemic. One important question is often asked: what would have happened if the disease control officer had competent problem solving skills and had timely reacted in line with this epidemic development? It is our task to simulate the problem solving process in the SARS war by deploying the Theory of Constraints (TOC) Problem Solving Model. We also discuss and demonstrate some of findings about TOC application in problem solving.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129987364","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}
{"title":"A novel face recognition method based on the local color vector binary patterns of features localization","authors":"Qiangqiang Song, Liquan Zhang","doi":"10.1109/ICNC.2014.6975954","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975954","url":null,"abstract":"LCVBP (Local Color Vector Binary Patterns) approach extracts multi-signal channel characteristics from color norm patterns and color angular patterns of a color image. As a result, feature dimension is higher and computational cost is greater. Hence, this paper presents a novel region-based LCVBP feature extraction method for face recognition. Firstly, we locate the feature points in a face image, such as eyes, nose and mouth, and obtain feature region by utilizing the location of feature points. Secondly, the LCVBP histograms of these feature regions are extracted, and sequentially put together as the final histogram characteristics of an image. Experimental results show that by abandoning this redundant information in a face image, we can also obtain the approximately equal identification rate with the LCVBP approach, but the dimension of characteristic vector is reduced greatly, the calculation cost is reduced significantly, and face recognition can be achieved faster.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130721230","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}