{"title":"Multilayer Architecture Based on HMM and SVM for Fault Classification","authors":"Yujun Pang, Zhentao Ma, Yuan Li","doi":"10.1109/ICICIC.2009.273","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.273","url":null,"abstract":"In order to solve the problems of current machine learning in fault diagnosing system of the chemical plants, a better and effective multilayer architecture model is used in this paper. Hidden Markov Model (HMM) is good at dealing with dynamic continuous data and Support Vector Machine (SVM) shows superior performance for classification, especially for limited samples. Combining their respective virtues, we propose a new multilayer architecture model to improve classification accuracy for a fault diagnosis example. The simulation result shows that this two level architecture framework combining HMM and SVM is better than the single HMM method in high classification accuracy with small training samples.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114201647","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":"Human Support Interface for Operation of Industrial Crane Based on RNN","authors":"M. Kimura, M. Konishi","doi":"10.1109/ICICIC.2009.232","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.232","url":null,"abstract":"A mobile crane is widely used not only to daily work but also to carry heavy materials efficiently in a construction site after the time of disaster. However, the decreases in the number of expert persons induce crane accident. As it is well known, the crane operation is highly complicated even for experts. It is required to realize a partial automation of crane operation and human support technology in the near future. In this paper, a human interface system is proposed intended to human support of a mobile crane operation. We developed the crane simulator which is composed of the hydraulic circuit model and the model for a crane dynamics. Based on the simulator, the human interface system for the operation support based on recurrent neural network (RNN) is proposed.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"223 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114986502","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":"Camera Position Estimation from Image by ANFIS","authors":"K. Hsia, S. Lien, C. C. Wang, J. Su","doi":"10.1109/ICICIC.2009.132","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.132","url":null,"abstract":"For picture taking, 3D scenery is projected to a 2D image through the camera center. Different position of the camera will cause a different image. It is hard to estimate the camera position only from an image. In this paper, a method to estimate the position of the camera by extracting features from a 2D image and using adaptive neuro-fuzzy inference systems (ANFIS) is proposed. From the experimental results, it is evidently that the proposed method can estimate the center of the camera correctly in a short time.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115516974","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 Multilevel Thresholding Method Based on Cross Entropy and Genetic Algorithms","authors":"Shu-Chien Huang","doi":"10.1109/ICICIC.2009.29","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.29","url":null,"abstract":"Threshold selection is one of the most important issues in image processing. In this paper, a general technique for multilevel thresholding based on cross entropy is proposed. Then, a genetic algorithm is designed especially for searching for the near-optimal or optimal thresholds. The effectiveness and efficiency of the proposed method is demonstrated by using well-known images.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123198809","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":"GA Approach for Designing Fuzzy Model with Wavelet Transforms","authors":"Shian-Tang Tzeng","doi":"10.1109/ICICIC.2009.223","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.223","url":null,"abstract":"In this paper, an efficient method is proposed to design fuzzy model with wavelet transforms for function learning. The structure is based on the basis of fuzzy rules including wavelet functions in the consequent parts of rules. In order to improve the function approximation accuracy and general capability of the system, an efficient genetic algorithm (GA) approach is used to adjust the parameters of dilation, translation, weights, and membership functions. By minimizing a quadratic measure of the error derived from the output of the system, the design problem can be characterized by the proposed GA formulation. The performance of our approximation is superior to that of the existing methods. Also, one numerical design example is presented to demonstrate the design flexibility and usefulness of this presented approach.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122978344","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 Hybrid Particle Swarm Optimization Algorithm with Diversity for Flow-Shop Scheduling Problem","authors":"Shin-Ying Huang, Chuen-Lung Chen","doi":"10.1109/ICICIC.2009.21","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.21","url":null,"abstract":"This paper proposed a hybrid particle swarm optimization algorithm (Shadow hybrid PSO, SHPSO) to solve the flow-shop scheduling problem (FSSP). SHPSO adopts the idea of Kuoa's HPSO model[4] by not only combines the random-key (RK) encoding scheme, individual enhancement (IE) scheme, but also adds the diversification mechanism such as ARPSO model and competitive shadow particles to prevent premature convergence. Computation experiments results of Taillard's [10] seven representative instances of FSSP show that the SHPSO perform close to HPSO for FSSP to minimize makespan. Further recommendations and improved ideas will be discussed later in this paper.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"112 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122939393","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":"Dynamic Load Balancing Algorithm Based on FCFS","authors":"Wenzheng Li, Hongyan Shi","doi":"10.1109/ICICIC.2009.182","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.182","url":null,"abstract":"In a load balancing cluster, the core of task distribution is the load balance algorithm. This paper briefly discusses load balancing, algorithms and their merits and demerits, then introduces a kind of load balancing algorithm that every node sends a corresponding request stream to remark its real-time load based on FCFS principle. The results of experience and measurement show that this dynamic load balancing algorithm is more effective than static algorithm.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123634596","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":"Case Adaptation in Case-Based Reasoning System for Back-Light Modules","authors":"C. Ho, Lifai Wang","doi":"10.1109/ICICIC.2009.134","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.134","url":null,"abstract":"Back-light module (BLM) is one of the critical components in TFT-LCD display. It provides homogeneous light source for the display. The utilization of case-based reasoning system (CBRS) in the design process greatly improves the productivity of the design engineers. When designing a new BLM, the engineer would retrieve the existing case with best fit to the customer's specifications. The retrieved design case needs to be adapted to the current specifications if no perfect match is found. The objective of this paper is to develop a case adaptation technology to assist product engineers in developing the new product after retrieval of existing case from CBRS. The first phase of the case-adaption technology development is to search for the most important factors affecting the final product performance. These factors are implemented in the attributes of the case-based reasoning system. The light guide plate is identified as the most critical component in the back-light module system in second phase of the development. This paper developed an empirical formula to assist the design engineers to adapt their design after the first design been measured as off the target specification. In most of the cases, the new suggested design parameters, suggested by the system based on the retrieved reference case and first adapted design developed by engineer, would successfully meet the specifications of customer.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121708760","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":"Computer Aided Diagnosis for Traditional Chinese Medicine Based on Information Retrieval Techniques","authors":"Shun Long, Weiheng Zhu","doi":"10.1109/ICICIC.2009.144","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.144","url":null,"abstract":"Traditional expert systems for medical diagnosis have a major pitfall, in that it is difficult to adopt new knowledge, particularly from descriptive, incomplete and unstructured information. Information retrieval sounds a natural answer to these problems. This paper presents an information retrieval based approach to computer aided diagnosis in order to tackle this problem. Preliminary experimental results with traditional Chinese medicine expertise show that it can effectively help to diagnose based on the descriptive symptoms provided.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123895616","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":"Adaptive Control Algorithm for Speed Regulation of Marine Engine","authors":"Xiaoyan Xu, Yinzhong Ye","doi":"10.1109/ICICIC.2009.68","DOIUrl":"https://doi.org/10.1109/ICICIC.2009.68","url":null,"abstract":"Firstly, the mathematical model of marine engine speed regulation system is established based on the analysis of the operation of a marine engine. Then, with the consideration of low efficiency of marine engine speed regulation process due to the complexity of its operational condition, both PI control algorithm and adaptive control algorithm are designed for improvement of the speed regulation process. The adaptive control algorithm adopts self-organizing through modification of weighted values come from correlative search, as well as a coefficient of smoothness is adopted to smooth the modification process of weighted values. Finally, simulation experiments of speed regulation process under the application of the abovementioned two different control algorithm are fulfilled and related experimental results are compared. The results show that the designed adaptive control algorithm is with better properties for speed regulation of marine engine.","PeriodicalId":240226,"journal":{"name":"2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123971791","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}