Pub Date : 2013-05-23DOI: 10.1109/SACI.2013.6608948
J. Gáti, G. Kártyás
As a continuous development of former results in teaching of engineering using advanced computer systems, this paper is an attempt to contribute to education of engineers by an analysis and discussion of multi-discipline representation of theory and practice. A new method is introduced, explained and discussed to integrate a formerly published course modeling and a course model evaluation method with the product modeling. The purpose of this integration is the communication of demand for examples supporting given course content and on the other hand, demand for course content needed by the definition of given product objects in the product model. Integration is realized through a new element called as issue distributor. Product modeling which is applied at recent advanced product lifecycle management is considered.
{"title":"Theory and practice in multi-discipline representations for engineering education","authors":"J. Gáti, G. Kártyás","doi":"10.1109/SACI.2013.6608948","DOIUrl":"https://doi.org/10.1109/SACI.2013.6608948","url":null,"abstract":"As a continuous development of former results in teaching of engineering using advanced computer systems, this paper is an attempt to contribute to education of engineers by an analysis and discussion of multi-discipline representation of theory and practice. A new method is introduced, explained and discussed to integrate a formerly published course modeling and a course model evaluation method with the product modeling. The purpose of this integration is the communication of demand for examples supporting given course content and on the other hand, demand for course content needed by the definition of given product objects in the product model. Integration is realized through a new element called as issue distributor. Product modeling which is applied at recent advanced product lifecycle management is considered.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115620520","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 : 2013-05-23DOI: 10.1109/SACI.2013.6609013
H. Charafeddine, V. Groza
This paper describes the utilization of wideband LMS adaptive beamforming with QMF filterbanks decomposition for sonar application. The obtained results showed that applying the new algorithm yields better results than the time domain LMS beamformer (TD). The subband LMS beamformer isolates the interference in its own band and uses that band beamformer to steer a null in its direction, this capability enables the other beamformers to adapt independently without been affected by the interference.
{"title":"Wideband adaptive LMS beamforming using QMF subband decomposition for sonar","authors":"H. Charafeddine, V. Groza","doi":"10.1109/SACI.2013.6609013","DOIUrl":"https://doi.org/10.1109/SACI.2013.6609013","url":null,"abstract":"This paper describes the utilization of wideband LMS adaptive beamforming with QMF filterbanks decomposition for sonar application. The obtained results showed that applying the new algorithm yields better results than the time domain LMS beamformer (TD). The subband LMS beamformer isolates the interference in its own band and uses that band beamformer to steer a null in its direction, this capability enables the other beamformers to adapt independently without been affected by the interference.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115734478","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 : 2013-05-23DOI: 10.1109/SACI.2013.6608959
E. Voisan, Oscar Marginean, R. Precup, Florin Dragan, C. Purcaru
Face detection is an important area of research in robotics and mechatronics applications as an essential subsystem in the interaction with human subjects. As face detection-based applications are moving into mobile and embedded systems, processing performance, among other characteristics, plays a major role. This paper reviews the current status concerning the development of face detection algorithms in order to evaluate the performance evaluation of a set of algorithms. A face detection algorithm based on the Viola-Jones face detector is implemented such that to run on three different operating systems and two hardware platforms in order to represent the basis for the development of a computing environment to run on robotic devices and on mechatronics applications.
{"title":"Performance evaluation of a face detection algorithm running on general purpose operating systems","authors":"E. Voisan, Oscar Marginean, R. Precup, Florin Dragan, C. Purcaru","doi":"10.1109/SACI.2013.6608959","DOIUrl":"https://doi.org/10.1109/SACI.2013.6608959","url":null,"abstract":"Face detection is an important area of research in robotics and mechatronics applications as an essential subsystem in the interaction with human subjects. As face detection-based applications are moving into mobile and embedded systems, processing performance, among other characteristics, plays a major role. This paper reviews the current status concerning the development of face detection algorithms in order to evaluate the performance evaluation of a set of algorithms. A face detection algorithm based on the Viola-Jones face detector is implemented such that to run on three different operating systems and two hardware platforms in order to represent the basis for the development of a computing environment to run on robotic devices and on mechatronics applications.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114738109","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 : 2013-05-23DOI: 10.1109/SACI.2013.6608993
Elisa Valentina Moisi, B. Nagy, V. Cretu
We present an algorithm for reconstruction of binary images represented on the triangular grid. The algorithm takes into consideration the three natural projections on the triangular grid and uses the maximum flow minimum cost algorithm for the reconstruction of the image. For testing our algorithm we used some artificial binary hexagon shaped images.
{"title":"Maximum flow minimum cost algorithm for reconstruction of images represented on the triangular grid","authors":"Elisa Valentina Moisi, B. Nagy, V. Cretu","doi":"10.1109/SACI.2013.6608993","DOIUrl":"https://doi.org/10.1109/SACI.2013.6608993","url":null,"abstract":"We present an algorithm for reconstruction of binary images represented on the triangular grid. The algorithm takes into consideration the three natural projections on the triangular grid and uses the maximum flow minimum cost algorithm for the reconstruction of the image. For testing our algorithm we used some artificial binary hexagon shaped images.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122485616","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 : 2013-05-23DOI: 10.1109/SACI.2013.6609007
D. Andrei, D. Poenaru, D. Nemes, M. Vida, L. Stoicu-Tivadar, N. Gal
The diagnosis and treatment of lumbar spine pathology represents a complex process involving many and diverse parameters that should to be investigated and processed. In order to properly approach the computer assisted treatment and diagnosis this paper presents a model of the process using BPMN and also a UML model for implementation. The data is supplied directly from the keyboard or from the Zebris equipment. The parameters investigated are: demographic data, disability status (4 degrees), daily activity expressed in calories (24 possibilities), Zebris mobility degree (minimum/ maximum-6 values), and Zebris position rate (expressed as an angle). The inference engine of the presented method is created using fuzzy inference system. The data collected from the patients and the Zebris equipment is transformed in linguistic variables and the appropriate fuzzy inference rules are constructed. The consequences of the rules encode the actions that should be taken. Relating the values of the investigated parameters screening values for each measurement can be established. Future work will result in prediction of recovery rate and also developing educational tools related to recovery domain.
{"title":"Process modeling and assisted diagnosis in spinal recovery","authors":"D. Andrei, D. Poenaru, D. Nemes, M. Vida, L. Stoicu-Tivadar, N. Gal","doi":"10.1109/SACI.2013.6609007","DOIUrl":"https://doi.org/10.1109/SACI.2013.6609007","url":null,"abstract":"The diagnosis and treatment of lumbar spine pathology represents a complex process involving many and diverse parameters that should to be investigated and processed. In order to properly approach the computer assisted treatment and diagnosis this paper presents a model of the process using BPMN and also a UML model for implementation. The data is supplied directly from the keyboard or from the Zebris equipment. The parameters investigated are: demographic data, disability status (4 degrees), daily activity expressed in calories (24 possibilities), Zebris mobility degree (minimum/ maximum-6 values), and Zebris position rate (expressed as an angle). The inference engine of the presented method is created using fuzzy inference system. The data collected from the patients and the Zebris equipment is transformed in linguistic variables and the appropriate fuzzy inference rules are constructed. The consequences of the rules encode the actions that should be taken. Relating the values of the investigated parameters screening values for each measurement can be established. Future work will result in prediction of recovery rate and also developing educational tools related to recovery domain.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128264814","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 : 2013-05-23DOI: 10.1109/SACI.2013.6608963
D. Ionescu, V. Suse, C. Gadea, B. Solomon, B. Ionescu, S. Islam
Since the introduction Gesture Control technology in the electronic gaming technology a series of attempts have been made to deploy it also on other domains such as robotics, teaching, medical, automotive and many others. Human gesture used for Man-Machine Interaction became attractive as it offers a simpler way of controlling sophisticated devices, in a sci-fi-like scenario, in return of an increasingly computational power required by the artificial intelligence algorithms needed to detect, track and recognize them. There have been attempts to bring a solution to it by using 2D or 3D based image processing methods. There is a clear balance incline towards 3D methods in the consumer product as besides the almost insurmountable difficulties for producing robust and stable results, the price constraint added supplementary hurdles. As perfect illumination conditions are core factors in obtaining the above results, the infrared light was unanimously adopted by the domain technologies. In this paper, a novel real-time depth-mapping principle and a corresponding hardware solution for an IR depth-mapping camera is introduced. The new IR camera architecture comprises an illuminator module which is pulsed and modulated via a monotonic function using a phase-locked loop control for the laser intensity, while the reflected infrared light is captured during the increasing and decreasing monotonic function. A reconfigurable hardware architecture (RHA) unit calculates the depth and controls the IR waves in synchronism with the infrared sensor. The resolution of the depth map is variable depending on the resolution and gating possibilities of the image sensor. A sensor of 1 megapixel is used, providing a resolution of 1024×1024. Images of real objects are reconstructed in 3D based on the data obtained by the laser controlled by the RHA. A corresponding image processing algorithm builds the 3D map of the object in real-time. In this paper the camera is used to control consumer electronic products such as TV sets, laptops and others.
{"title":"A new NIR camera for gesture control of electronic devices","authors":"D. Ionescu, V. Suse, C. Gadea, B. Solomon, B. Ionescu, S. Islam","doi":"10.1109/SACI.2013.6608963","DOIUrl":"https://doi.org/10.1109/SACI.2013.6608963","url":null,"abstract":"Since the introduction Gesture Control technology in the electronic gaming technology a series of attempts have been made to deploy it also on other domains such as robotics, teaching, medical, automotive and many others. Human gesture used for Man-Machine Interaction became attractive as it offers a simpler way of controlling sophisticated devices, in a sci-fi-like scenario, in return of an increasingly computational power required by the artificial intelligence algorithms needed to detect, track and recognize them. There have been attempts to bring a solution to it by using 2D or 3D based image processing methods. There is a clear balance incline towards 3D methods in the consumer product as besides the almost insurmountable difficulties for producing robust and stable results, the price constraint added supplementary hurdles. As perfect illumination conditions are core factors in obtaining the above results, the infrared light was unanimously adopted by the domain technologies. In this paper, a novel real-time depth-mapping principle and a corresponding hardware solution for an IR depth-mapping camera is introduced. The new IR camera architecture comprises an illuminator module which is pulsed and modulated via a monotonic function using a phase-locked loop control for the laser intensity, while the reflected infrared light is captured during the increasing and decreasing monotonic function. A reconfigurable hardware architecture (RHA) unit calculates the depth and controls the IR waves in synchronism with the infrared sensor. The resolution of the depth map is variable depending on the resolution and gating possibilities of the image sensor. A sensor of 1 megapixel is used, providing a resolution of 1024×1024. Images of real objects are reconstructed in 3D based on the data obtained by the laser controlled by the RHA. A corresponding image processing algorithm builds the 3D map of the object in real-time. In this paper the camera is used to control consumer electronic products such as TV sets, laptops and others.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130560154","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 : 2013-05-23DOI: 10.1109/SACI.2013.6608971
M. Miki, T. Yoshii, Hisanori Ikegami, Y. Azuma, Akihiko Emi, K. Yoshida
We are conducting research and development on an intelligent lighting system that can improve both workers' comfort in an office and reduce electricity consumption of lighting. Upon introducing this system in an actual office, we were able to realize both an improvement in workers' comfort and a drastic reduction in electricity consumption. Furthermore, in office lighting design of recent years, brightness design has become important in addition to illuminance design. In this paper, we propose a new lighting control system that realizes a brightness distribution within a worker's field of vision that they find comfortable. A mechanism that considers brightness distribution in a worker's field of vision has been added to the previous model of an intelligent lighting system that only controlled desk-surface illuminance. Experimental results indicated that target brightness distribution within a worker's field of vision could be realized. It was also shown that, by having a system learn the brightness distribution a worker finds most comfortable, the targeted brightness distribution can be provided even when display brightness is changed during work.
{"title":"A lighting control system to optimize brightness distribution on working places","authors":"M. Miki, T. Yoshii, Hisanori Ikegami, Y. Azuma, Akihiko Emi, K. Yoshida","doi":"10.1109/SACI.2013.6608971","DOIUrl":"https://doi.org/10.1109/SACI.2013.6608971","url":null,"abstract":"We are conducting research and development on an intelligent lighting system that can improve both workers' comfort in an office and reduce electricity consumption of lighting. Upon introducing this system in an actual office, we were able to realize both an improvement in workers' comfort and a drastic reduction in electricity consumption. Furthermore, in office lighting design of recent years, brightness design has become important in addition to illuminance design. In this paper, we propose a new lighting control system that realizes a brightness distribution within a worker's field of vision that they find comfortable. A mechanism that considers brightness distribution in a worker's field of vision has been added to the previous model of an intelligent lighting system that only controlled desk-surface illuminance. Experimental results indicated that target brightness distribution within a worker's field of vision could be realized. It was also shown that, by having a system learn the brightness distribution a worker finds most comfortable, the targeted brightness distribution can be provided even when display brightness is changed during work.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130530087","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 : 2013-05-23DOI: 10.1109/SACI.2013.6609015
I. Simonics
Since 2011, we have been educating mentor teachers at buda University Trefort Ágoston Centre for Engineering Education - ÓOU TÁcEE. Mentor teachers can support the preparation process of our engineering teacher students in secondary vocational schools by coaching their teaching practice. In this article I present examples how have changed my teaching methodology according to the recognized needs and share my experience on teaching-learning process.
{"title":"New wave for learning with mentor teachers","authors":"I. Simonics","doi":"10.1109/SACI.2013.6609015","DOIUrl":"https://doi.org/10.1109/SACI.2013.6609015","url":null,"abstract":"Since 2011, we have been educating mentor teachers at buda University Trefort Ágoston Centre for Engineering Education - ÓOU TÁcEE. Mentor teachers can support the preparation process of our engineering teacher students in secondary vocational schools by coaching their teaching practice. In this article I present examples how have changed my teaching methodology according to the recognized needs and share my experience on teaching-learning process.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129545215","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 : 2013-05-23DOI: 10.1109/SACI.2013.6608983
Gelu Laurentiu Ioanas, T. Dragomir
Hydraulic processes with turbulent flow are usually highly nonlinear and common rail (CR) systems make no exception. Since the performances of diesel CR engines are directly dependent on the rail pressure, and on its values used in control, a prediction model which can lead to better performances is presented. The prediction makes use of Hammerstein dynamic models integrated into a multilevel Neuro-Fuzzy structure. The process input space decomposition is performed axis orthogonal for a large region using Local Linear Model Tree (LOLIMOT) algorithm and the local dynamic models parameters are adapted using recursive last squares method. The practical final results are favorable.
{"title":"Common-rail pressure estimation using a Neuro-Fuzzy architecture with local Hammerstein models","authors":"Gelu Laurentiu Ioanas, T. Dragomir","doi":"10.1109/SACI.2013.6608983","DOIUrl":"https://doi.org/10.1109/SACI.2013.6608983","url":null,"abstract":"Hydraulic processes with turbulent flow are usually highly nonlinear and common rail (CR) systems make no exception. Since the performances of diesel CR engines are directly dependent on the rail pressure, and on its values used in control, a prediction model which can lead to better performances is presented. The prediction makes use of Hammerstein dynamic models integrated into a multilevel Neuro-Fuzzy structure. The process input space decomposition is performed axis orthogonal for a large region using Local Linear Model Tree (LOLIMOT) algorithm and the local dynamic models parameters are adapted using recursive last squares method. The practical final results are favorable.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"80 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133287840","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 : 2013-05-23DOI: 10.1109/SACI.2013.6608966
András London, T. Csendes
Two popular and widely used webpage ranking algorithms are PageRank and HITS. We considered the 2009 Szeged Wine Fest data and another reliable data set of wines from the famous region Villány and, on basis of each data set, constructed a directed and weighted bipartite graph of wine tasters and wines. We applied an extended version of PageRank and HITS, the Co-HITS algorithm to wine tasting graph in order to rank tasters according to their ability and professional skill. The results of our technique were compared to other simple statistical methods. In general we observed that our ranking method performed better: it can filter out incompetent tasters, who, for example, gave the average score of some other tasters for the wines she or he tasted. Furthermore, our method gives a clearer picture about the competence of wine tasters.
{"title":"HITS based network algorithm for evaluating the professional skills of wine tasters","authors":"András London, T. Csendes","doi":"10.1109/SACI.2013.6608966","DOIUrl":"https://doi.org/10.1109/SACI.2013.6608966","url":null,"abstract":"Two popular and widely used webpage ranking algorithms are PageRank and HITS. We considered the 2009 Szeged Wine Fest data and another reliable data set of wines from the famous region Villány and, on basis of each data set, constructed a directed and weighted bipartite graph of wine tasters and wines. We applied an extended version of PageRank and HITS, the Co-HITS algorithm to wine tasting graph in order to rank tasters according to their ability and professional skill. The results of our technique were compared to other simple statistical methods. In general we observed that our ranking method performed better: it can filter out incompetent tasters, who, for example, gave the average score of some other tasters for the wines she or he tasted. Furthermore, our method gives a clearer picture about the competence of wine tasters.","PeriodicalId":304729,"journal":{"name":"2013 IEEE 8th International Symposium on Applied Computational Intelligence and Informatics (SACI)","volume":"172 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114623632","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：