矿物信息检索实验

D. Cakmakov, D. Davcev
{"title":"矿物信息检索实验","authors":"D. Cakmakov, D. Davcev","doi":"10.1145/166266.166272","DOIUrl":null,"url":null,"abstract":"The problems in image retrieval derive from the difficulty to exactly define and interpret the image content. In most cases, image retrieval techniques are based on the data base system techniques [6], [19] or the information retrieval systems, where the image content is represented in a text form [20]. Other image retrieval techniques require that the images belong to a specific domain which must be described in advance [25]. In this paper, we present the experiments in retrieval of multimedia mineral information using AMCIRS (A Multimedia Cognitive-based Information Retrieval System) [13]. The AMCIRS query based mechanism is based on a multimedia objects content search using the vector model. Each vector is composed of text and image objects. The image objects in the vectors are image object contours, represented by polygonal approximations. The content search process is deduced to the similarity estimation between the MM query and MM index vectors. The similarity function for image objects is based on the polygon similarity estimation. The basic elements of A Multimedia Cognitive-based Information Retrieval System called AMCIRS which integrates image and text information have been described elsewhere [12], [13]. In AMCIRS, the content search process is performed using the vector model [26], where the user query and the MM information are presented by the MM query and index vectors respectively. Each vector contains text and image objects. The image objects in the vectors are image object contours, represented by polygonal approximations [24]. The information selection in AMCIRS is based on the similarity estimation between the MM query and MM index vectors. The similarity function for the image objects is deduced to the polygon similarity estimation [14]. The experimental evaluation of AMCIRS retrieval effectiveness is expressed by the recall and precision parameters. Possible advantages of multiple media retrieval with respect to the single medium retrieval are also investigated and explicitly represented by the recall-precision diagrams.","PeriodicalId":412458,"journal":{"name":"MULTIMEDIA '93","volume":"28 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1993-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"Experiments in retrieval of mineral information\",\"authors\":\"D. Cakmakov, D. Davcev\",\"doi\":\"10.1145/166266.166272\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The problems in image retrieval derive from the difficulty to exactly define and interpret the image content. In most cases, image retrieval techniques are based on the data base system techniques [6], [19] or the information retrieval systems, where the image content is represented in a text form [20]. Other image retrieval techniques require that the images belong to a specific domain which must be described in advance [25]. In this paper, we present the experiments in retrieval of multimedia mineral information using AMCIRS (A Multimedia Cognitive-based Information Retrieval System) [13]. The AMCIRS query based mechanism is based on a multimedia objects content search using the vector model. Each vector is composed of text and image objects. The image objects in the vectors are image object contours, represented by polygonal approximations. The content search process is deduced to the similarity estimation between the MM query and MM index vectors. The similarity function for image objects is based on the polygon similarity estimation. The basic elements of A Multimedia Cognitive-based Information Retrieval System called AMCIRS which integrates image and text information have been described elsewhere [12], [13]. In AMCIRS, the content search process is performed using the vector model [26], where the user query and the MM information are presented by the MM query and index vectors respectively. Each vector contains text and image objects. The image objects in the vectors are image object contours, represented by polygonal approximations [24]. The information selection in AMCIRS is based on the similarity estimation between the MM query and MM index vectors. The similarity function for the image objects is deduced to the polygon similarity estimation [14]. The experimental evaluation of AMCIRS retrieval effectiveness is expressed by the recall and precision parameters. Possible advantages of multiple media retrieval with respect to the single medium retrieval are also investigated and explicitly represented by the recall-precision diagrams.\",\"PeriodicalId\":412458,\"journal\":{\"name\":\"MULTIMEDIA '93\",\"volume\":\"28 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1993-09-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"MULTIMEDIA '93\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/166266.166272\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"MULTIMEDIA '93","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/166266.166272","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 5

摘要

图像检索中存在的问题是难以准确地定义和解释图像内容。在大多数情况下,图像检索技术是基于数据库系统技术[6],[19]或信息检索系统,其中图像内容以文本形式表示[20]。其他图像检索技术要求图像属于一个特定的领域,必须提前描述[25]。在本文中,我们提出了使用基于多媒体认知的信息检索系统(AMCIRS)检索多媒体矿物信息的实验[13]。基于AMCIRS查询的机制是基于矢量模型的多媒体对象内容搜索。每个向量由文本和图像对象组成。矢量中的图像对象是图像对象的轮廓,用多边形近似表示。将内容搜索过程推导为MM查询与MM索引向量之间的相似度估计。图像对象的相似度函数是基于多边形相似度估计的。集成了图像和文本信息的基于多媒体认知的信息检索系统(称为AMCIRS)的基本要素已在其他地方描述[12],[13]。在AMCIRS中,内容搜索过程使用向量模型进行[26],其中用户查询和MM信息分别由MM查询和索引向量表示。每个向量包含文本和图像对象。矢量中的图像对象是图像对象轮廓,用多边形近似表示[24]。AMCIRS中的信息选择基于MM查询与MM索引向量之间的相似度估计。将图像对象的相似度函数推导为多边形相似度估计[14]。AMCIRS检索效果的实验评价用查全率和查准率参数来表示。多媒体检索相对于单媒体检索可能具有的优势也被调查,并通过召回精度图明确地表示出来。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Experiments in retrieval of mineral information
The problems in image retrieval derive from the difficulty to exactly define and interpret the image content. In most cases, image retrieval techniques are based on the data base system techniques [6], [19] or the information retrieval systems, where the image content is represented in a text form [20]. Other image retrieval techniques require that the images belong to a specific domain which must be described in advance [25]. In this paper, we present the experiments in retrieval of multimedia mineral information using AMCIRS (A Multimedia Cognitive-based Information Retrieval System) [13]. The AMCIRS query based mechanism is based on a multimedia objects content search using the vector model. Each vector is composed of text and image objects. The image objects in the vectors are image object contours, represented by polygonal approximations. The content search process is deduced to the similarity estimation between the MM query and MM index vectors. The similarity function for image objects is based on the polygon similarity estimation. The basic elements of A Multimedia Cognitive-based Information Retrieval System called AMCIRS which integrates image and text information have been described elsewhere [12], [13]. In AMCIRS, the content search process is performed using the vector model [26], where the user query and the MM information are presented by the MM query and index vectors respectively. Each vector contains text and image objects. The image objects in the vectors are image object contours, represented by polygonal approximations [24]. The information selection in AMCIRS is based on the similarity estimation between the MM query and MM index vectors. The similarity function for the image objects is deduced to the polygon similarity estimation [14]. The experimental evaluation of AMCIRS retrieval effectiveness is expressed by the recall and precision parameters. Possible advantages of multiple media retrieval with respect to the single medium retrieval are also investigated and explicitly represented by the recall-precision diagrams.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Specification of multimedia composition and a visual programming environment News on-demand for multimedia networks A synchronization and communication model for distributed multimedia objects Multimedia CD-Rom conference publishing (panel session) Where were we: making and using near-synchronous, pre-narrative video
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1