基于颜色的图像检索在艺术教育中的应用
发布时间：2019-06-26 21:28
【摘要】：当今，素质教育让艺术教育在教育中的地位越来越重要，艺术教育尤其是美术教育，需要储备丰富的图像资源才能让学生的思维空间变得更为宽阔。计算机技术和多媒体技术高度发展的今天，互联网为我们提供了丰富的资源，，我们自己搜集的资源也可以放在校园网上与人共享。网络已经成为人们获取信息的重要途径，网上信息的检索变得尤为重要。 目前图像检索方法主要有两类。基于关键词的图像检索和基于内容的图像检索。基于内容的图像检索(CBIR)是当前图像检索技术的研究热点之一，它通过对图像的内容特征(如图像的颜色、形状、纹理等)进行分析和提取，建立图像特征索引库，然后根据图像内容特征的相似性检索图像，而使检索结果在视觉特征上具有更好的一致性。在基于内容的图像检索中，颜色是最直观、最明显的视觉特征。基于颜色的图像检索通过提取图像的颜色特征，利用颜色特征间的相似度实现图像的检索。 颜色特征的提取，首先需要选取合理的颜色模型表示颜色。RGB模型较常用，但不符合人们对颜色相似性的主观判断，故在图像颜色分析中常采用HSV模型。然后，在对颜色空间进行合理的量化后，利用颜色直方图法分析图像的颜色特征。常用的颜色特征分析方法有统计直方图法，累加直方图法，直方图相交法，比例直方图法，距离法，参考颜色表法，聚类算法和HSI中心矩法等。通过颜色聚类提取出图像的主色调，存入图像的颜色特征索引库。 论文以几个著名的基于内容的图像检索实验系统为例，阐述了基于颜色特征的图像检索方法，并探讨了基于颜色的图像检索在美术教学中应用的问题。通过教学实例，体会到图像检索技术给美术教学带来的积极作用，对培养我们的美术学生在绘画方面的颜色选取能力、颜色的合理搭配能力将会起到积极的效果，更能体现信息技术教育与艺术教育整合的优越性，有利于促进艺术教育的发展。
[Abstract]:Nowadays, quality education makes art education more and more important in education. Art education, especially art education, needs to reserve rich image resources in order to make students' thinking space wider. With the high development of computer technology and multimedia technology, the Internet provides us with rich resources, and the resources we collect can also be shared with people on the campus network. The network has become an important way for people to obtain information, and the retrieval of online information has become particularly important. At present, there are two main image retrieval methods. Keyword-based image retrieval and content-based image retrieval. Content-based image retrieval (CBIR) is one of the research hotspots in current image retrieval technology. by analyzing and extracting the content features (such as image color, shape, texture, etc.), the image feature index library is established, and then the image is searched according to the similarity of image content features, so that the retrieval results are more consistent in visual features. In content-based image retrieval, color is the most intuitive and obvious visual feature. The color-based image retrieval realizes the image retrieval by extracting the color features of the image and using the similarity between the color features. In order to extract color features, it is necessary to select a reasonable color model to represent color. RGB model is more commonly used, but it does not accord with people's subjective judgment of color similarity, so HSV model is often used in image color analysis. Then, after the color space is reasonably quantified, the color features of the image are analyzed by color histogram method. The commonly used color feature analysis methods are statistical histogram method, cumulative histogram method, histogram intersection method, proportional histogram method, distance method, reference color table method, clustering algorithm and HSI central moment method. The main tone of the image is extracted by color clustering and stored in the color feature index library of the image. Taking several famous content-based image retrieval experimental systems as examples, this paper expounds the image retrieval method based on color features, and discusses the application of color-based image retrieval in art teaching. Through teaching examples, we can realize the positive effect of image retrieval technology on art teaching, which will play a positive effect on cultivating the color selection ability and color reasonable collocation ability of our art students in painting, and can better reflect the advantages of the integration of information technology education and art education, and is conducive to promoting the development of art education.
【学位授予单位】：山东师范大学
【学位级别】：硕士
【学位授予年份】：2005
【分类号】：G354;J0-4
本文编号：2506507
[Abstract]:Nowadays, quality education makes art education more and more important in education. Art education, especially art education, needs to reserve rich image resources in order to make students' thinking space wider. With the high development of computer technology and multimedia technology, the Internet provides us with rich resources, and the resources we collect can also be shared with people on the campus network. The network has become an important way for people to obtain information, and the retrieval of online information has become particularly important. At present, there are two main image retrieval methods. Keyword-based image retrieval and content-based image retrieval. Content-based image retrieval (CBIR) is one of the research hotspots in current image retrieval technology. by analyzing and extracting the content features (such as image color, shape, texture, etc.), the image feature index library is established, and then the image is searched according to the similarity of image content features, so that the retrieval results are more consistent in visual features. In content-based image retrieval, color is the most intuitive and obvious visual feature. The color-based image retrieval realizes the image retrieval by extracting the color features of the image and using the similarity between the color features. In order to extract color features, it is necessary to select a reasonable color model to represent color. RGB model is more commonly used, but it does not accord with people's subjective judgment of color similarity, so HSV model is often used in image color analysis. Then, after the color space is reasonably quantified, the color features of the image are analyzed by color histogram method. The commonly used color feature analysis methods are statistical histogram method, cumulative histogram method, histogram intersection method, proportional histogram method, distance method, reference color table method, clustering algorithm and HSI central moment method. The main tone of the image is extracted by color clustering and stored in the color feature index library of the image. Taking several famous content-based image retrieval experimental systems as examples, this paper expounds the image retrieval method based on color features, and discusses the application of color-based image retrieval in art teaching. Through teaching examples, we can realize the positive effect of image retrieval technology on art teaching, which will play a positive effect on cultivating the color selection ability and color reasonable collocation ability of our art students in painting, and can better reflect the advantages of the integration of information technology education and art education, and is conducive to promoting the development of art education.
【学位授予单位】：山东师范大学
【学位级别】：硕士
【学位授予年份】：2005
【分类号】：G354;J0-4
【参考文献】
相关期刊论文 前6条
1 王惠锋 ,孙正兴 ,王箭;语义图像检索研究进展[J];计算机研究与发展;2002年05期
2 曹莉华,柳伟,李国辉;基于多种主色调的图像检索算法研究与实现[J];计算机研究与发展;1999年01期
3 李岚,冯刚;MPEG——7与基于内容的图像检索[J];计算机工程与应用;2002年17期
4 卢汉清,孔维新,廖明,马颂德;基于内容的视频信号与图像库检索中的图像技术[J];自动化学报;2001年01期
5 陈立娜;因特网上的图像搜索引擎[J];情报理论与实践;2001年04期
6 傅华胜,周洞汝;Internet上的图像检索技术[J];计算机工程与设计;2003年11期
相关硕士学位论文 前1条
1 徐曼;基于内容的图像检索技术的研究与系统实现[D];南京理工大学;2002年
本文编号：2506507
本文链接：https://www.wllwen.com/wenyilunwen/yishull/2506507.html
