...
首页> 外文期刊>Journal of Computers >An Effective Image Retrieval Method Based on Multi-features
【24h】

An Effective Image Retrieval Method Based on Multi-features

机译:一种基于多特征的有效图像检索方法

获取原文
           

摘要

To improve the accuracy of image retrieval methods,an effective image retrieval method based on multifeaturesis proposed. The color feature of the image isextracted by constructing a (16:4:4) quantization schemeof image color. Perform the marked-watershed to achievethe segmentation of the image, and then extract Jan Flusserinvariant moments. Gaussian model is applied to normalizethe different subcharacters distance. The final multi-featuressimilarity consists of the color similarity and the shapesimilarity. Experiments demonstrate the efficiency of theproposed method.
机译:为了提高图像检索方法的准确性,提出了一种基于多功能特征的有效图像检索方法。通过构造图像颜色的(16:4:4)量化方案来提取图像的颜色特征。执行标记分水岭以实现图像分割,然后提取Jan Flusser不变矩。应用高斯模型对不同的子字符距离进行归一化。最终的多特征相似度由颜色相似度和形状相似度组成。实验证明了该方法的有效性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号