• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2006, Vol. 28 ›› Issue (7): 60-61.

• 论文 • 上一篇    下一篇

基于图像边缘小波矩和支持向量机的目标识别

梅雪 林锦国   

  • 出版日期:2006-07-01 发布日期:2010-05-20

  • Online:2006-07-01 Published:2010-05-20

摘要:

小波矩结合了矩特征和小波特征,既反映了图像的全局性信息,又反映了图像的局域性信息,并且具有旋转、平移和缩放不变性.利用小波矩与支持向量机进行目标识别,不但解 决了图像识别中特征量随图像旋转、平移和缩放而变化的问题,而且提高了对近似物体的识别能力,是解决小样本、近似图像识别的有效方法.

关键词: 小波矩 目标识别 支持向量机

Abstract:

The wavelet moment which combines the moment features with the wavelet features, reflects both the global and local information simultaneously and is   invariant to translation, scaling and rotation. Support Vector Machine and wavelet moments are used to identify the targets. This is an effective method for image recognition with similar features, which improves the capability of distinguishing similar images, and resolves the problems for images to va ry with translation, scaling and rotation.

Key words: wavelet moment, target recognition, support vector machine