计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (16): 156-158.

• 图形、图像、模式识别 • 上一篇    下一篇

利用SIFT算子与图像插值实现图像匹配

卜凡艳1,檀结庆2   

  1. 1.合肥工业大学 计算机与信息学院,合肥 230009
    2.合肥工业大学 应用数学研究所,合肥 230009
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-06-01 发布日期:2011-06-01

Image matching approach using SIFT descriptor and image interpolation

BU Fanyan1,TAN Jieqing2   

  1. 1.College of Computer and Information,Hefei University of Technology,Hefei 230009,China
    2.Institute of Applied Mathematics,Hefei University of Technology,Hefei 230009,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-06-01 Published:2011-06-01

摘要: 尺度不变特征算子SIFT具有良好的尺度、旋转、光照不变特性,广泛应用于图像匹配领域。该算子利用128维的特征描述向量来表征每个特征点,由于维数过高,影响图像匹配的速度。为此,提出了利用图像插值技术对特征描述向量进行降维处理,提高匹配速度,并通过实验验证了该方法的有效性。

关键词: 尺度空间, 特征提取, 插值方法, 图像匹配

Abstract: SIFT(Scale Invariant Feature Transform) is one of the most effective local features of scale,rotation and illumination invariant,which is widely used in the field of image matching.SIFT indicates every feature point using feature vectors with 128 dimension.But high dimension affects the speed of image matching.For the above mentioned issues,a new method using image interpolation techniques is proposed to reduce the dimensions of feature vectors.The experimental results verify the validity of the method.

Key words: scale space, feature extraction, interpolation method, image matching