An image matching algorithm based on difference measure and improved SIFT algorithm

作者：Gao Qiang; Yang Hongye^*; Yang Wu

来源：Journal of Information and Computational Science, 2014, 11(10): 3631-3642.

DOI：10.12733/jics20104641

• Difference
• Feature matching
• Image
• Membership
• SIFT

In this paper, detailed process of Scale-invariant Feature Transformation (SIFT) algorithm is analyzed. For its weaknesses, the difference measure theory is introduced, and then a new improved method is proposed. At first, the gray value matrix of image is converted to the difference value matrix based on the knowledge of fuzzy membership. The non-linear relationship between human eye sensitivity and differences of image gray value changes into a linear relationship. And then SIFT feature points are extracted and matched based on the difference value of pixel. The experimental results based on standard image datasets demonstrate that, this method can reduce the complexity of the algorithm and improve the matching rate at the same time; the efficiency of the algorithm is also improved.

• 出版日期2014
• 单位华北电力大学