摘要
提出了一种采用分数阶微分与尺度不变特征变换算法(SIFT)相结合的方式进行图像识别及匹配方法。该方法首先采用分数阶微分方法对图像的细节纹理部分进行加强,从而提高图像的分辨率,然后采用尺度不变特征变换算法对旋转缩放后的图像进行特征关键点提取和匹配,从而提高图像识别的准确率。应用该方法对Lena图像进行图像处理实验,结果表明:采用阶次为0.6的分数阶微分算法与SIFT相结合可最大化地提取图像的关键点和提高图像匹配的准确率(94.59%)。
In this paper,a novel method of image recognition and matching based on fractional order differential and scale invariant feature transform（SIFT）was proposed.In this method,the details of the image were improved by using the fractional order differential method firstly,then the SIFT algorithm was used to extract and match the feature points of the image,and under the above method,the accuracy of image recognition was improved.An experimental study on Lena image was carried out in this paper,the experimental results show that the accuracy of image matching can be improved to 94.59% with the proposed method based on fractional order（0.6）differential and SIFT algorithm.
出处
《半导体光电》
CAS
北大核心
2016年第6期890-893,898,共5页
Semiconductor Optoelectronics
基金
国家自然基金青年基金项目(51409290)
关键词
分数阶微分
SIFT
关键点
图像匹配
fractional order differentiation
SIFT
key points
image matching
