| 摘要: |
| 针对SURF（Speeded Up Robust Features）算法在图像拼接过程中配准精度低和运算复杂度较高的问题，提出了一种基于SURF分割圆形区域的图像拼接方法。首先，对图像进行特征点的检测和提取过程中，在传统SURF算法的基础上采用了分割圆形区域提取描述符；接着，计算了区域归一化的灰度均值和韦伯局部描述细节信息，形成新的描述符；然后，通过改进的RANSAC（Random Sample Consensus）算法对每个模块的误匹配点进行消除，得到精确的匹配特征点。实验结果表明，相较基于改进的SURF算法的图像拼接，该方法匹配正确率平均提高了6.05%，拼接时间平均减少了24.47%，匹配速度和效果有了较大提升，基本满足图像拼接速度快、精准度高、稳定可靠等要求。 |
| 关键词: 图像拼接 特征检测 图像配准 韦伯描述符 SURF算法 |
| DOI： |
|
| 基金项目:公安部科技强警基础工作专项项目（2018GABJC39） |
|
| Image stitching based on SURF segmentation of circular regions |
| HONG Pengcheng,TANG Yao,REN Zhong |
| (School of Communication and Information Engineering,Xian University of Posts & Telecommunications,Xian 710121,China) |
| Abstract: |
| In order to solve the problem of low stitching accuracy and high computational complexity of Speeded Up Robust Features(SURF) algorithm in the process of image stitching,a SURF-based image stitching method for segmenting circular regions is proposed.First，in the process of detecting and extracting the characteristic points of the image,a modular circular area extraction descriptor is adopted on the basis of the traditional SURF algorithm.The gray mean value of the area normalization and the Weber local description detail information are calculated to from a new descriptor.Then,the false matching points of each module are eliminated by the improved Random Sample Consensus(RANSAC) algorithm to obtain the accurate matching feature points.The experimental results show that,compared with that of the improved SURF algorithm,the matching accuracy of the proposed method has been improved by 6.05%,the matching time is reduced by 24.47%,matching speed and effect is greatly improved.It can satisfy the requirements of fast,high precision stable and reliable image stitching. |
| Key words: image mosaic feature detection image registration Webber narrator SURF algorithm |
