.png){kind=icon}
返回
Research on image stitching method based on fuzzy inference
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-022-12748-9.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
This paper concern the problem of ghosting caused by parallax and moving objects in image stitching. Previous approaches have used local homography or optimal seam lines to avoid ghosting. In this work, we propose an image stitching method based on fuzzy inference. At first, our use of the contrast limited adaptive histogram equalization (CLAHE) increase the matching points of the object surface in the low-contrast images. Then, to reduce the number of mismatching points, we combine the orientation of feature points to improve the zero-mean normalized cross-correlation (ZNCC) for filter matching points. Furthermore, by viewing the distance weight and gray difference of pixels in overlapping region as the first input and the second input of fuzzy inference respectively, and regarding the output of the fuzzy inference as the weight of image fusion. We generate high-quality stitching images. The experimental results show that our approach can reduce the ghosting phenomenon and improve the quality of the stitching.
Keyword:
Image stitching
Fuzzy inference
ZNCC
Eliminate ghosting
期刊
IF:
3
论文数:
2.0W
被引数:
3.2W
机构
引用论文
.png){kind=icon}
Image mosaicking using improved auto-sorting algorithm and local difference-based harris features
PREAI
分享收藏Variation of the electron energy distribution with He dilution in an inductively coupled argon discharge在感应耦合氩气放电中，电子能量分布随氦气稀释的变化
OAAI
OAAI分享收藏
查看更多内容