Image stitching based on human visual system and SIFT algorithm

The image stitching process often produces many undesirable effects. Solving problems such as discontinuity and dislocation of pictures has always been the focus of people’s research. From the perspective of human vision, this dislocation situation can be easily perceived and found. In this paper, w...

Full description

Saved in:
Bibliographic Details
Published inThe Visual computer Vol. 40; no. 1; pp. 427 - 439
Main Authors Zhang, Jindong, Xiu, Ying
Format Journal Article
LanguageEnglish
Published Berlin/Heidelberg Springer Berlin Heidelberg 01.01.2024
Subjects
Artificial Intelligence
Computer Graphics
Computer Science
Image Processing and Computer Vision
Original Article
Human visual system
Optimal seamline detection
Image stitching
Image fusion
Online AccessGet full text

Cover

More Information
Summary:The image stitching process often produces many undesirable effects. Solving problems such as discontinuity and dislocation of pictures has always been the focus of people’s research. From the perspective of human vision, this dislocation situation can be easily perceived and found. In this paper, we propose a stitching strategy based on the human visual system (HVS) and scale-invariant feature transform (SIFT) algorithm. We preprocess the brightness difference and contrast of the stitched images, combining SIFT algorithm and HVS to divide the overlapping areas of the stitched images and establish an attribute relationship model. We use dynamic programming to find the optimal seamline according to the attribute relationship model, and the final result makes the optimal seamline almost invisible under the discriminative vision of human eyes. The experimental results show that our method has more advantages in the HVS.
ISSN:0178-2789
1432-2315
DOI:10.1007/s00371-023-02791-4