Premium
An efficient intensity‐based ready‐to‐use X‐ray image stitcher
Author(s) -
Wang Junchen,
Zhang Xiaohui,
Sun Zhen,
Yuan Fuzhen
Publication year - 2018
Publication title -
the international journal of medical robotics and computer assisted surgery
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.556
H-Index - 53
eISSN - 1478-596X
pISSN - 1478-5951
DOI - 10.1002/rcs.1925
Subject(s) - image stitching , image warping , computer science , computer vision , artificial intelligence , image (mathematics) , feature (linguistics) , philosophy , linguistics
Abstract Background: The limited field of view of the X‐ray image intensifier makes it difficult to cover a large target area with a single X‐ray image. X‐ray image stitching techniques have been proposed to produce a panoramic X‐ray image. Methods: This paper presents an efficient intensity‐based X‐ray image stitcher, which does not rely on accurate C‐arm motion control or auxiliary devices and hence is ready to use in clinic. The stitcher consumes sequentially captured X‐ray images with overlap areas and automatically produces a panoramic image. The gradient information for optimization of image alignment is obtained using a back‐propagation scheme so that it is convenient to adopt various image warping models. Results: The proposed stitcher has the following advantages over existing methods: (1) no additional hardware modification or auxiliary markers are needed; (2) it is robust against feature‐based approaches; (3) arbitrary warping models and shapes of the region of interest are supported; (4) seamless stitching is achieved using multi‐band blending. Experiments have been performed to confirm the effectiveness of the proposed method. Conclusion: The proposed X‐ray image stitcher is efficient, accurate and ready to use in clinic.