Seamless Stitching Algorithm with Strong Structure Preservation for Wide-Parallax Images

doi:10.19678/j.issn.1000-3428.0252122

Abstract

Abstract:

To address the key issues in image stitching caused by wide-parallax, such as geometric distortion, ghosting artifacts, and visible seams, a seamless stitching algorithm with strong structure preservation for wide-parallax images is proposed. This algorithm achieves full-process structure preservation, from optimized registration to precise fusion, by constructing a multidimensional constraint model. First, during the registration phase, a scene-guided adaptive mesh deformation model is established, and dynamic weighted vectors are constructed based on the feature distribution and proportion of overlapping areas, significantly improving the spatial registration accuracy. Second, a dual-edge detection mechanism combining strong and weak detection is proposed to capture residual structural misalignment, and a novel significant texture measurement model is designed using feature-coordinate convolution, enabling precise perception of significant structures in the overlapping region. Finally, in the fusion phase, an optimal seam cost function is constructed considering color difference, structural difference, and significant texture degree as smoothness constraints, yielding a seam mask that balances structural integrity and naturalness of the image. Experimental results demonstrate that this algorithm effectively solve the problem of structural distortion in wide-parallax stitched images by generating high-definition and visually coherent panoramic images with a large viewing angle. Compared with six advanced image stitching algorithms, this algorithm improves the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) by at least 8.14% and 10.44%, respectively, verifying its technical advantages.

Key words: image stitching, structure preservation, adaptive registration, optimal stitching seam, energy function optimization

摘要：

针对图像拼接中因视差效应引发的几何畸变、重影伪影及拼接缝显露等关键问题, 提出一种强结构保护的宽视差图像无缝拼接算法。该算法通过构建多维度约束模型, 实现了从优化配准到精准融合的全流程结构保护: 首先, 在配准阶段建立场景引导的自适应网格变形模型, 基于特征分布与重叠区域比例构造动态加权向量, 显著提升了配准空间准确性; 其次, 提出强弱双重边缘检测机制以捕捉残余结构错位, 并基于特征坐标卷积设计一种新型显著纹理度量模型, 实现对重叠区域显著结构的精准感知; 最后, 在融合阶段构建最优拼接缝代价函数, 将色差、结构差和显著纹理度联合作为平滑度约束条件, 得到兼顾图像结构完整性和自然性的拼接缝掩膜。实验结果表明, 该算法能有效解决宽视差拼接图像中存在的结构失真问题, 生成清晰度高且视觉连贯性强的大视角全景图像。与6种先进的图像拼接算法相比, 该算法的峰值信噪比(PSNR)和结构相似度指数度量(SSIM)指标分别提升了8.14%和10.44%以上, 验证了其技术优势。

关键词: 图像拼接, 结构保护, 自适应配准, 最优拼接缝, 能量函数优化

YANG Ruoyi, LIU Lidong, QU Jingkang. Seamless Stitching Algorithm with Strong Structure Preservation for Wide-Parallax Images[J]. Computer Engineering, 2026, 52(9): 242-252.

杨若怡, 刘立东, 曲敬康. 强结构保护的宽视差图像无缝拼接算法[J]. 计算机工程, 2026, 52(9): 242-252.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0252122

https://www.ecice06.com/EN/Y2026/V52/I9/242

Figures/Tables 16

Fig.1 Stitching seam in overlapping area

Fig.2 Procedure of the method in this paper

Fig.3 Image alignment examples

Fig.4 Comparison of edge extraction effect

Fig.5 Significant texture degree contrast results

Fig.6 Comparison of stitching results of park scene images

Fig.7 Comparison of stitching results of intersection scene images

Fig.8 Comparison of stitching results of signboard scene images

Fig.9 Wide-parallax image data

Fig.10 Experimental results of PSNR and SSIM indexes for different algorithms

Fig.11 Stitching results and details of ablation experiments

Fig.12 Stitching results with small differences in scene depth

Fig.13 Failed stitching results with large differences in scene depth

Fig.14 Comparison of stitching result of square scene images

References 32

1	胡育诚, 芮挺, 杨成松, 等. 基于改进SIFT的无人机航拍图像快速配准研究. 计算机科学, 2021, 48 (8): 134- 138.
	HU Y C , RUI T , YANG C S , et al. Study on aerial image fast registration from UAV. Computer Science, 2021, 48 (8): 134- 138.
2	郭泓邑, 张红民, 李萍萍. 一种视差伪影最小化的视频拼接方法. 计算机工程与应用, 2020, 56 (20): 186- 190.
	GUO H Y , ZHANG H M , LI P P . Video stitching method for minimizing parallax artifacts. Computer Engineering and Applications, 2020, 56 (20): 186- 190.
3	ZARAGOZA J, CHIN T J, BROWN M S, et al. As-projective-as-possible image stitching with moving DLT[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Washington D.C., USA: IEEE Press, 2013: 2339-2346.
4	LI J , DENG B S , TANG R F , et al. Local-adaptive image alignment based on triangular facet approximation. IEEE Transactions on Image Processing, 2020, 29, 2356- 2369. doi: 10.1109/TIP.2019.2949424
5	LEE K Y, SIM J Y. Warping residual based image stitching for large parallax[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2020: 8195-8203.
6	ZHANG Z H , YANG X Q , XU C . Natural image stitching with layered warping constraint. IEEE Transactions on Multimedia, 2023, 25, 329- 338. doi: 10.1109/TMM.2021.3126157
7	XUE W L , CHEN Y Y , YANG Y H , et al. Image stitching via augmentation selection and comprehensive optimization of homography models. IEEE Transactions on Consumer Electronics, 2024, 70 (1): 3299- 3307. doi: 10.1109/TCE.2023.3309271
8	LIN M Y , LIU T B , LI Y , et al. Image stitching by disparity-guided multi-plane alignment. Signal Processing, 2022, 197, 108534. doi: 10.1016/j.sigpro.2022.108534
9	JIANG Z Y , ZHANG Z X , LIU J Y , et al. Multispectral image stitching via global-aware quadrature pyramid regression. IEEE Transactions on Image Processing, 2024, 33, 4288- 4302. doi: 10.1109/TIP.2024.3430532
10	LIAO T L , LI N . Single-perspective warps in natural image stitching. IEEE Transactions on Image Processing, 2020, 29, 724- 735. doi: 10.1109/TIP.2019.2934344
11	JIA Q, LI Z J, FAN X, et al. Leveraging line-point consistence to preserve structures for wide parallax image stitching[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2021: 12181-12190.
12	LI Y X, ZHAO H T, JIA Q, et al. Image stitching based on multi-scale meshes[C]//Proceedings of the IEEE International Conference on Image Processing (ICIP). Washington D.C., USA: IEEE Press, 2023: 2895-2899.
13	庞荣来, 林静, 张磊. 网格驱动的双向图像拼接算法. 计算机科学, 2020, 47 (3): 130- 136.
	PANG R L , LIN J , ZHANG L . Grid-driven bi-directional image stitching algorithm. Computer Science, 2020, 47 (3): 130- 136.
14	ZHANG J D , JIANG N N . Image stitching algorithm based on semantics-preserving warps. Signal, Image and Video Processing, 2024, 19 (1): 40.
15	李婕, 周顺. 基于超像素和光流引导的正射影像拼接方法. 计算机工程, 2022, 48 (3): 263- 270. doi: 10.19678/j.issn.1000-3428.0060568
	LI J , ZHOU S . Orthophoto splicing method guided by superpixel and optical flow. Computer Engineering, 2022, 48 (3): 263- 270. doi: 10.19678/j.issn.1000-3428.0060568
16	GAO J, LI Y, CHIN T J, et al. Seam-driven image stitching[EB/OL]. [2025-01-11]. https://www.semanticscholar.org/paper/Seam-Driven-Image-Stitching-Gao-Li/0c97bbc90dd710e3de69cb75d7b4bb00edc1f086.
17	WEN S P , WANG X L , ZHANG W C , et al. Structure preservation and seam optimization for parallax-tolerant image stitching. IEEE Access, 2022, 10, 78713- 78725. doi: 10.1109/ACCESS.2022.3194245
18	LIN K M, JIANG N J, CHEONG L F, et al. SEAGULL: seam-guided local alignment for parallax-tolerant image stitching[C]//Proceedings of ECCV 2016. Berlin, Germany: Springer, 2016: 370-385.
19	LI N , LIAO T L , WANG C . Perception-based seam cutting for image stitching. Signal, Image and Video Processing, 2018, 12 (5): 967- 974. doi: 10.1007/s11760-018-1241-9
20	HAN G J , ZHANG Y Z , ZHOU M C . Image stitching algorithm based on two-stage optimal seam line search. Journal of King Saud University (Computer and Information Sciences), 2024, 36 (10): 102256. doi: 10.1016/j.jksuci.2024.102256
21	ZHANG Z H , HE J , SHEN M Q , et al. Multimodel fore-/background alignment for seam-based parallax-tolerant image stitching. Computer Vision and Image Understanding, 2024, 240, 103912. doi: 10.1016/j.cviu.2023.103912
22	MIAO X P , QU T , CHEN X , et al. Superpixel-based foreground-preserving image stitching. Machine Vision and Applications, 2022, 34 (1): 17.
23	NIE L , LIN C Y , LIAO K , et al. Unsupervised deep image stitching: reconstructing stitched features to images. IEEE Transactions on Image Processing, 2021, 30, 6184- 6197. doi: 10.1109/TIP.2021.3092828
24	NIE L, LIN C Y, LIAO K, et al. Parallax-tolerant unsupervised deep image stitching[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). Washington D.C., USA: IEEE Press, 2024: 7365-7374.
25	田丽丽, 郭敏, 徐秋平. 基于图割与GVF Snake的凹型目标快速提取算法. 计算机应用, 2008, 28 (10): 2633- 2635.
	TIAN L L , GUO M , XU Q P . Fast concave object extraction algorithm based on graph cuts and GVF Snake. Journal of Computer Applications, 2008, 28 (10): 2633- 2635.
26	XIE S N , TU Z W . Holistically-nested edge detection. International Journal of Computer Vision, 2017, 125 (1): 3- 18.
27	VON GIOI R G , JAKUBOWICZ J , MOREL J M , et al. LSD: a fast line segment detector with a false detection control. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32 (4): 722- 732. doi: 10.1109/TPAMI.2008.300
28	ZHANG J M, SCLAROFF S, LIN Z, et al. Minimum barrier salient object detection at 80 FPS[C]//Proceedings of the IEEE International Conference on Computer Vision (ICCV). Washington D.C., USA: IEEE Press, 2016: 1404-1412.
29	YUAN Y T , FANG F M , ZHANG G X . Superpixel-based seamless image stitching for UAV images. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59 (2): 1565- 1576. doi: 10.1109/TGRS.2020.2999404
30	LI J , WANG Z M , LAI S M , et al. Parallax-tolerant image stitching based on robust elastic warping. IEEE Transactions on Multimedia, 2018, 20 (7): 1672- 1687. doi: 10.1109/TMM.2017.2777461
31	PENG Z Y , MA Y , ZHANG Y J , et al. Seamless UAV hyperspectral image stitching using optimal seamline detection via graph cuts. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61, 5512213.
32	KIM M, LEE J, LEE B, et al. Implicit neural image stitching with enhanced and blended feature reconstruction[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). Washington D.C., USA: IEEE Press, 2024: 4075-4084.

[1]	Jintao ZHOU, Diju GAO, Zhiquan LIU. Detection Method of Water-Surface Obstacles for Unmanned Ships Based on Panoramic Vision [J]. Computer Engineering, 2024, 50(2): 113-121.
[2]	WU Lushen, CHEN Xiaodu. An Image Stitching Algorithm Based on Improved AKAZE Feature and RANSAC [J]. Computer Engineering, 2021, 47(1): 246-254.
[3]	QI Xiangming, WANG Jiaqi. Large Parallax Image Stitching Algorithm for Mesh Deformation Subdivision [J]. Computer Engineering, 2020, 46(1): 236-242.
[4]	ZHANG Jingjing,ZHAI Donghai,HUANG Lizhi,YU Qiang. Parallax Image Stitching Algorithm Based on Feature Blocking [J]. Computer Engineering, 2018, 44(5): 220-226.
[5]	XU Min,MO Dongming. Application of 18-dimensional Descriptor of Canny Edge Feature in Image Stitching [J]. Computer Engineering, 2017, 43(9): 310-315.
[6]	WU Jiana, LAN Shi-yonga,b, HUANG Fei-hua. Design and Implementation of a Multi-channel Fast Video Stitching System [J]. Computer Engineering, 2014, 40(2): 208-211,218.
[7]	YANG Xiao-Hui, WANG Min. Seamless Image Stitching Method Based on ASIFT [J]. Computer Engineering, 2013, 39(2): 241-244.
[8]	ZHENG Jing-Bi, WANG Chun-Xia. Improved Method for Generating Global View Face Texture Image [J]. Computer Engineering, 2012, 38(9): 228-230,251.
[9]	ZHANG Gui-hua; WANG Cheng-liang. Sequence Annular Image Processing Technology and Its Application [J]. Computer Engineering, 2009, 35(24): 208-210.

Please choose a citation manager

Content to export