Image Stitching Combining Enhanced Optimal Seam and Optimized Brightness

Xiangtan Yu; Yaohong Zhao; Wei Xiang

doi:10.3788/LOP232779

Laser & Optoelectronics Progress, Volume. 61, Issue 18, 1837006(2024)

Image Stitching Combining Enhanced Optimal Seam and Optimized Brightness

Xiangtan Yu^1,2,3,4, Yaohong Zhao^1,2、*, and Wei Xiang^1,2

Author Affiliations

¹Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, Liaoning , China

²Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang 110016, Liaoning , China

³Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110169, Liaoning , China

⁴University of Chinese Academy of Sciences, Beijing 100049, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(11)

Fig. 1. Flow chart of proposed algorithm

Download full size

Fig. 2. Comparison of registration effects. (a) Before registration; (b) after registration

Download full size

Fig. 3. Schematic diagrams of seam search. (a) Traditional algorithm; (b) enhancement algorithm

Download full size

Fig. 4. Definition and application diagrams of direction matrix D. (a) Definition diagram; (b) application diagram

Download full size

Fig. 5. Relationship between $β$ and the number of iterations/energy value

Download full size

Fig. 6. Comparison of results between traditional algorithm and enhanced algorithm. (a) Schematic diagram of traditional algorithm for searching seam in overlapping area; (b) traditional algorithm result; (c) schematic diagram of enhanced algorithm for searching seam in overlapping area; (d) enhanced algorithm result

Download full size

Fig. 7. Comparison of large parallax processing effects. (a)‒(c) QEISE; (d)‒(f) LPC; (g)‒(i) proposed algorithm; (j)‒(l) proposed algorithm with seam

Download full size

Fig. 8. Images with exposure differences

Download full size

Fig. 9. Comparison of exposure difference processing effects. (a) (g) Before optimization; (b) (h) after optimization; (c) (i) HMM; (d)‒(f) (j)‒(l) enlarged images of details

Download full size

Table 1. Comparison of SSIM/PSNR on different algorithms

View table

Table 1. Comparison of SSIM/PSNR on different algorithms

Image	QEISE^［10］		LPC^［6］		Algorithm of reference［11］		Proposed algorithm
Image	SSIM	PSNR	SSIM	PSNR	SSIM	PSNR	SSIM	PSNR
gym	0.7205	51.2564	0.6479	44.5121	0.8776	47.1115	0.9345	53.8936
temple	0.6843	54.6188	0.5602	36.6599	0.8640	46.1547	0.9284	58.2805
standing-he	0.5970	51.5301	0.4961	40.3860	0.5055	41.3309	0.6536	53.5421
four	0.6697	51.5301	0.5678	40.8646	0.9295	50.9267	0.9512	54.6080
foundation	0.5452	55.9699	0.5583	42.9826	0.7465	51.5919	0.8085	59.0526
fence	0.6199	52.7863	0.4906	40.9433	0.8032	52.0495	0.8222	52.9691
building	0.7432	48.2974	0.6193	39.1166	0.8819	50.8518	0.8992	53.2437
corner	0.6252	55.1316	0.5066	32.8505	0.8649	54.8767	0.9075	55.4981
carpark	0.6636	54.5507	0.6702	45.9438	0.9046	55.0998	0.9072	55.6421
train	0.4636	48.6395	0.5282	49.4773	0.6129	49.1685	0.6248	50.0992

Table 2. Performance evaluation indicators for panoramic image before and after optimization
View table
Table 2. Performance evaluation indicators for panoramic image before and after optimization
Parameter Before optimization After optimization
AGIC 0.8595 0.7314
AG 7.3086 8.1921

Tools

Get Citation

Copy Citation Text

Xiangtan Yu, Yaohong Zhao, Wei Xiang. Image Stitching Combining Enhanced Optimal Seam and Optimized Brightness[J]. Laser & Optoelectronics Progress, 2024, 61(18): 1837006

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites