Quantum Image Stitching Algorithm Based on Improved Harris and Quadratic Normalized Cross Correlation

Zetian Tang; Zhao Ding; Ruimin Zeng; Yang Wang; Dengwei Zhu; Yuhao Wang; Minzhe Zhong; Chen Yang

doi:10.3788/LOP57.101005

Journals >Laser & Optoelectronics Progress >Volume 57 >Issue 10 >Page 101005 > Article

Laser & Optoelectronics Progress
Vol. 57, Issue 10, 101005 (2020)

Quantum Image Stitching Algorithm Based on Improved Harris and Quadratic Normalized Cross Correlation

Zetian Tang^1、*, Zhao Ding¹, Ruimin Zeng¹, Yang Wang¹, Dengwei Zhu¹, Yuhao Wang¹, Minzhe Zhong¹, and Chen Yang^1、2、**

Author Affiliations

¹College of Big Data and Information Engineering, Guizhou University, Key Laboratory of Micro-Nano-Electronics and Software Technology of Guizhou Province, Power Semiconductor Device Reliability Engineering Center of the Ministry of Education, Guiyang, Guizhou 550025, China

²Longmaster Information & Technology Co., Guiyang, Guizhou 550022, China;

show less

DOI: 10.3788/LOP57.101005 Cite this Article Set citation alerts

Zetian Tang, Zhao Ding, Ruimin Zeng, Yang Wang, Dengwei Zhu, Yuhao Wang, Minzhe Zhong, Chen Yang. Quantum Image Stitching Algorithm Based on Improved Harris and Quadratic Normalized Cross Correlation[J]. Laser & Optoelectronics Progress, 2020, 57(10): 101005 Copy Citation Text

show less

Fig. 1. Flow chart of quantum image stitching

Download full size

Fig. 2. Impact on corner detection by threshold. (a) Small threshold; (b) excessive threshold

Download full size

Fig. 3. Error matching

Download full size

Fig. 4. Flow chart of statistical method for the number of quantum dots or rings

Download full size

Fig. 5. Gray value logarithmic histogram

Download full size

Fig. 6. Statistical graph of the relationship between quantum dots or rings and subregions with preserved corner points

Download full size

Fig. 7. Original images. (a)(b) Quantum wire images; (c)(d) quantum ring images; (e)(f) quantum dot images

Download full size

Fig. 8. Corner detection results. (a)(c)(e) Traditional Harris algorithm; (b)(d)(f) proposed algorithm

Download full size

Fig. 9. Results of corner matching. (a) Quantum wire stitching image; (b) quantum ring stitching image; (c) quantum dot stitching image

Download full size

Fig. 10. Image stitching results. (a) Quantum wire stitching image; (b) quantum ring stitching image; (c) quantum dot stitching image

Download full size

		1	2	3	4	5	6
Image
Number ofquantumdots or rings	OTSU	3	2	23	22	19	21
	Ref. [12]	11	10	40	46	28	26
	Proposed	8	10	38	47	27	25
	Manual counting	11	10	44	46	28	26
	OTSU	0.028	0.023	0.021	0.021	0.016	0.010
Time /s	Ref. [12]	0.517	0.513	0.737	0.705	0.703	0.779
	Proposed	0.082	0.075	0.269	0.260	0.296	0.302

Table 1. Comparison of quantum dot or ring counting methods

Algorithm	Fig. 7(a)	Fig. 7(b)	Fig. 7(c)	Fig. 7(d)	Fig. 7(e)	Fig. 7(f)
Harris	1277	1099	157	174	90	252
Proposed	678	513	238	335	592	514

Table 2. Comparison of corner detection results

Window size	Fig. 7(a)(b)		Fig. 7(c)(d)	Fig. 7(e)(f)
Window size	Error rate /%	Time /s		Error rate /%	Time /s		Error rate /%	Time /s
5×5	66.88		3.929		82.86	1.608	67.35	3.551
9×9	61.10		4.026		72.32	1.738	66.77	3.728
13×13	56.01		4.154		54.17	1.876	65.79	3.941
17×17	50.53		4.409		38.56	1.966	63.24	4.134
21×21	45.33		4.830		27.59	2.068	61.67	4.569
25×25	41.22		5.188		25.00	2.179	58.50	4.916
Proposed	4.82		3.982		27.27	1.688	10.17	3.682

Table 3. Comparison results of mismatch rate and time

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information