Panoramic image stitching using cross correlation and phase correlation methods

1st International E-Conference on Recent Developments in Science, Engineering and Information Technology on 23rd to 25th September, 2020 Department of Computer Science, DDE, Madurai Kamaraj University, Tamil Nadu, India. International Journal of Computer Science (IJCS) Published by SK Research Group of Companies (SKRGC)

Abstract

Image stitching / mosaicking is the process of aligning multiple images together to form a single high resolution image known as panoramic image. Stitching multiple images of same scene captured from different positions to form a high resolution single image is a challenging task. Images taken by a single snap will permit only a limited view whereas panoramic image allows a wide angle view. In order to create a panoramic image, the simple method is to stitch two or more images of a scene into single composite image. Mainly there are two approaches for image stitching such as “direct approach and feature based approach”. In this paper, we discuss two basic simple approaches of direct method for image stitching. First method is correlation based and second one is phase correlation. The proposed cross correlation method can stitch multiple images having horizontal displacement or vertical displacement with each other. But it cannot handle images having both horizontal and vertical displacement simultaneously. In order to solve this problem, we proposed phase correlation based image stitching that can be used for measuring the motion parameter i.e. both horizontal displacement and vertical displacement present in images. Hence, we can stitch multiple images having both horizontal and vertical displacement simultaneously. In cross correlation based method, three images of same scene with different orientations were put together to form a high resolution panoramic view. But a pair of images were used to stitch together to form a panoramic image using phase correlation based method. The proposed cross correlation method is not completely translation invariant, whereas the phase correlation based stitching is invariant to translation. But the main disadvantages of both methods are that they are not invariant to image scale and rotation.

References

[1] Adel, E., M. Elmogy, and H. Elbakry (2014). Article: Image stitching based on feature extraction techniques: A survey. International Journal of Computer Applications, 99(6), 1–8. Full text available.

[2] Arya, S. (2015).Article: A review on image stitching and its different methods. International Journal of Advanced Research in Computer Science and Software Engineering.

[3] Bannore, V. (2009). Iterative-interpolation super-resolution image reconstruction. Studies in Computational Intelligence, 195, 1-8.

[4] Brown, L. G. (1992). A survey of image registration techniques. ACM computing surveys (CSUR), 24(4), 325–376.

[5] Shashank, K., N. S. Chaitanya, G. Manikanta, C. N. Balaji, and V. Murthy (2014). A survey and review over image alignment and stitching methods. J. Electron. Commun. Technol, 5, 50– 52.

[6] Pangarkar, P., V. B. Gaikwad, and R. Prakash (2015). A survey on panorama image stitching. International Journal of Engineering Research & Management Technology.

[7] Bhat, A. S., Shivaprakash, A. V., Prasad, N. S., & Nagaraj, C. (2013, July). Template matching technique for panoramic image stitching. In 2013 7th Asia Modelling Symposium (pp. 111-115). IEEE.

[8] Swaroop, P., & Sharma, N. (2016). An overview of various template matching methodologies in image processing. International Journal of Computer Applications, 153(10), 8-14.

[9] Zitova, B., & Flusser, J. (2003). Image registration methods: a survey. Image and vision computing, 21(11), 977-1000.

[10] Szeliski, R. et al. (2007). Image alignment and stitching: A tutorial. Foundations and Trends® in Computer Graphics and Vision, 2(1), 1–104.

[11] Mistry, S., & Patel, A. (2016). Image Stitching using Harris Feature Detection. International Research Journal of Engineering and Technology (IRJET), 3(04), 2220-6.

[12] Kale, P., & Singh, K. R. (2015). A technical analysis of image stitching algorithm. International Journal of Computer Science and Information Technologies, 6(1), 284-288.

[13] Harris, C. G., & Stephens, M. (1988, August). A combined corner and edge detector. In Alvey vision conference (Vol. 15, No. 50, pp. 10-5244).

[14] Lindeberg, T. (2012). Scale invariant feature transform.

[15] Juan, L., & Gwon, L. (2007). A comparison of sift, pca-sift and surf. International Journal of Signal Processing, Image Processing and Pattern Recognition, 8(3), 169-176.

[16] Bay, H., Tuytelaars, T., & Van Gool, L. (2006, May). Surf: Speeded up robust features. In European conference on computer vision (pp. 404-417). Springer, Berlin, Heidelberg.

[17]6Rosten, E., & Drummond, T. (2006, May). Machine learning for high-speed corner detection. In European conference on computer vision (pp. 430-443). Springer, Berlin, Heidelberg.

[18] Chen, J., Luo, L., Wang, S., & Wu, H. (2018, July). Automatic Panoramic UAV Image Mosaic Using ORB Features and Robust Transformation Estimation. In 2018 37th Chinese Control Conference (CCC) (pp. 4265-4270). IEEE

[19] Raghavachari, C., & GA, S. S. (2014, April). Efficient use of bandwidth by image compression for vision-based robotic navigation and control. In 2014 International Conference on Communication and Signal Processing (pp. 513-517). IEEE

[20] Zyada, Z., & Fukuda, T. (2007, September). 3D template based automatic landmine detection from GPR data. In SICE Annual Conference 2007 (pp. 1552-1557). IEEE

[21] Briechle, K., & Hanebeck, U. D. (2001, March). Template matching using fast normalized cross correlation. In Optical Pattern Recognition XII (Vol. 4387, pp. 95-102). International Society for Optics and Photonics.

[22] Ray, R., Kumar, V., Banerji, D., & Shome, S. N. (2012, February). Simultaneous Localisation and Image Intensity Based Occupancy Grid Map Building--A New Approach. In 2012 Third International Conference on Intelligent Systems Modelling and Simulation (pp. 143-148). IEEE

[23] Roche, A., Malandain, G., Pennec, X., & Ayache, N. (1998, October). The correlation ratio as a new similarity measure for multimodal image registration. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 1115-1124). Springer, Berlin, Heidelberg.

[24] Kang, J., Kim, J., Lee, I., & Kim, K. (2019). Minimum Error Seam-Based Efficient Panorama Video Stitching Method Robust to Parallax. IEEE Access, 7, 167127-167140.

[25] Deng, Y., & Zhang, T. (2003, November). Generating panorama photos. In Internet Multimedia Management Systems IV (Vol. 5242, pp. 270-279). International Society for Optics and Photonics.

[26] Douini, Y. et al (2017). Solving sub-pixel image registration problems using phase correlation and Lucas- Kanade optical flow method.21 (11), 977–1000. Processing (IJIP) Volume (3), Issue (4).

Keywords

Panoramic image, image stitching, cross – correlation, template matching, phase correlation, registration, blending.