Indian Journal of Science and Technology
Year: 2023, Volume: 16, Issue: 45, Pages: 4225-4232
Original Article
Manjusha Deshmukh1*
1Principal, Sarasati College of Engineering, Kharghar, Mumbai, Maharashtra, India
*Corresponding Author
Email: [email protected]
Received Date:27 April 2023, Accepted Date:30 October 2023, Published Date:06 December 2023
Objective: Registration is a critical step in any image analysis. In the Radon Transform, rotation can be expressed as translation, and the Fourier Transform is invariant to translation. In this paper we present an efficient combination of Radon Transform and Fourier Transform for obtaining transformation to correct scale, translation and rotation. This concept is used for image matching. Method : This study presents an innovative Radon Transform to register two images. The method involves calculating a Radon Transform first, followed by a Fourier description for each image that needs to be matched. Mono-modal visuals are used in experiments. The Lena dataset is used for correcting scale, translation, and rotation. The same algorithm is used on around 36 images to correct scale, translation, rotation, scale and rotation, rotation and translation, scale and translation etc. Findings : This algorithm is tested successfully with different datasets, which differ by translation and/or rotation. Lena image dataset is used that was scaled, rotated and translated with the objective of template matching. Around 36 datasets are used for experimentation. Novelty : Innovative method of combination of Fourier and Radon Transform for image registration is attempted with 98 % accuracy.
Keywords: Image registration, image fusion, Radon Transform, Fourier transforms, affine transform
© 2023 Deshmukh. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Published By Indian Society for Education and Environment (iSee)
Subscribe now for latest articles and news.