Analysis of Descent-Based Image Registration

Download

index.pdf

Date

2013-01-01

Author

Vural, Elif

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

168
views

0
downloads

We present a performance analysis for image registration with gradient descent. We consider a typical multiscale registration setting where the global two-dimensional translation between a pair of images is estimated by smoothing the images and minimizing the distance between them with gradient descent. Our study particularly concentrates on the effect of noise and low-pass filtering on the alignment accuracy. We analyze the well-behavedness of the image distance function by estimating the neighborhood of translations for which it is free of undesired local minima. This is the neighborhood of translations that are correctly computable with a simple gradient descent minimization. We show that the area of this neighborhood increases at least quadratically with the smoothing filter size. We then examine the effect of noise on the alignment accuracy and derive an upper bound for the alignment error in terms of the noise properties and filter size. Our main finding is that the error increases at a rate that is at least linear with respect to the filter size. Therefore, smoothing improves the well-behavedness of the distance function; however, this comes at the cost of amplifying the alignment error in noisy settings. Our results provide a mathematical insight into why hierarchical techniques are effective in image registration, suggesting that the multiscale alignment strategy of these techniques is very suitable from the perspective of the tradeoff between the well-behavedness of the objective function and the registration accuracy. To the best of our knowledge, this is the first such study for descent-based image registration.

Subject Keywords

Image registration, Hierarchical registration methods, Image smoothing, Gradient descent, Performance analysis

URI

https://hdl.handle.net/11511/39999

Journal

SIAM JOURNAL ON IMAGING SCIENCES

DOI

https://doi.org/10.1137/130909858

Collections

Department of Electrical and Electronics Engineering, Article

Suggestions

OpenMETU
Core

Analysis of Image Registration with Tangent Distance Vural, Elif (2014-01-01) The computation of the geometric transformation between a reference and a target image, known as registration or alignment, corresponds to the projection of the target image onto the transformation manifold of the reference image (the set of images generated by its geometric transformations). However, it often takes a nontrivial form such that the exact computation of projections on the manifold is difficult. The tangent distance method is an effective algorithm for solving this problem by exploiting a line...
Optimization of Mesa Structured InGaAs Based Photodiode Arrays Dolas, M. Halit; Çırçır, Kübra; Kocaman, Serdar (2017-04-13) We design lattice matched InP/In0.53Ga0.47As mesa structured heterojunction p-n photodiodes with a novel passivation methodology based on a fully depleted thin p-InP layer. Mesa-structured detectors are targeted due to their competitive advantages for applications such as multicolor/hyperspectral imaging. Test detector pixels with different perimeter/area ratios are fabricated with and without etching thin InP passivation layer between pixels in order to comparatively examine passivating behavior. I-V chara...
Towards Effective Image Classification Using Class-Specific Codebooks and Distinctive Local Features Altintakan, Umit Lutfu; Yazıcı, Adnan (2015-03-01) Local image features, which are robust to scale, view, and orientation changes in images, play a key factor in developing effective visual classification systems. However, there are two main limitations to exploit these features in image classification problems: 1) a large number of key-points are located during the feature detection process, and 2) most of the key-points arise in background regions, which do not contribute to the classification process. In order to decrease the inverse effects of these lim...
Robust, object-based high-resolution image reconstruction from low-resolution video Eren, Pekin Erhan; Sezan, MI; Tekalp, AM (1997-10-01) We propose a robust, object-based approach to high-resolution image reconstruction from video using the projections onto convex sets (POCS) framework. The proposed method employs a validity map and/or a segmentation map. The validity map disables projections based on observations with inaccurate motion information for robust reconstruction in the presence of motion estimation errors; while the segmentation map enables object-based processing where more accurate motion models can be utilized to improve the q...
Analysis of correlated circular and extremal data with a flexible cylindrical distribution Kalaylıoğlu Akyıldız, Zeynep Işıl (2021-08-01) In this article, we introduce a flexible cylindrical distribution for modeling and analysis of dependent extremal and directional observations. The distribution can be used to investigate the connection between two related phenomena, such as the daily fastest wind speed and its direction. The proposed model is applicable for the analysis of a wide variety of cylindrical data, including datasets with asymmetrically distributed directional observations. The model enjoys the advantages of interpretable model p...

Citation Formats

E. Vural, “Analysis of Descent-Based Image Registration,” SIAM JOURNAL ON IMAGING SCIENCES, pp. 2310–2349, 2013, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/39999.