المؤلف
Ndigande, Alain P., Ozer, Sedat
تاريخ النشر
2024-01-01
مكان النشر
-
IEEE
الموضوع
Unmanned aerial vehicle (UAV) image processing, Multimodal image registration, Lukas-kanade (LK) algorithms, Infrared image registration, Image alignment, Deep learning, Corner-matching, Computer architecture, Deep learning, Image resolution, Prediction algorithms, Autonomous aerial vehicles, Cameras, Feature extraction
النوع
دورية
اللغة
الإنجليزية
رقمي
نعم
مخطوط
لا
المكتبة
جامعة اوزيجين
معرف أصل المكتبة
0196-2892
رقم السجل
15e104c3-e1f5-4c7c-952c-3eed8fdd8337
موقع المكتبة
International Relations
التاريخ
2024-01-01
نص عينة
This article proposes a deep-learning-based solution for multimodal image alignment regarding unmanned aerial vehicle (UAV)-taken images. Many recently proposed state-of-the-art alignment techniques rely on using Lucas-Kanade (LK)-based solutions for a successful alignment. However, we show that we can achieve state-of-the-art results without using LK-based methods. Our approach carefully utilizes a two-branch-based convolutional neural network (CNN) based on feature embedding blocks. We propose two variants of our approach, where in the first variant (Model A), we directly predict the new coordinates of only the four corners of the image to be aligned; and in the second one (Model B), we predict the homography matrix directly. Applying alignment on the image corners forces the algorithm to match only those four corners as opposed to computing and matching many (key) points, since the latter may cause many outliers, yielding less accurate alignment. We test our proposed approach on four aerial datasets and obtain state-of-the-art results when compared to the existing recent deep LK-based architectures.
DOI
10.1109/TGRS.2024.3367986
Cilt
62
