Image Registration and Change Detection under Rolling Shutter Motion Blur

Abstract

In this paper, we address the problem of registering a distorted image and a reference image of the same scene by estimating the camera motion that had caused the distortion. We simultaneously detect the regions of changes between the two images. We attend to the coalesced effect of rolling shutter and motion blur that occurs frequently in moving CMOS cameras. We first model a general image formation framework for a 3D scene following a layered approach in the presence of rolling shutter and motion blur. We then develop an algorithm which performs layered registration to detect changes. This algorithm includes an optimisation problem that leverages the sparsity of the camera trajectory in the pose space and the sparsity of changes in the spatial domain. We create a synthetic dataset for change detection in the presence of motion blur and rolling shutter effect covering different types of camera motion for both planar and 3D scenes. We compare our method with existing registration methods and also show several real examples captured with CMOS cameras.

Obtaining World Coordinate Information of UAV in GNSS Denied Environments

“…There are two imaging methods for the camera: One is to take pictures with a global shutter, but if the exposure time is too long, the pictures will produce image paste phenomenon [37,38]. The other way is shooting with a rolling shutter, but if the progressive scanning speed is not enough, the shooting result may appear in any situation such as "tilt", "swing uncertain", or "partial exposure" [39,40]. The best way to overcome the blur is to reduce the relative motion between the camera and the target [37][38][39][40].…”

“…The other way is shooting with a rolling shutter, but if the progressive scanning speed is not enough, the shooting result may appear in any situation such as "tilt", "swing uncertain", or "partial exposure" [39,40]. The best way to overcome the blur is to reduce the relative motion between the camera and the target [37][38][39][40]. Therefore, the PBVS system keeps the target on the optical axis of the camera all the time through the pan-tilt control camera, to ensure the effectiveness of target tracking to the greatest extent.…”

Blind First-Order Perspective Distortion Correction Using Parallel Convolutional Neural Networks

“…Some works use multiple images with different orientations to properly estimate distortion parameters [ 24 , 25 , 26 ]. To some extent, methods that combine multiple images for enhancement require some perspective transformation technique [ 27 , 28 , 29 , 30 ]. The same technique is implemented for performing image stitching [ 31 , 32 , 33 , 34 , 35 ].…”

Разработка Методов Удаления Дефектов На Изображениях Энергообъектов, Полученных В Процессе Дистанционного Мониторинга С Помощью Бпла

“…Следует, однако, отметить, что представленные авторами алгоритмы автокалибровки и оценки частоты колебаний требуют значительных вычислительных затрат и не всегда применимы в случае встроенной микроконтроллерной реализации бортового компьютера БПЛА. В работах [7,8] также анализируется последовательность кадров, и предпринимаются попытки выделения областей изображения, подверженных, артефактам связанных с эффектом построчного переноса. Приведенные алгоритмы тестировались на реальных видеопотоках, в том числе и полученных с БПЛА.…”