Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 2

Image Registration

mage registration is the spatial alignment of two or more images with mathematical mapping

I from points in one view to points in another. Image registration is widely used in remote
sensing, medical imaging,and computer vision. The majority of image registration methods consists of key point
detection, key point matching, transform model estimation, and image re-sampling and estimation (Zitova & Flusser, 2003).

I t is possible that the two images being compared may not align exactly. There are several
situations where there may be misalignment due to several geometric transformations such as
translation, rotation, scaling, and perspective transformations. In order to deal with these
transformations, a transformation matrix of homography will be used which will warp one image
with respect to another ( Pimentel-Alarcon & Nowak, 2017).

Homography

The heart of image registration technique is homography. It relates the pixel co-ordinates
in the two images (Roth, 2009). However, this homography has outlier and noise problem. An
outlier is an observation that is (a) so far from the majority of observation, or (b) somehow
can be differentiated by the general behavior (pattern) of the majority, that is should be
treated differently because this a point that is incorrectly measured or obtained due to the
presence of noise. On the other hand, noise is the meaningless data which is present during
image acquisition, coding, transmission, and processing steps and it produces undesirable
effects such as anti-facts,unrealistic edges, unseen lines, corners, and disturb backgrounds
scenes (Talafha, 2015; Boyat & Joshi, 2015).

THEORETICAL FRAMEWORK
New Related Algorithm
The New Related Algorithm computes 4 corresponding points as there is 4 points in the answer sheet. It sorts the points based on their
x-coordinates by the left-most and right-most points from the sorted x-coordinates points. Then,it sorts the left-most coordinates according to their
y-coordinates by the top-left and bottom-left points, respectively. It uses top-left coordinates as an anchor to calculate the Euclidean distance
between the top-left and right-most points; by the Pythagorean theorem, the point with largest distance will be the bottom-right point. It returns the
coordinates in top-left, top-right, bottom-right,and bottom-left order. It obtains a consistent order of the points by unpacking them individually. It
computes the width of the new image, which will be the maximum distance between bottom-right and bottom-left x-coordinates or the top-right and
top-left x-coordinates. It computes the height of the new image, which will be the maximum distance between the top-right and bottom-right
y-coordinates or the top-left and bottom-left y-coordinates. Dimensions of the new image is already there so its constructs the set of dimension
points of the image by specifying points in the top-left, top-right, bottom-left, and bottom-right order. It computes the perspective transform matrix.
As homography uses warpPerspective function in OpenCV, this New Related Algorithm also uses this warPerspective function in OpenCV to warp
the image as this algorithm is bieng implemented in Phyton.

You might also like