opencv 43 image alignment

picture is shot tilted

template

key feature matching

aligned image (left), template (right)

stack aligned on top of template

#alignment2.py

# import the necessary packages

from alignment import align_images

import numpy as np

import argparse

import imutils

import cv2

# construct the argument parser and parse the arguments

ap = argparse.ArgumentParser()

ap.add_argument("-i", "--image", required=True,

                help="path to input image that we'll align to template")

ap.add_argument("-t", "--template", required=True,

                help="path to input template image")

args = vars(ap.parse_args())

# load the input image and template from disk

print("[INFO] loading images...")

image = cv2.imread(args["image"])

template = cv2.imread(args["template"])

# align the images

print("[INFO] aligning images...")

aligned = align_images(image, template, debug=True)

# resize both the aligned and template images so we can easily

# visualize them on our screen

aligned = imutils.resize(aligned, width=700)

template = imutils.resize(template, width=700)

# our first output visualization of the image alignment will be a

# side-by-side comparison of the output aligned image and the

# template

stacked = np.hstack([aligned, template])

# our second image alignment visualization will be *overlaying* the

# aligned image on the template, that way we can obtain an idea of

# how good our image alignment is

overlay = template.copy()

output = aligned.copy()

cv2.addWeighted(overlay, 0.5, output, 0.5, 0, output)

# show the two output image alignment visualizations

cv2.imshow("Image Alignment Stacked", stacked)

cv2.imshow("Image Alignment Overlay", output)

cv2.waitKey(0)

---------------------------

#alignment.py

import numpy as np

import imutils

import cv2

def align_images(image, template, maxFeatures=500, keepPercent=0.2,

                 debug=False):

    # convert both the input image and template to grayscale

    imageGray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

    templateGray = cv2.cvtColor(template, cv2.COLOR_BGR2GRAY)

    # use ORB to detect keypoints and extract (binary) local

    # invariant features

    orb = cv2.ORB_create(maxFeatures)

    (kpsA, descsA) = orb.detectAndCompute(imageGray, None)

    (kpsB, descsB) = orb.detectAndCompute(templateGray, None)

    # match the features

    method = cv2.DESCRIPTOR_MATCHER_BRUTEFORCE_HAMMING

    matcher = cv2.DescriptorMatcher_create(method)

    matches = matcher.match(descsA, descsB, None)

    # sort the matches by their distance (the smaller the distance,

    # the "more similar" the features are)

    matches = sorted(matches, key=lambda x: x.distance)

    # keep only the top matches

    keep = int(len(matches) * keepPercent)

    matches = matches[:keep]

    # check to see if we should visualize the matched keypoints

    if debug:

        matchedVis = cv2.drawMatches(image, kpsA, template, kpsB,

                                     matches, None)

        matchedVis = imutils.resize(matchedVis, width=1000)

        cv2.imshow("Matched Keypoints", matchedVis)

        cv2.waitKey(0)

    # allocate memory for the keypoints (x, y)-coordinates from the

    # top matches -- we'll use these coordinates to compute our

    # homography matrix

    ptsA = np.zeros((len(matches), 2), dtype="float")

    ptsB = np.zeros((len(matches), 2), dtype="float")

    # loop over the top matches

    for (i, m) in enumerate(matches):

        # indicate that the two keypoints in the respective images

        # map to each other

        ptsA[i] = kpsA[m.queryIdx].pt

        ptsB[i] = kpsB[m.trainIdx].pt

    # compute the homography matrix between the two sets of matched

    # points

    (H, mask) = cv2.findHomography(ptsA, ptsB, method=cv2.RANSAC)

    # use the homography matrix to align the images

    (h, w) = template.shape[:2]

    aligned = cv2.warpPerspective(image, H, (w, h))

    # return the aligned image

    return aligned

------------------------------

#terminal

python alignment2.py --template assets/align2.jpg --image assets/align1.jpg

reference:

https://www.pyimagesearch.com/2020/08/31/image-alignment-and-registration-with-opencv/