In [1]:
%matplotlib inline
from matplotlib import pyplot as plt
In [2]:
import imutils
import cv2
import numpy as np
In [3]:
In [4]:
def detectAndDescribe(image):
    # convert the image to grayscale
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    descriptor = cv2.xfeatures2d.SIFT_create()
    (kps, features) = descriptor.detectAndCompute(image, None)
    # convert the keypoints from KeyPoint objects to NumPy
    # arrays
    kps = np.float32([ for kp in kps])
    # return a tuple of keypoints and features
    return (kps, features)
In [5]:
def matchKeypoints(kpsA, kpsB, featuresA, featuresB, ratio, reprojThresh):
    # compute the raw matches and initialize the list of actual
    # matches
    matcher = cv2.DescriptorMatcher_create("BruteForce")
    rawMatches = matcher.knnMatch(featuresA, featuresB, 2)
    matches = []
    # loop over the raw matches
    for m in rawMatches:
        # ensure the distance is within a certain ratio of each
        # other (i.e. Lowe's ratio test)
        if len(m) == 2 and m[0].distance < m[1].distance * ratio:
            matches.append((m[0].trainIdx, m[0].queryIdx))
    # computing a homography requires at least 4 matches
    if len(matches) > 4:
        # construct the two sets of points
        ptsA = np.float32([kpsA[i] for (_, i) in matches])
        ptsB = np.float32([kpsB[i] for (i, _) in matches])

        # compute the homography between the two sets of points
        (H, status) = cv2.findHomography(ptsA, ptsB, cv2.RANSAC,
        # return the matches along with the homograpy matrix
        # and status of each matched point
        return (matches, H, status)
    # otherwise, no homograpy could be computed
    return None
In [6]:
# load the two images
imageA = cv2.imread('1orig.png')
imageB = cv2.imread('2orig.png')


Habcam images are "along-track" to adapt to the existing code i'm rotating the images here this is bad because is an extra step in the workflow
In [7]:
#imageB = np.rot90(imageB_v, k=3)
#imageA = np.rot90(imageA_v, k=3)
In [8]:
(kpsA, featuresA) = detectAndDescribe(imageA)
(kpsB, featuresB) = detectAndDescribe(imageB)
In [9]:
# match features between the two images

M = matchKeypoints(kpsA, kpsB, featuresA, featuresB, ratio, reprojThresh)


I need to study the parameters ratio and reprojThresh
In [10]:
(matches, H, status) = M
result = cv2.warpPerspective(imageA, H, (imageA.shape[1], imageA.shape[0] + imageB.shape[0]))
result[0:imageB.shape[0], 0:imageB.shape[1]] = imageB
In [11]:
plt.title('My fisrt mosaic');
In [12]:
plt.title('First image');