Feature Matching

import cv2

img1 = cv2.imread("book.jpg", cv2.IMREAD_GRAYSCALE)
img2 = cv2.imread("shelf.jpg", cv2.IMREAD_GRAYSCALE)

orb = cv2.ORB_create(1000)
kp1, d1 = orb.detectAndCompute(img1, None)
kp2, d2 = orb.detectAndCompute(img2, None)

bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)
matches = bf.match(d1, d2)
matches = sorted(matches, key=lambda m: m.distance)[:30]

out = cv2.drawMatches(img1, kp1, img2, kp2, matches, None,
                      flags=cv2.DrawMatchesFlags_NOT_DRAW_SINGLE_POINTS)
cv2.imwrite("matches.jpg", out)

import cv2

sift = cv2.SIFT_create()
kp1, d1 = sift.detectAndCompute(img1, None)
kp2, d2 = sift.detectAndCompute(img2, None)

index_params  = dict(algorithm=1, trees=5)   # KDTree
search_params = dict(checks=50)
flann = cv2.FlannBasedMatcher(index_params, search_params)

matches = flann.knnMatch(d1, d2, k=2)

# Lowe's ratio test
good = [m for m,n in matches if m.distance < 0.75 * n.distance]

import cv2, numpy as np

if len(good) > 10:
    src = np.float32([kp1[m.queryIdx].pt for m in good]).reshape(-1,1,2)
    dst = np.float32([kp2[m.trainIdx].pt for m in good]).reshape(-1,1,2)
    H, mask = cv2.findHomography(src, dst, cv2.RANSAC, 5.0)

    h, w = img1.shape
    corners = np.float32([[0,0],[0,h],[w,h],[w,0]]).reshape(-1,1,2)
    projected = cv2.perspectiveTransform(corners, H)
    img2_box = cv2.polylines(img2.copy(), [np.int32(projected)], True, 255, 3)

import cv2

imgs = [cv2.imread(p) for p in ["left.jpg","mid.jpg","right.jpg"]]
stitcher = cv2.Stitcher_create()
status, pano = stitcher.stitch(imgs)
if status == cv2.Stitcher_OK:
    cv2.imwrite("panorama.jpg", pano)