CODEWITHSUNDEEP
pythonopencvimage-processingsiftorb
Ali110
Ali110

Reputation: 161

How to draw bounding box on best matches?

How can I draw a bounding box on best matches in BF MATCHER using Python?

Upvotes: 13

Views: 11609

Answers (1)

Howard GENG
Howard GENG

Reputation: 1105

Here is a summary of the approach it should be a proper solution:

  1. Detect keypoints and descriptors on the query image (img1)
  2. Detect keypoints and descriptors on the target image (img2)
  3. Find the matches or correspondences between the two sets of descriptors
  4. Use the best 10 matches to form a transformation matrix
  5. Transform the rectangle around img1 based on the transformation matrix
  6. Add offset to put the bounding box at the correct position
  7. Display the result image (as below).

BF Matching Result Image (Preview)

Here is the code:

import numpy as np
import cv2
from matplotlib import pyplot as plt

img1 = cv2.imread('box.png', 0)          # query Image
img2 = cv2.imread('box_in_scene.png',0)  # target Image

# Initiate SIFT detector
orb = cv2.ORB_create()

# find the keypoints and descriptors with ORB
kp1, des1 = orb.detectAndCompute(img1,None)
kp2, des2 = orb.detectAndCompute(img2,None)

# create BFMatcher object
bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)

# Match descriptors.
matches = bf.match(des1,des2)

# Sort them in the order of their distance.
matches = sorted(matches, key = lambda x:x.distance)

good_matches = matches[:10]

src_pts = np.float32([ kp1[m.queryIdx].pt for m in good_matches     ]).reshape(-1,1,2)
dst_pts = np.float32([ kp2[m.trainIdx].pt for m in good_matches ]).reshape(-1,1,2)
M, mask = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC,5.0)
matchesMask = mask.ravel().tolist()
h,w = img1.shape[:2]
pts = np.float32([ [0,0],[0,h-1],[w-1,h-1],[w-1,0] ]).reshape(-1,1,2)

dst = cv2.perspectiveTransform(pts,M)
dst += (w, 0)  # adding offset

draw_params = dict(matchColor = (0,255,0), # draw matches in green color
               singlePointColor = None,
               matchesMask = matchesMask, # draw only inliers
               flags = 2)

img3 = cv2.drawMatches(img1,kp1,img2,kp2,good_matches, None,**draw_params)

# Draw bounding box in Red
img3 = cv2.polylines(img3, [np.int32(dst)], True, (0,0,255),3, cv2.LINE_AA)

cv2.imshow("result", img3)
cv2.waitKey()
# or another option for display output
#plt.imshow(img3, 'result'), plt.show()

Upvotes: 25

Related Questions