diff options
Diffstat (limited to 'scan.py')
| -rw-r--r-- | scan.py | 87 |
1 files changed, 48 insertions, 39 deletions
@@ -1,56 +1,65 @@ -from __future__ import print_function - import cv2 +import imutils import numpy as np -MAX_MATCHES = 500 -GOOD_MATCH_PERCENT = 0.15 + +def order_points(pts): + rect = np.zeros((4, 2), dtype="float32") + s = pts.sum(axis=1) + rect[0] = pts[np.argmin(s)] + rect[2] = pts[np.argmax(s)] + diff = np.diff(pts, axis=1) + rect[1] = pts[np.argmin(diff)] + rect[3] = pts[np.argmax(diff)] + return rect -def align_images(im1, im2): - im1_gra = cv2.cvtColor(im1, cv2.COLOR_BGR2GRAY) - im2_gray = cv2.cvtColor(im2, cv2.COLOR_BGR2GRAY) +def four_point_transform(image, pts): + rect = order_points(pts) + (tl, tr, br, bl) = rect + widthA = np.sqrt(((br[0] - bl[0]) ** 2) + ((br[1] - bl[1]) ** 2)) + widthB = np.sqrt(((tr[0] - tl[0]) ** 2) + ((tr[1] - tl[1]) ** 2)) + maxWidth = max(int(widthA), int(widthB)) - orb = cv2.ORB_create(MAX_MATCHES) - keypoints1, descriptors1 = orb.detectAndCompute(im1_gra, None) - keypoints2, descriptors2 = orb.detectAndCompute(im2_gray, None) + heightA = np.sqrt(((tr[0] - br[0]) ** 2) + ((tr[1] - br[1]) ** 2)) + heightB = np.sqrt(((tl[0] - bl[0]) ** 2) + ((tl[1] - bl[1]) ** 2)) + maxHeight = max(int(heightA), int(heightB)) - matcher = cv2.DescriptorMatcher_create(cv2.DESCRIPTOR_MATCHER_BRUTEFORCE_HAMMING) - matches = matcher.match(descriptors1, descriptors2, None) + dst = np.array([ + [0, 0], + [maxWidth - 1, 0], + [maxWidth - 1, maxHeight - 1], + [0, maxHeight - 1]], dtype="float32") - matches.sort(key=lambda x: x.distance, reverse=False) + M = cv2.getPerspectiveTransform(rect, dst) + warped = cv2.warpPerspective(image, M, (maxWidth, maxHeight)) - num_good_matches = int(len(matches) * GOOD_MATCH_PERCENT) - matches = matches[:num_good_matches] + return warped - im_matches = cv2.drawMatches(im1, keypoints1, im2, keypoints2, matches, None) - cv2.imwrite("matches.jpg", im_matches) - points1 = np.zeros((len(matches), 2), dtype=np.float32) - points2 = np.zeros((len(matches), 2), dtype=np.float32) +image = cv2.imread("example.jpg") +ratio = image.shape[0] / 500.0 +orig = image.copy() - for i, match in enumerate(matches): - points1[i, :] = keypoints1[match.queryIdx].pt - points2[i, :] = keypoints2[match.trainIdx].pt +gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) +gray = cv2.GaussianBlur(gray, (5, 5), 0) +edged = cv2.Canny(gray, 75, 200) - h, mask = cv2.findHomography(points1, points2, cv2.RANSAC) +cnts = cv2.findContours(edged.copy(), cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE) +cnts = imutils.grab_contours(cnts) +cnts = sorted(cnts, key=cv2.contourArea, reverse=True)[:5] - height, width, channels = im2.shape - im1_reg = cv2.warpPerspective(im1, h, (width, height)) +for c in cnts: + peri = cv2.arcLength(c, True) + approx = cv2.approxPolyDP(c, 0.02 * peri, True) + if len(approx) == 4: + screenCnt = approx + break - return im1_reg, h +cv2.drawContours(image, [screenCnt], -1, (0, 255, 0), 2) +warped = four_point_transform(orig, screenCnt.reshape(4, 2) * ratio) -if __name__ == '__main__': - ref_filename = "mask.jpg" - print("Reading reference image : ", ref_filename) - im_reference = cv2.imread(ref_filename, cv2.IMREAD_COLOR) - im_filename = "example.jpg" - print("Reading image to align : ", im_filename) - im = cv2.imread(im_filename, cv2.IMREAD_COLOR) - print("Aligning images ...") - im_req, h = align_images(im, im_reference) - out_filename = "aligned.jpg" - print("Saving aligned image : ", out_filename) - cv2.imwrite(out_filename, im_req) - print("Estimated homography : \n", h) +cv2.imshow("Original", imutils.resize(orig, height=650)) +cv2.imshow("Scanned", imutils.resize(warped, height=650)) +cv2.waitKey(0) |