https://blog.csdn.net/weixin_43151193/article/details/125222481特征点的检测与匹配（ORB,SIFT,SURFT比较）
本文旨在总结opencv-python上特征点的检测和匹配。
1、特征点的检测（包括：ORB,SIFT,SURFT）
2、特侦点匹配方法 (包括：暴力法，FLANN，以及随机抽样一致性优化RANSAC算法)
注：由于SURF专利问题，所以opencv官方包目前不支持SURF但支持ORB和SIFT，安装opencv-contrib-python包就可以解决
pip uninstall opencv-python
pip install opencv-contrib-python==3.4.2.17
亲测效果：
import numpy as np
import cv2
from matplotlib import pyplot as plt
def SIFT(img):
# SIFT算法关键点检测
# 读取图像
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# SIFT关键点检测
# 1. 实例化sift
sift = cv2.xfeatures2d.SIFT_create()
# 2. 利用sift.detectAndCompute()检测关键点并计算
kp, des = sift.detectAndCompute(gray, None)
# gray: 进行关键带你检测的图像，注意是灰度图像
# kp: 关键点信息，包括位置，尺度，方向信息
# des: 关键点描述符，每个关键点对应128个梯度信息的特征向量
# 3. 将关键点检测结果绘制在图像上
# cv2.drawKeypoints(image, keypoints, outputimage, color, flags)
# image: 原始图像
# keypoints: 关键点信息，将其绘制在图像上
# outputimage: 输出图片，可以是原始图像
# color: 颜色设置，通过修改(b, g, r)的值，更改画笔的颜色，b = 蓝色, g = 绿色, r = 红色
# flags: 绘图功能的标识设置
# 1. cv2.DRAW_MATCHES_FLAGS_DEFAULT: 创建输出图像矩阵，使用现存的输出图像绘制匹配对象和特征点，对每一个关键点只绘制中间点
# 2. cv2.DRAW_MATCHES_FLAGS_DRAW_OVER_OUTIMG: 不创建输出图像矩阵，而是在输出图像上绘制匹配对
# 3. cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS: 对每一个特征点绘制带大小和方向的关键点图形
# 4. cv2.DRAW_MATCHES_FLAGS_NOT_DRAW_SINGLE_POINTS: 单点的特征点不被绘制
cv2.drawKeypoints(img, kp, img, (0, 255, 0))
cv2.imwrite("1.jpg", img)
# 图像显示
plt.figure(figsize=(10, 8), dpi=100)
plt.imshow(img[:, :, ::-1])
plt.xticks([]), plt.yticks([])
plt.show()
return kp, des
def ByFlann(img1, img2, kp1, kp2, des1, des2, flag="ORB"):
"""
（1）FLANN匹配器
:param img1: 匹配图像1
:param img2:
匹配图像2
:param kp1: 匹配图像1的特征点
:param kp2: 匹配图像2的特征点
:param des1: 匹配图像1的描述子
:param des2: 匹配图像2的描述子
:return:
"""
if (flag == "SIFT" or flag == "sift"):
# SIFT方法
FLANN_INDEX_KDTREE = 1
index_params = dict(algorithm=FLANN_INDEX_KDTREE,
trees=
5)
search_params = dict(check=50)
else:
# ORB方法
FLANN_INDEX_LSH = 6
index_params = dict(algorithm=FLANN_INDEX_LSH,
table_number=6,
key_size=12,
multi_probe_level=1)
search_params = dict(check=50)
# 定义FLANN参数
flann = cv2.FlannBasedMatcher(index_params, search_params)
matches = flann.match(des1, des2)
img3 = cv2.drawMatches(img1, kp1, img2, kp2, matches, None, flags=cv2.DrawMatchesFlags_NOT_DRAW_SINGLE_POINTS)
dim = (800, 400)
# 设定新的尺寸
resized = cv2.resize(img3, dim, interpolation=cv2.INTER_AREA)
#cv2.imshow("Matches", resized)
#cv2.imwrite("1.jpg", img3)
cv2.waitKey(0)
return matches
def RANSAC(img1, img2, kp1, kp2, matches):
MIN_MATCH_COUNT = 10
# store all the good matches as per Lowe's ratio test.
matchType = type(matches[0])
good = []
print(matchType)
if isinstance(matches[0], cv2.DMatch):
# 搜索使用的是match
good = matches
else:
# 搜索使用的是knnMatch
for m, n in matches:
if m.distance
0.7 * n.distance:
good.append(m)
if len(good) > MIN_MATCH_COUNT:
src_pts = np.float32([kp1[m.queryIdx].pt for m in good]).reshape(-1, 1, 2)
dst_pts = np.float32([kp2[m.trainIdx].pt for m in good]).reshape(-1, 1, 2)
# M: 3x3 变换矩阵.
M, mask = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC, 5.0)
matchesMask = mask.ravel().tolist()
# 计算拼接后图像的尺寸
H=M
h1, w1 = img1.shape[:2]
h2, w2 = img2.shape[:2]
corners1 = np.float32([[0, 0], [0, h1], [w1, h1], [w1, 0]]).reshape(-1, 1, 2
)
corners2 = np.float32([[0, 0], [0, h2], [w2, h2], [w2, 0]]).reshape(-1, 1, 2)
transformed_corners = cv2.perspectiveTransform(corners1, H)
# 计算拼接图的宽高
all_corners = np.concatenate((corners2, transformed_corners), axis=0)
x_min, y_min = np.int32(all_corners.min(axis=0).ravel() - 0.5)
x_max, y_max = np.int32(all_corners.max(axis=0).ravel() + 0.5)
output_width = x_max - x_min
output_height = y_max - y_min
# 平移变换矩阵，使图像完整显示
T = np.array([[1, 0, -x_min], [0, 1, -y_min], [0, 0, 1]])
result = cv2.warpPerspective(img1, T.dot(H), (output_width, output_height))
result[-y_min:h1 - y_min, -x_min:w1 - x_min] = img2
dim = (1600, 800) # 设定新的尺寸
resized = cv2.resize(result, dim, interpolation=cv2.INTER_AREA)
cv2.imshow('result', resized)
# h, w = img1.shape
# pts = np.float32([[0, 0], [0, h - 1], [w - 1, h - 1], [w - 1, 0]]).reshape(-1, 1, 2)
# dst = cv2.perspectiveTransform(pts, M)
#
# img2 = cv2.polylines(img2, [np.int32(dst)], True, 255, 3, cv2.LINE_AA)
else:
print
"Not enough matches are found - %d/%d" % (len(good), MIN_MATCH_COUNT)
matchesMask = None
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, None, **draw_params)
draw_params1 = dict(matchColor=(0, 255, 0), # draw matches in green color
singlePointColor=None,
matchesMask=None, # draw only inliers
flags=2)
img33 = cv2.drawMatches(img1, kp1, img2, kp2, good, None, **draw_params1)
cv2.imwrite("1.jpg", img33)
cv2.imwrite("2.jpg", img3)
dim = (1600, 800) # 设定新的尺寸
resized = cv2.resize(img33, dim, interpolation=cv2.INTER_AREA)
#cv2.imshow("before", resized)
cv2.imwrite("match.jpg", img3)
resized = cv2.resize(img3, dim, interpolation=cv2.INTER_AREA)
cv2.imshow(
"now", resized)
cv2.waitKey(0)
img1 = cv2.imread("./1.png")
img2 = cv2.imread("./2.png")
kp1, des1 = SIFT(img1)
kp2, des2 = SIFT(img2)
img1 = cv2.imread("./1.png")
img2 = cv2.imread("./2.png")
matches = ByFlann(img1, img2, kp1, kp2, des1, des2, "SIFT")
RANSAC(img1, img2, kp1, kp2, matches)
