opencv-常用代码

以下是一些OpenCV中常用的代码片段，涵盖了一些基本的图像处理和计算机视觉任务。

加载和显示图像:

import cv2# 读取图像
img = cv2.imread('image.jpg')# 显示图像
cv2.imshow('Image', img)
cv2.waitKey(0)
cv2.destroyAllWindows()

调整图像大小:

resized_img = cv2.resize(img, (width, height))

灰度转换:

gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

图像平滑:

smoothed_img = cv2.GaussianBlur(img, (kernel_size, kernel_size), 0)

边缘检测:

edges = cv2.Canny(gray_img, threshold1, threshold2)

图像阈值处理:

ret, binary_img = cv2.threshold(gray_img, threshold_value, max_value, cv2.THRESH_BINARY)

图像轮廓检测:

contours, hierarchy = cv2.findContours(edges, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

绘制轮廓:

cv2.drawContours(img, contours, -1, (0, 255, 0), 2)

人脸检测:

face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
faces = face_cascade.detectMultiScale(gray_img, scaleFactor=1.3, minNeighbors=5)

图像旋转:

rows, cols = img.shape[:2]
M = cv2.getRotationMatrix2D((cols / 2, rows / 2), angle, scale)
rotated_img = cv2.warpAffine(img, M, (cols, rows))

图像腐蚀与膨胀:

kernel = np.ones((kernel_size, kernel_size), np.uint8)
eroded_img = cv2.erode(img, kernel, iterations=1)
dilated_img = cv2.dilate(img, kernel, iterations=1)

图像直方图均衡化:
```
equ_img = cv2.equalizeHist(gray_img)
```

图像混合:

blended_img = cv2.addWeighted(img1, alpha, img2, beta, gamma)

图像拼接:

stitched_img = cv2.hconcat([img1, img2])

图像截取:
```
roi = img[y:y+h, x:x+w]
```
图像相加:
```
added_img = cv2.add(img1, img2)
```

图像减法:

subtracted_img = cv2.subtract(img1, img2)

图像位运算:

bitwise_and = cv2.bitwise_and(img1, img2)
bitwise_or = cv2.bitwise_or(img1, img2)
bitwise_xor = cv2.bitwise_xor(img1, img2)
bitwise_not = cv2.bitwise_not(img)

图像平均模糊:

averaged_img = cv2.blur(img, (kernel_size, kernel_size))

中值滤波:

median_blurred_img = cv2.medianBlur(img, ksize)

自适应阈值:

adaptive_thresh = cv2.adaptiveThreshold(gray_img, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 11, 2)

图像旋转缩放:

M = cv2.getRotationMatrix2D((cols / 2, rows / 2), angle, scale)
rotated_resized_img = cv2.warpAffine(img, M, (new_cols, new_rows))

SIFT 特征提取和匹配:

sift = cv2.SIFT_create()
keypoints1, descriptors1 = sift.detectAndCompute(img1, None)
keypoints2, descriptors2 = sift.detectAndCompute(img2, None)

SURF 特征提取和匹配:

surf = cv2.SURF_create()
keypoints1, descriptors1 = surf.detectAndCompute(img1, None)
keypoints2, descriptors2 = surf.detectAndCompute(img2, None)

ORB 特征提取和匹配:

orb = cv2.ORB_create()
keypoints1, descriptors1 = orb.detectAndCompute(img1, None)
keypoints2, descriptors2 = orb.detectAndCompute(img2, None)

使用 FLANN 匹配器进行特征匹配:

FLANN_INDEX_KDTREE = 1
index_params = dict(algorithm=FLANN_INDEX_KDTREE, trees=5)
search_params = dict(checks=50)
flann = cv2.FlannBasedMatcher(index_params, search_params)
matches = flann.knnMatch(descriptors1, descriptors2, k=2)

简单图像深度学习任务:

net = cv2.dnn.readNet('model.weights', 'model.cfg')
blob = cv2.dnn.blobFromImage(img, scalefactor=1.0, size=(300, 300), mean=(104.0, 177.0, 123.0))
net.setInput(blob)
detections = net.forward()

HOG 特征提取与行人检测:

hog = cv2.HOGDescriptor()
hog.setSVMDetector(cv2.HOGDescriptor_getDefaultPeopleDetector())
boxes, weights = hog.detectMultiScale(img, winStride=(8, 8), padding=(0, 0), scale=1.05)