接上篇，本文將繼續分享OpenCV實用技巧，涵蓋圖像處理、目標檢測、3D視覺等進階領域！

六、圖像變換進階

17. 圖像金字塔

# 高斯金字塔下采樣
smaller = cv2.pyrDown(img)# 高斯金字塔上采樣
larger = cv2.pyrUp(img)

用于多尺度圖像處理，構建圖像金字塔

18. 極坐標變換

polar = cv2.warpPolar(img, (300,300), (w//2, h//2), radius, cv2.WARP_POLAR_LINEAR
)

實現直角坐標系與極坐標系轉換

19. 非均勻縮放

M = np.float32([[1, 0.5, 0], [0, 1, 0]])
sheared = cv2.warpAffine(img, M, (int(w*1.5), h))

實現圖像剪切變形效果

20. 網格扭曲效果

map_x = np.zeros(img.shape[:2], np.float32)
map_y = np.zeros(img.shape[:2], np.float32)
# 創建網格扭曲映射...
distorted = cv2.remap(img, map_x, map_y, cv2.INTER_LINEAR)

實現自定義的網格變形效果

七、形態學操作大全

21. 膨脹操作

kernel = np.ones((5,5), np.uint8)
dilated = cv2.dilate(img, kernel)

擴大亮區域，連接斷裂對象

22. 腐蝕操作

eroded = cv2.erode(img, kernel)

縮小亮區域，去除小噪點

23. 開運算

opening = cv2.morphologyEx(img, cv2.MORPH_OPEN, kernel)

先腐蝕后膨脹，去除小物體

24. 閉運算

closing = cv2.morphologyEx(img, cv2.MORPH_CLOSE, kernel)

先膨脹后腐蝕，填充小孔洞

25. 形態學梯度

gradient = cv2.morphologyEx(img, cv2.MORPH_GRADIENT, kernel)

獲取物體輪廓邊緣

26. 頂帽操作

tophat = cv2.morphologyEx(img, cv2.MORPH_TOPHAT, kernel)

提取亮色小特征（如白紙黑字）

27. 黑帽操作

blackhat = cv2.morphologyEx(img, cv2.MORPH_BLACKHAT, kernel)

提取暗色小特征（如黑底白點）

八、圖像濾波與增強

28. 雙邊濾波

bilateral = cv2.bilateralFilter(img, 9, 75, 75)

保持邊緣銳利的同時降噪

29. 自定義濾波核

kernel = np.array([[0,-1,0], [-1,5,-1], [0,-1,0]])
sharpened = cv2.filter2D(img, -1, kernel)

創建銳化濾波核增強細節

30. 導向濾波

# 使用PyTorch實現導向濾波更佳
# 偽代碼：guided_filter = guideFilter(guide, img, radius, eps)

保持邊緣的先進濾波技術

31. 非局部均值去噪

denoised = cv2.fastNlMeansDenoisingColored(img, None, 10, 10, 7, 21)

高效去除高斯噪聲

32. 拉普拉斯銳化

laplacian = cv2.Laplacian(img, cv2.CV_64F)
sharp = cv2.convertScaleAbs(laplacian)

增強圖像邊緣和細節

九、視頻處理技巧

33. 視頻幀讀取

cap = cv2.VideoCapture('video.mp4')
while cap.isOpened():ret, frame = cap.read()if not ret: break# 處理幀
cap.release()

34. 視頻寫入

fourcc = cv2.VideoWriter_fourcc(*'XVID')
out = cv2.VideoWriter('output.avi', fourcc, 20.0, (640,480))
out.write(processed_frame)
out.release()

35. 幀差分法

prev_frame = None
while True:ret, frame = cap.read()gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)if prev_frame is not None:diff = cv2.absdiff(gray, prev_frame)_, motion = cv2.threshold(diff, 30, 255, cv2.THRESH_BINARY)prev_frame = gray

檢測視頻中的運動區域

36. 背景建模

bg_subtractor = cv2.createBackgroundSubtractorMOG2()
fg_mask = bg_subtractor.apply(frame)

分離前景運動物體

十、3D視覺與相機標定

37. 棋盤格角點檢測

pattern_size = (9, 6)
ret, corners = cv2.findChessboardCorners(gray, pattern_size)
if ret:cv2.drawChessboardCorners(img, pattern_size, corners, ret)

相機標定的關鍵步驟

38. 相機標定

# 準備標定數據
obj_points = []  # 3D點
img_points = []  # 2D點# 執行標定
ret, mtx, dist, rvecs, tvecs = cv2.calibrateCamera(obj_points, img_points, gray.shape[::-1], None, None
)

39. 畸變校正

undistorted = cv2.undistort(img, mtx, dist)

40. 距離測量

# 已知物體大小和焦距
focal_length = 500  # 相機焦距（像素）
known_width = 20.0  # 物體實際寬度（cm）
pixel_width = 150   # 圖像中的物體寬度（像素）
distance = (known_width * focal_length) / pixel_width

單目視覺距離測量

十一、目標檢測進階

41. HOG行人檢測

hog = cv2.HOGDescriptor()
hog.setSVMDetector(cv2.HOGDescriptor_getDefaultPeopleDetector())
rects, weights = hog.detectMultiScale(frame, winStride=(4,4), padding=(8,8), scale=1.05)

42. YOLO目標檢測

net = cv2.dnn.readNet("yolov3.weights", "yolov3.cfg")
blob = cv2.dnn.blobFromImage(img, 1/255, (416,416), swapRB=True, crop=False)
net.setInput(blob)
outputs = net.forward(net.getUnconnectedOutLayersNames())

43. 二維碼檢測

detector = cv2.QRCodeDetector()
data, points, _ = detector.detectAndDecode(img)
if points is not None:points = points[0].astype(int)cv2.polylines(img, [points], True, (0,255,0), 3)

十二、圖像分析與測量

44. 輪廓近似

epsilon = 0.01 * cv2.arcLength(cnt, True)
approx = cv2.approxPolyDP(cnt, epsilon, True)
cv2.drawContours(img, [approx], 0, (0,255,0), 2)

45. 最小外接矩形

rect = cv2.minAreaRect(cnt)
box = cv2.boxPoints(rect)
box = np.int0(box)
cv2.drawContours(img, [box], 0, (0,0,255), 2)

46. 最小外接圓

(x,y), radius = cv2.minEnclosingCircle(cnt)
center = (int(x), int(y))
radius = int(radius)
cv2.circle(img, center, radius, (255,0,0), 2)

47. 橢圓擬合

if len(cnt) >= 5:ellipse = cv2.fitEllipse(cnt)cv2.ellipse(img, ellipse, (0,255,0), 2)

十三、高級處理技巧

48. 圖像修復

damaged = img.copy()
mask = np.zeros(img.shape[:2], np.uint8)
# 創建受損區域...
restored = cv2.inpaint(damaged, mask, 3, cv2.INPAINT_TELEA)

49. 高動態范圍成像

# 準備不同曝光度的圖像
images = [img1, img2, img3]  
times = [0.5, 1.0, 2.0]  # 曝光時間# 合并HDR圖像
calibrate = cv2.createCalibrateDebevec()
response = calibrate.process(images, times)
merge = cv2.createMergeDebevec()
hdr = merge.process(images, times, response)

50. 圖像拼接

stitcher = cv2.Stitcher_create()
status, panorama = stitcher.stitch([img1, img2])
if status == cv2.Stitcher_OK:cv2.imshow('Panorama', panorama)

十四、實用工具集錦（51-100）

完整100例速查表

?關注我們并獲取更多【OpenCV100】獲取更多示例