一、效果

二、程式碼實現

1、python程式碼，基於opencv庫和imutils庫

來源：https://blog.csdn.net/qq_34199383/article/details/79571318

from imutils import perspective
from skimage.filters import threshold_local
import cv2
import imutils
 
# 邊緣掃描
image = cv2.imread("./picture/5.png")
ratio = image.shape[0] / 500.0                                 # 比例
orig = image.copy()
image = imutils.resize(image, height = 500)
 
# 灰度轉換及邊緣查詢
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (5, 5), 0)
edged = cv2.Canny(gray, 75, 200)                               # 邊緣檢測
 
# 只保留輪廓
cnts = cv2.findContours(edged.copy(), cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)#通過邊緣影象找到輪廓
cnts = cnts[0] if imutils.is_cv2() else cnts[1]                # 用以區分OpenCV2.4和OpenCV3
cnts = sorted(cnts, key = cv2.contourArea, reverse = True)[:5] # 保留最大輪廓
 
for c in cnts:
    peri = cv2.arcLength(c, True)
    approx = cv2.approxPolyDP(c, 0.02 * peri, True)            # 輪廓點    
    if len(approx) == 4:                                       # 表明找到四個輪廓點
        screenCnt = approx
        break
    
# 轉為鳥瞰圖
warped = perspective.four_point_transform(orig, screenCnt.reshape(4, 2) * ratio)
warped = cv2.cvtColor(warped, cv2.COLOR_BGR2GRAY)              # 灰度轉換
T = threshold_local(warped, 11, offset = 10, method = "gaussian")
warped = (warped > T).astype("uint8") * 255
 
cv2.imshow("Original", imutils.resize(orig, height = 650))
cv2.imshow("Scanned", imutils.resize(warped, height = 650))
cv2.waitKey(0)
 

2、C++實現

基於opencv的getPerspectiveTransform()函式和warpPerspective()函式

程式碼：https://blog.csdn.net/t6_17/article/details/78729097

效果：

3、Image perspective transformation and text recognition

功能：

•  Image cropping automation
•  Bird's eye view transformation
• Optical character recognition

Github：https://github.com/linghugoogle/cv-birdview