1.ROI感興趣區域的操作

尋找感興趣的區域主要就是利用矩陣的切片功能來提取.
如face = image[100:200, 300:400]

import cv2 as cv
image = cv.imread('./data/lena.jpg', 1)
cv.imshow('source image', image)
# 提取感興趣區域
face = image[250:400, 200:350]
# 將感興趣區域轉換成灰度圖
gray = cv.cvtColor(face, cv.COLOR_BGR2GRAY)
# 將灰度圖轉換成BGR格式為了後邊賦值操作保持通道一致
# 這裡灰度轉換並不會變成彩色一定要注意
 

backface = cv.cvtColor(gray, cv.COLOR_GRAY2BGR)
# 重新賦值
image[250:400, 200:350] = backface
cv.imshow('face_part', image)
cv.waitKey(0)
cv.destroyAllWindows()

2.泛洪填充

泛洪填充,如何填充一個物件內部區域

• FLOODFILL_FIXED_RANGE- 改變影象,泛洪填充
• FLOODFILL_MASK_ONLY - 不改變影象,只填充遮罩層本身,忽略新的顏色值引數
• floodFill(Mat image, Mat mask, Point seedPoint, Scalar newVal)
• floodFill(image, mask, seedPoint, newVal, rect, loDiff, upDiff, flags)
  src(x,y)=[src(seed.x, seed,y)-loDiff, src(seed.x, seed,y)+upDiff]

import cv2 as cv
import  numpy as np

# 彩色影象的填充
def fill_color_demo(image):
    copy_image = image.copy()
    h, w = image.shape[:2]
    mask = np.zeros([
 
h+2, w+2], np.uint8)
    cv.floodFill(copy_image, mask, (30, 30), (0, 255, 255),
                 (100, 100, 100), (50, 50, 50), cv.FLOODFILL_FIXED_RANGE)
    cv.imshow("flood image demo", copy_image)

# 灰度圖填充
# 灰度圖填充
def fill_binary():
    img = np.zeros([400, 400, 3], np.uint8)
    img[100:300, 100:300, :] = 255
    cv.imshow("fill_binary", img)
    mask = np.ones([402, 402, 1], np.uint8)
    mask[101:301, 101:301] = 0
    cv.floodFill(img, mask, (200, 200),(0, 0, 255), cv.FLOODFILL_MASK_ONLY)
    cv.imshow("filled binary", img)

image = cv.imread('./data/lena.jpg', 1)
cv.imshow('source image', image)
fill_color_demo(image)
fill_binary()
cv.waitKey(0)
cv.destroyAllWindows()

後續:
遞迴演算法和掃描非遞迴演算法,後者更快
泛洪填充中的影象與掩碼做與操作