1

img = cv2.imread(img_path)

1

cv2.imwrite(save_path, img)

1

cv2.putText(img, "text test", (startX, startY), cv2.FONT_HERSHEY_SIMPLEX, font_size, (B,G,R), thickness)

1

cv2.rectangle(img,(x,y), (x+h,y+w), (0,255,0), thickness)

1
2
3
4
5

keypress = cv2.waitKey(200) # 200为当前图片的显示持续时间

if keypress == ord('c') # keypress为按键的整数形式, 所以需要用ord将字符类型转换

if cv2.waitKey(200) == 27: # Decimal 27 = Esc

1
2
3
4

import cv2
img = cv2.imread('./test.jpg')
print(type(img)) # <class 'numpy.ndarray'>
print(img.shape) # (500, 1069, 3)  (高, 宽, 通道)

1
2
3
4
5
6
7
8
9
10

# 图片的分辨率为300*200(宽*高)，这里b, g, r设为随机值，注意dtype属性
b = np.random.randint(0, 255, (200, 300), dtype=np.uint8)
g = np.random.randint(0, 255, (200, 300), dtype=np.uint8)
r = np.random.randint(0, 255, (200, 300), dtype=np.uint8)
# 合并通道，形成图片
img = cv2.merge([b, g, r])  # opencv的通道是b在最前,r在最后
# 显示图片
cv2.imshow('test', img)
cv2.waitKey(0)
cv2.destroyWindow('test')

1
2
3
4
5
6
7
8
9
10
11
12

# 图片的分辨率为800*200(宽*高)，这里b, g, r设为随机值，注意dtype属性
b = np.random.randint(0, 255, (200, 800), dtype=np.uint8)
g = np.random.randint(0, 255, (200, 800), dtype=np.uint8)
r = np.random.randint(0, 255, (200, 800), dtype=np.uint8)
# 合并通道，形成图片
img = cv2.merge([b, g, r])  # opencv的通道是b在最前,r在最后
# 显示图片
cv2.imshow('test', img)
cv2.waitKey(0)
cv2.destroyWindow('test')
# 拆分通道, 每个通道都变成了单通道数组
[blue, green, red] = cv2.split(img)

1
2
3
4
5

# 方法一:
rgb_img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# 方法二:
rgb_img = img[:, :, [2, 1, 0]]   # img[h,w,v]
rgb_img = img[:, :, ::-1]

1
2
3
4
5
6
7
8
9
10
11

from PIL import Image
import numpy as np

# PIL to cv2
pil_img = Image.open(img_path)
print(pil_img.size) # (w, h)
np_img = np.array(pil_img)
cv2_img = np_img[:, :, ::-1] # 交换通道

# cv2 to PIL
pil_img = Image.fromarray(cv2_img[:, :, ::-1])

1
2
3
4

b,g,r=cv2.split(img)
img2=cv2.merge([r,g,b])
plt.imshow(img2)
plt.show()

1
2

# 已知子图左上角坐标 (x1, y1), 右下角坐标(x2, y2)
crop_img = img[y1:y2, x1:x2, :]

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

import cv2
image_path = './test.jpg'
src_image = cv2.imread(image_path) # 读取图片

size = src_image.shape  # 获取图片的尺寸, 返回一个元组: (height, width, depth)

copy_image = src_image.copy() # 复制图片

cv2.imwrite('./dst_test.jpg', copy_image) # 保存图片

cv2.imshow('image', src_image) # 显示图片

# 利用下标访问指定像素
for x in range(src_image.shape[0]): # 以行为主, 行数=图片height
  for y in range(src_image.shape[1]):  # 列数 = 图片width
    src_image[x,y] = (255,0,255)   # (blue, green, red)  值越高表示对应颜色越显著, 全0为黑, 全255为白