day1：python图像处理

看的b站这个博主的视频https://b23.tv/FPm8Y5m学习

1，python图像处理，提取图像rgb通道并显示，灰度图像显示，剪切图像并显示

import cv2

# #打印cv2的版本号
# print(cv2.getVersionString())

image = cv2.imread("F:/enhancement/Test/Fusion/5.bmp")
# #图片大小
# print(image.shape)

# #提取RGB通道并显示
# cv2.imshow("b", image[:, :, 0])
# cv2.imshow("g", image[:, :, 1])
# cv2.imshow("r", image[:, :, 2])
# cv2.waitKey()

# #灰度图像
# gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# cv2.imshow("gray", gray)
# cv2.waitKey()

# # crop
crop = image[100:500, 300:800]
cv2.imshow("crop", crop)
cv2.waitKey()

2.绘制

import cv2
import numpy as np

# #创建画布
image = np.zeros([300, 300, 3], dtype=np.uint8)

# #线段 起始坐标，终点坐标，线的颜色，线段粗细
cv2.line(image, (100, 200), (250, 250), (255, 0, 0), 2)
# #矩形 对角坐标
cv2.rectangle(image, (30, 100), (40, 150), (0, 255, 0), 2)
# #圆形 圆心，半径，颜色，粗细
cv2.circle(image, (150, 150), 20, (0, 0, 255), 3)
# #字符串，字符串内容，坐标，字体序号（0：默认），缩放系数，颜色，粗细，线条类型（1：实线）
cv2.putText(image, "hello", (100, 50), 0, 2, (255, 255, 255), 2, 1)
cv2.imshow("image", image)
cv2.waitKey()

结果

3.图像的特征点提取

import cv2

image = cv2.imread("F:/image enhancement/python/Low-light-Image-Enhancement-master/demo/2.bmp")
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

corners = cv2.goodFeaturesToTrack(gray, 500, 0.1, 10)
for corner in corners:
    x, y = corner.ravel()
    cv2.circle(image, (int(x), int(y)), 3, (255, 0, 255), -1)

cv2.imshow("corners", image)
cv2.waitKey()

结果

4.图像的梯度、canny边缘检测

import cv2

# 灰度图像方法1
gray = cv2.imread("F:/image enhancement/images/9.bmp", cv2.IMREAD_GRAYSCALE)

# #灰度图像方法2
# image = cv2.imread("F:/image enhancement/images/9.bmp")
# gray2 = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

# 拉普拉斯算子梯度算法
laplacian = cv2.Laplacian(gray, cv2.CV_64F)
# canny边缘提取（梯度区间，大于50是边缘，小于10不是边缘，之间的待定）
canny = cv2.Canny(gray, 10, 50)

cv2.imshow("gray", gray)
# cv2.imshow("gray2", gray2)
cv2.imshow("laplacian", laplacian)
cv2.imshow("canny", canny)
cv2.waitKey()

结果

5图像的二值化

import cv2

gray = cv2.imread("F:/image enhancement/images/9.bmp", cv2.IMREAD_GRAYSCALE)
# 固定阈值，阈值为10， 最大灰度值255
ret, binary = cv2.threshold(gray, 20, 255, cv2.THRESH_BINARY)
# 自适应阈值
binary_adaptive = cv2.adaptiveThreshold(
    gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 115, 1)
# otsu算法
ret1, binary_otsu = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
cv2.imshow("gray", gray)
cv2.imshow("binary", binary)
cv2.imshow("adaptive", binary_adaptive)
cv2.imshow("otsu", binary_otsu)

cv2.waitKey()

结果

6，对图像进行膨胀腐蚀操作

import cv2
import numpy as np

gray = cv2.imread("F:/image enhancement/images/3.bmp", cv2.IMREAD_GRAYSCALE)

# 反向阈值，把背景变为黑色，物体变成白色
_, binary = cv2.threshold(gray, 20, 255, cv2.THRESH_BINARY_INV)
# 内核 5*5像素的正方形
kernel = np.ones((5, 5), np.uint8)
# 腐蚀
erosion = cv2.erode(binary, kernel)
# 膨胀
dilation = cv2.dilate(binary, kernel)

cv2.imshow("gray", gray)
cv2.imshow("binary", binary)
cv2.imshow("erode", erosion)
cv2.imshow("dilate", dilation)
cv2.waitKey()

结果