week1 基本图像处理

a=[1,2,3] print("Hello friends，welcome week1's class .........",29*"-","%s"%(a))

环境配置

!pip install matplotlib -i https://pypi.tuna.tsinghua.edu.cn/simple import cv2 import numpy as np import matplotlib

matplotlib 使用例子

https://www.cnblogs.com/yanghailin/p/11611333.html, import matplotlib.pyplot as plt i=0 print(image_data[i%10]) plt.imshow(image_data[i%10],cmap = 'gray') i=i+1 plt.imshow(cv2.imread('lena.jpg')) img = cv2.imread("lena.jpg") img= cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

直接用生成矩阵的方式生成图片

img0 = np.array([[0,0,1],[0,1,0],[1,0,0]]) #img0 = np.array([[0,0,1],[0,1,0],[1,0,0]])

查看矩阵数值以及大小

print(img0) print(img0.shape) print("img0 size = %s,%s"%(img0.shape[0],img0.shape[1])) import matplotlib.pyplot as plt import matplotlib.pylab as plt plt.imshow(img0,cmap = 'gray' )

color map

plt.imshow(img0) #pycharm要加一句：plt.show() 加一个%matplotlib inline就会显示 ## 从摄像头采集图像 # read camera cap = cv2.VideoCapture(0) # read video #cap = cv2.VideoCapture("/Users/zhaomingming/Documents/HTC/核心课/CVFundamentals/week1/How Computer Vision Works.mp4") #cap = cv2.VideoCapture("../How Computer Vision Works.mp4") cap = cv2.VideoCapture("How Computer Vision Works.mp4") print(cap.isOpened()) return_value=True if 1: while return_value: return_value,frame = cap.read() print(cap.isOpened()) # frame 图像帧 #cv2.cvtColor() #cv2.COLOR_BGR2RGB plt.imshow(frame,cmap = 'gray') #plt.imshow(cv2.cvtColor(frame,cv2.COLOR_BGR2RGB)) ### frame 的大小是多少？ print(frame.shape) # 关闭capture cap.release() ## 从文件读取图像数据 img = cv2.imread("lena.jpg") #cv2.cvtColor() #cv2.COLOR_BGR2RGB img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) print(img.shape) # range of instrest roi = img[100:200,300:400] plt.imshow(img) #plt.imshow(roi) ## 图像颜色空间变换 # 黑白图像 img_gray=cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) img_gray.shape plt.imshow(img_gray,cmap='gray') ### 彩色图像的颜色空间转换 # https://blog.csdn.net/zhang_cherry/article/details/88951259 #### rgb2hsv import cv2 img_BGR = cv2.imread('lena.jpg') img_hsv=cv2.cvtColor(img_BGR,cv2.COLOR_BGR2HSV) plt.imshow(img_hsv,cmap='') #### 二值化 # 二值化 import cv2 img = cv2.imread('lena.jpg') plt.imshow(cv2.threshold(img,128,200,cv2.THRESH_BINARY)) ## 图像的刚体变换 ### 图像的放大与缩小 # 列，行 img =cv2.resize(img,(50,30)) plt.imshow(img) img.shape # 列，行 img = cv2.imread('lena.jpg') img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) img =cv2.resize(img,(500,300)) plt.imshow(img) img.shape ### 获取感兴趣区域 # roi img_roi =img[100:300,0:200] plt.imshow(img_roi) import numpy as np # 移动： M = np.float32([[1,0,300],[0,1,200]]) print(img[0+50,0]) img_1=cv2.warpAffine(img,M,(1000,1000)) plt.imshow(img_1) print(img_1[200+50,300]) ### 图像的旋转变换与拉伸变换 #pts1 = np.float32([[50,50],[200,50],[50,200]]) #pts2 = np.float32([[10,100],[200,50],[100,250]]) #M = cv2.getAffineTransform(pts1,pts2) #print(M) #180/3.14 theta=0.5 #M = np.float32([[np.cos(theta),-np.sin(theta),100],[np.sin(theta),np.cos(theta),200]]) M = np.float32([[0.1,0,100],[0,2,100]]) # 变换矩阵，平移，斜切，旋转 # affine cols=800 rows=800 dst = cv2.warpAffine(img,M,(cols,rows)) plt.imshow(dst) pts1 = np.float32([[56,65],[368,52],[28,387],[389,390]]) pts2 = np.float32([[0,0],[100,0],[0,300],[300,300]]) M = cv2.getPerspectiveTransform(pts1,pts2) print(M) # 拉伸变换后者透视变换 dst = cv2.warpPerspective(img,M,(300,300)) plt.imshow(dst) ## 图像模糊与锐化 img= cv2.GaussianBlur(img, (11, 11), 1, 0) # 高斯模糊 plt.imshow(img) cv2.Canny(pil_img3,30,150) ### 图像滤波/卷积 kernel = np.ones((3,3),np.float32)/8 kernel=-kernel kernel[0,:]=[-1,-1,-1] kernel[1,:]=[0,0,0] kernel[2,:]=[1,1,1] print(kernel) plt.imshow(img) #dst=cv.filter2D(src, ddepth, kernel[, dst[, anchor[, delta[, borderType]]]])；当ddepth=-1时，表示输出图像与原图像有相同的深度。 print(img.shape) result = cv2.filter2D(img,-1,kernel) result.shape print(result[0,0]) plt.imshow(result*255) result = cv2.filter2D(result,-1,kernel) plt.imshow(result) result.shape ## add watermask wm = cv2.imread("water1.png") wm = cv2.resize(wm,(300,300)) wm = 255-wm img1 = cv2.resize(img,(300,300)) #img1 = cv2.cvtColor(img1, cv2.COLOR_BGR2RGB) print(wm.shape) plt.imshow(cv2.add(wm,img1)) plt.imshow(cv2.addWeighted(wm,0.9,img1,0.5,0)) plt.imshow(wm)