opencv|OpenCV---阈值与平滑处理图像阈值计算机视觉|python|opencv

【OpenCV+TensorFlow】清华博士带你做项目！计算机视觉实战+深度学习，项目可写进简历！（附源码资料）-人工智能/深度学习框架/RNN/池化层/感受_哔哩哔哩_bilibili import cv2# 读取的格式是BGR import matplotlib.pyplot as plt import numpy as npdef cv_show(name, img): cv2.imshow('image', img)# 第一个为窗口名字 cv2.waitKey(0)# 等待时间，毫秒级，0表示任意键终止 cv2.destroyAllWindows()''' 阈值与平滑处理 ret.dst=cv2.threshold(src,thresh,maxval,type) src：输出图，只能输入单通道图像，通常来说为灰度图 dst：输出图 thresh：阈值 maxval：当像素超过了阈值（或者小于阈值，根据type来决定），所赋予的值 type：二值化操作的类型，包含以下5种类型：cv2.THRESH_BINARY超过阈值部分取maxval（最大值），否则取0; cv2.THRESH_BINARY_INY THRESH_BINARY的反转; cv2.THRESH_TRUNC 大于阈值部分设为阈值，否则不变; cv2.THRESH_TOZERO 大于阈值部分不改变，否则设为0; cv2.THRESH_TOZERO_INVTHRESH_TOZERO的反转； '''# flower = cv2.imread('flower.jpg') # # cv_show('flower', flower) # ret, thresh1 = cv2.threshold(flower, 127, 255, cv2.THRESH_BINARY) # ret, thresh2 = cv2.threshold(flower, 127, 255, cv2.THRESH_BINARY_INV) # ret, thresh3 = cv2.threshold(flower, 127, 255, cv2.THRESH_TRUNC) # ret, thresh4 = cv2.threshold(flower, 127, 255, cv2.THRESH_TOZERO) # ret, thresh5 = cv2.threshold(flower, 127, 255, cv2.THRESH_TOZERO_INV) # # titles = ['Original Image', 'BINARY', 'BINARY_INV', 'TRUNC', 'TOZERO', 'TOZERO_INV'] # images = [flower, thresh1, thresh2, thresh3, thresh4, thresh5] # for i in range(6): #plt.subplot(2, 3, i + 1), plt.imshow(images[i], 'gray') #plt.title(titles[i]) #plt.xticks([]), plt.yticks([]) # plt.show()# 图像平滑处理 moon = cv2.imread('timg.jpg') # cv_show('moon', moon)# # 均值滤波 # # 简单的平均卷积操作 # blur = cv2.blur(moon, (3, 3)) # cv_show('blur', blur)# # 方框滤波 # # 基本和均值一样，可以选择归一化 # box = cv2.boxFilter(moon, -1, (3, 3), normalize=True)# normalize为True时和均值滤波一致，而为False时超过255的值设为255 # cv_show('box', box)# # 高斯滤波 # # 高斯模糊的卷积核里的数值是满足高斯分布，相当于更重视中间的 # aussian = cv2.GaussianBlur(moon, (5, 5), 1) # cv_show('aussian', aussian)# # 中值滤波 # # 相当于中值代替 # median = cv2.medianBlur(moon, 5)# 中值滤波 # cv_show('median', median)# 展示所有 blur = cv2.blur(moon, (3, 3)) aussian = cv2.GaussianBlur(moon, (5, 5), 1) median = cv2.medianBlur(moon, 5) res = np.vstack((blur, aussian, median)) cv_show('median vs average', res)