something fantastic in life

CV/Background Replacement.py

@@ -0,0 +1,154 @@
+#!/usr/bin/env python3
+# _*_coding:utf-8_*_
+# @Author : Xnhyacinth
+# @Time : 2022/3/8 17:38
+import matplotlib.pyplot as plt
+import numpy as np
+import cv2
+
+# Read in the image
+image = cv2.imread('img/1.jpg')
+
+# Print out the type of image data and its dimensions (height, width, and color)
+print('This image is:', type(image),
+      ' with dimensions:', image.shape)
+
+
+'''RGB'''
+# RGB
+# Make a copy of the image
+image_copy = np.copy(image)
+
+# OpenCV 会读取 BGR 格式（而不是 RGB 格式）的图像，因为一开始开发 OpenCV 的时候，BGR 颜色格式对相机制造商和图像软件提供商来说很受欢迎。
+# 红色通道被认为是最不重要的颜色通道，因此列在最后面。但是，现在标准改变了，很多图像软件和相机都使用 RGB 格式
+
+# Change color to RGB (from BGR)
+image_copy_rgb = cv2.cvtColor(image_copy, cv2.COLOR_BGR2RGB)
+
+# Display the image copy
+plt.imshow(image_copy_rgb)
+
+# 定义颜色阈值
+lower_blue = np.array([0, 0, 250])
+upper_blue = np.array([250, 250, 255])
+
+# 创建遮罩
+mask = cv2.inRange(image_copy_rgb, lower_blue, upper_blue)
+
+# Vizualize the mask
+plt.imshow(mask, cmap='gray')
+
+masked_image_rgb = np.copy(image_copy_rgb)
+
+masked_image_rgb[mask != 0] = [0, 0, 0]
+
+# Display it!
+plt.imshow(masked_image_rgb)
+# Load in a background image, and convert it to RGB
+background_image = cv2.imread('img/10.jpeg')
+# print(background_image.shape)
+background_image_rgb = cv2.cvtColor(background_image, cv2.COLOR_BGR2RGB)
+
+# Crop it to the right size (1482x988)
+crop_background = background_image_rgb[0:1482, 0:988]
+
+crop_background[mask == 0] = [0, 0, 0]
+
+# Display the background
+plt.imshow(crop_background)
+
+# Add the two images together to create a complete image!
+complete_image = masked_image_rgb + crop_background
+
+# Display the result
+plt.imshow(complete_image)
+
+
+
+'''HSV'''
+# Change color to HSV (from BGR)
+image_copy_hsv = cv2.cvtColor(image_copy, cv2.COLOR_BGR2HSV)
+
+# Display the image copy
+plt.imshow(image_copy_hsv)
+
+#提取图片的蓝色
+lower_hsv=np.array([100,43,46])
+upper_hsv=np.array([124,255,255])
+mask=cv2.inRange(image_copy_hsv,lowerb=lower_hsv,upperb=upper_hsv)
+plt.imshow(mask, cmap='gray')
+# cv2.imshow("mask",mask)
+
+masked_image_hsv = np.copy(image_copy_hsv)
+
+masked_image_hsv[mask != 0] = [0, 0, 0]
+
+# Display it!
+plt.imshow(masked_image_hsv)
+
+# convert image it to HSV
+
+# print(background_image.shape)
+background_image_hsv = cv2.cvtColor(background_image, cv2.COLOR_BGR2HSV)
+
+# Crop it to the right size (1482x988)
+crop_background = background_image_hsv[0:1482, 0:988]
+
+crop_background[mask == 0] = [0, 0, 0]
+plt.imshow(crop_background)
+
+# Add the two images together to create a complete image!
+complete_image_hsv = masked_image_hsv + crop_background
+
+# Display the result
+plt.imshow(complete_image_hsv)
+
+# Change color to RGB (from HSV)
+complete_image_hsv = cv2.cvtColor(complete_image_hsv, cv2.COLOR_HSV2RGB)
+# Display the result
+plt.imshow(complete_image_hsv)
+
+
+
+'''HSL'''
+# Change color to HLS (from BGR)
+image_copy_hls = cv2.cvtColor(image_copy, cv2.COLOR_BGR2HLS)
+
+# Display the image copy
+plt.imshow(image_copy_hls)
+
+#提取图片的蓝色
+lower_hls=np.array([100,43,46])
+upper_hls=np.array([124,255,255])
+mask=cv2.inRange(image_copy_hls,lowerb=lower_hls,upperb=upper_hls)
+plt.imshow(mask, cmap='gray')
+
+masked_image_hls = np.copy(image_copy_hls)
+
+masked_image_hls[mask != 0] = [0, 0, 0]
+
+# Display it!
+plt.imshow(masked_image_hls)
+
+# convert image it to HLS
+
+# print(background_image.shape)
+background_image_hls = cv2.cvtColor(background_image, cv2.COLOR_BGR2HLS)
+
+# Crop it to the right size (1482x988)
+crop_background = background_image_hls[0:1482, 0:988]
+
+crop_background[mask == 0] = [0, 0, 0]
+plt.imshow(crop_background)
+
+
+# Add the two images together to create a complete image!
+complete_image_hls = masked_image_hls + crop_background
+
+# Display the result
+plt.imshow(complete_image_hls)
+
+# Change color to RGB (from HSV)
+complete_image_hls = cv2.cvtColor(complete_image_hls, cv2.COLOR_HLS2RGB)
+# Display the result
+plt.imshow(complete_image_hls)

CV/Gaussian Filter.py

@@ -0,0 +1,95 @@
+#!/usr/bin/env python3
+# _*_coding:utf-8_*_
+# @Author : Xnhyacinth
+# @Time : 2022/3/8 17:23
+
+from matplotlib import pyplot as plt
+import cv2
+import numpy as np
+
+
+# 定义高斯噪声函数
+def noise_Gaussian(img, mean, var):
+    """
+    img:原始图像
+    mean：均值
+    var：方差，值越大噪声越大
+    """
+
+    # 创建均值为mean方差为var呈高斯分布的图像矩阵
+    noise = np.random.normal(mean, var ** 0.5, img.shape)
+
+    # 将原始图像的像素值进行归一化，除以255使像素值在0-1之间
+    img = np.array(img / 255, dtype=float)
+
+    # 噪声和图片合并即加噪后的图像
+    out = img + noise
+
+    # 解除归一化，乘以255将加噪后的图像的像素值恢复
+    out = np.uint8(out * 255)
+    return out
+
+
+# 高斯滤波器
+
+def gaussian_filter(img, K_size=3, sigma=1.0):
+    img = np.asarray(np.uint8(img))
+    if len(img.shape) == 3:
+        H, W, C = img.shape
+    else:
+        img = np.expand_dims(img, axis=-1)
+        H, W, C = img.shape
+
+    ## Zero padding
+    pad = K_size // 2
+    out = np.zeros((H + pad * 2, W + pad * 2, C), dtype=np.float64)
+    out[pad: pad + H, pad: pad + W] = img.copy().astype(np.float64)
+
+    ## prepare Kernel
+    K = np.zeros((K_size, K_size), dtype=np.float64)
+    for x in range(-pad, -pad + K_size):
+        for y in range(-pad, -pad + K_size):
+            K[y + pad, x + pad] = np.exp(-(x ** 2 + y ** 2) / (2 * (sigma ** 2)))
+    K /= (2 * np.pi * sigma * sigma)
+    K /= K.sum()
+    tmp = out.copy()
+
+    # filtering
+    for y in range(H):
+        for x in range(W):
+            for c in range(C):
+                out[pad + y, pad + x, c] = np.sum(K * tmp[y: y + K_size, x: x + K_size, c])
+    out = np.clip(out, 0, 255)
+    out = out[pad: pad + H, pad: pad + W].astype(np.uint8)
+    return out
+
+
+if __name__ == "__main__":
+    # 上传图片
+    # read the image
+    img = cv2.imread("D:/photo/11.jpeg")
+
+    # Print out the type of image data and its dimensions (height, width, and color)
+    print('This image is:', type(img),
+          ' with dimensions:', img.shape)
+
+    img_copy = np.copy(img)
+    # change color to rgb(from bgr)
+    img_copy = cv2.cvtColor(img_copy, cv2.COLOR_BGR2RGB)
+    # display
+    cv2.imshow("img", img_copy)
+
+    ## 添加高斯噪声
+    # 显示添加高斯噪声的结果图像
+    out = noise_Gaussian(img_copy, mean=0, var=0.003)
+    cv2.imshow("out", out)
+
+    # sigma=0.5 k_size=3
+    out_05 = gaussian_filter(img_copy, sigma=0.5)
+    cv2.imshow("out_05", out_05)
+
+    # sigma=0.5 k_size=5
+    out_05_5 = gaussian_filter(img_copy, K_size=5, sigma=0.5)
+    cv2.imshow("out_05_5", out_05_5)
+
+    cv2.waitKey(0)