图像处理python代码（自用）
1、二值图求差并计算比例
import cv2
import numpy as np
import os
from PIL import Image
# 需要输入图像的路径
img_path = "./gtFine/train/raw/"
mask_path = "./mask/train/"
imgROI_path = "./diff/train/"
if not os.path.exists(imgROI_path):
os.makedirs(imgROI_path)
sum = 0
count = 0
for _image in os.listdir(img_path):
if _image.split('.')[1] == "png":
print(_image)
# 将同名的图片和mask相减
img_path_ = os.path.join(img_path, _image)
portion = os.path.splitext(_image)
#mask_name = portion[0] + "_color" + portion[1]
mask_name = portion[0] + portion[1]
mask_path_ = os.path.join(mask_path, mask_name)
img = cv2.imread(img_path_)
mask = cv2.imread(mask_path_)
result1 = cv2.subtract(img, mask)
result2 = cv2.subtract(mask, img)
result = cv2.add(result1, result2)
result = cv2.cvtColor(result, cv2.COLOR_BGR2GRAY)
color = np.where(result == 255)[0].shape[0]
pixel_sum = result.shape[0] * result.shape[1]
sum += color / pixel_sum
count += 1
result = Image.fromarray(result)
result.save(imgROI_path + _image)
print(1-sum/count)
2、孔洞填充及膨胀腐蚀操作
from PIL import Image
import os
import numpy as np
import cv2
img_path = './mask/train/'
result_path = './mask_result/train/'
# 先采用开运算去除小的噪点 然后膨胀 最后使用孔洞填充
# 孔洞填充
def im_fill(imgray):
# 原图取补得到MASK图像
mask = 255 - imgray
# 构造Marker图像
marker = np.zeros_like(imgray)
marker[0, :] = 255
marker[-1, :] = 255
marker[:, 0] = 255
marker[:, -1] = 255
marker_0 = marker.copy()
# 形态学重建
SE = cv2.getStructuringElement(shape=cv2.MORPH_CROSS, ksize=(3, 3))
while True:
marker_pre = marker
dilation = cv2.dilate(marker, kernel=SE)
marker = np.min((dilation, mask), axis=0)
if (marker_pre == marker).all():
break
dst = 255 - marker
return dst
if not os.path.exists(result_path):
os.makedirs(result_path)
for _image in os.listdir(img_path):
print(_image)
img_path_ = os.path.join(img_path, _image)
img = cv2.imread(img_path_)
# Use closing operate
# closing_kernel = np.ones((20, 20), np.uint8)
# img = cv2.morphologyEx(img, cv2.MORPH_CLOSE, closing_kernel, iterations=2)
# img = np.hstack((img, closing))
# Use opening operate
opening_kernel = np.ones((10, 10), np.uint8)
img = cv2.morphologyEx(img, cv2.MORPH_OPEN, opening_kernel, iterations=1)
# img = np.hstack((img, opening))
# Use dilate operate
dilate_kernel = np.ones((60, 60), np.uint8)
img = cv2.dilate(img, dilate_kernel, iterations=1)
img = im_fill(img)
res_img = Image.fromarray(img)
res_img.save(result_path + _image)
3、获取目录下所有图片像素值
from PIL import Image
import os
def checkColor(dir, file, recursion):
files = os.listdir(dir)
for name in files:
fullname = os.path.join(dir, name)
if (os.path.isdir(fullname) & recursion):
checkColor(fullname, file, recursion)
else:
image = Image.open(fullname)
file.write(name)
file.write(str(image.getcolors()))
file.write("\n")
def Test():
dir = "./train"
# txt文件名
outfile = "train.txt"
file = open(outfile, "w")
checkColor(dir, file, 1)
file.close()
Test()
4、高斯模糊
from PIL import Image, ImageFilter
# 打开图像
im1 = Image.open('iron_man.jpg')
# 创建一个im1两倍宽的图像
img = Image.new('RGB', (im1.width*2, im1.height), 'red')
# 高斯模糊处理
im2 = im1.filter(ImageFilter.GaussianBlur)
# 将im1粘贴到img上
img.paste(im1, (0, 0))
# 将im2（高斯模糊后的图像）粘贴到img上
img.paste(im2, (im1.width, 0))
img.show()
5、减法运算
from PIL import Image, ImageChops
# 打开图像
im1 = Image.open('xscn.jpg')
im2 = Image.open('xscn2.jpg')
# 合成图像并显示
im3 = ImageChops.subtract(im1, im2)
im3.show()
6、ImageEnhance模块（色彩、亮度）
from PIL import Image, ImageEnhance
# 打开im1
im1 = Image.open("gtx.jpg")
# 获取颜色（各种）调整器
enhance_im1 = ImageEnhance.Color(im1)
#enhance_im1 = ImageEnhance.Contrast(im1)
#enhance_im1 = ImageEnhance.Brightness(im1)
#enhance_im1 = ImageEnhance.Sharpness(im1)
# 减弱颜色（以及其它属性）
im2 = enhance_im1.enhance(0.5)
# 增强颜色（以及其它属性）
im3 = enhance_im1.enhance(1.5)
# 获取原图大小
w, h = im1.size
# 创建一个原图大小3倍的图片
img = Image.new("RGB", (w*3, h))
# 将减弱的图片放在最左边
img.paste(im2, (0, 0))
# 将原图放在中间
img.paste(im1, (w, 0))
# 将增强后的图片放在最右边
img.paste(im3, (w*2, 0))
# 显示图片
img.show()
7、图像缩小
#!/usr/bin/env python
# encoding: utf-8
"""
@author: H
@software: Pycharm
@file: ImageInterpolation.py
@time: 2020/10/16 8:28
"""
import os
import numpy as np
from PIL import Image
file_path = r'D:\postgraduate\first_year\数字图像处理\作业\homework1\practice1\image\jerry.jpg'
img = Image.open(file_path) # 读取图片，格式为Image
# 把Image格式的图片转化为numpy格式进性图像处理
img = np.array(img, dtype=np.uint8)
height, width, mode = img.shape[0], img.shape[1], img.shape[2] # 取出高、宽、通道数
print(height, width, mode) # (275 183 3)
# 缩放的目标大小，这里以缩放为原图的1/2为例
desWidth = int(width * 0.5)
desHeight = int(height * 0.5)
desImage = np.zeros((desHeight, desWidth, mode), np.uint8) # 定义一个目标图片代表的array，纯黑图片
# 像素填充
# 方法1：最近邻插值法
for des_x in range(0, desHeight):
for des_y in range(0, desWidth):
# 判断新像素点在原图中的像素点坐标
src_x = int(des_x * (height/desHeight))
src_y = int(des_y * (width/desWidth))
desImage[des_x, des_y] = img[src_x, src_y] # 填充
print(desImage.shape)
des_img = Image.fromarray(desImage)
des_img.save('./image/jerry.jpg') # 图片另存为
8、图像放大
#!/usr/bin/env python
# encoding: utf-8
"""
@author: H
@software: Pycharm
@file: ImageInterpolation1.py
@time: 2020/10/17 20:31
"""
import os
import numpy as np
from PIL import Image
"""
图像缩放常见三种方法的实现：
- 最近邻插值法
- 双线性插值法
- 双立方插值法
"""
def Interpolation_NNI(filepath):
"""
最近邻插值法(nearest neighbor interpolation)
:param filepath:
:return:
"""
img = Image.open(file_path) # 读取图片，格式为Image
# img.show() # 显示图片
# 把Image格式的图片转化为numpy格式进性图像处理
img = np.array(img, dtype=np.uint8)
height, width, mode = img.shape[0], img.shape[1], img.shape[2] # 取出高、宽、通道数
print(height, width, mode) # (275 183 3)
# 缩放的目标大小，这里以缩放为原图的1/2为例
desWidth = int(width * 0.5)
desHeight = int(height * 0.5)
desImage = np.zeros((desHeight, desWidth, mode), np.uint8) # 定义一个目标图片代表的array，纯黑图片
# 像素填充
# 方法1：最近邻插值法
for des_x in range(0, desHeight):
for des_y in range(0, desWidth):
# 判断新像素点在原图中的像素点坐标
src_x = int(des_x * (height / desHeight))
src_y = int(des_y * (width / desWidth))
desImage[des_x, des_y] = img[src_x, src_y] # 填充
print(desImage.shape)
des_img = Image.fromarray(desImage)
des_img.save('./image/jerry.jpg')
def Interpolation_Bilinear(filepath, desHeight, desWidth):
# 双线性插值法
img = Image.open(filepath) # 读取图片
img = np.array(img, np.uint8) # 转化为numpy数组
desImageNumpy = np.zeros(img.shape, np.uint8) # 生成一个大小相同的全0的numpy数组
height, width, mode = img.shape[0], img.shape[1], img.shape[2] # 高、宽、channel数
# 找出目标位置在源图中的位置
scale_x = float(width)/desWidth # x轴缩放比例
scale_y = float(height)/desHeight # y轴缩放比例
des_image = np.zeros((desHeight, desWidth, mode), np.uint8)
for n in range(mode):
for des_y in range(desHeight):
for des_x in range(desWidth):
# 确定四个近邻点坐标
src_x = (des_x + 0.5) * scale_x - 0.5 #
src_y = (des_y + 0.5) * scale_y - 0.5
src_x_1 = int(np.floor(src_x)) #
src_y_1 = int(np.floor(src_y))
src_x_2 = min(src_x_1+1, width-1) # 防止坐标点寻找溢出
src_y_2 = min(src_y_1+1, height-1)
# 两次x轴线性插值
value_1 = (src_x_2 - src_x)*img[src_y_1, src_x_1, n]+(src_x - src_x_1)*img[src_y_1, src_x_2, n]
value_2 = (src_x_2 - src_x)*img[src_y_2, src_x_1, n]+(src_x - src_x_1)*img[src_y_2, src_x_2, n]
# y轴线性插值
des_image[des_y, des_x, n] = (src_y_2 - src_y)*value_1 + (src_y - src_y_1)*value_2
print(des_image.shape)
des_img = Image.fromarray(des_image)
des_img.save('./image/new_bilinear.jpg')
if __name__ == '__main__':
file_path = r'D:\postgraduate\first_year\数字图像处理\作业\homework1\practice1\image\kobe1.jpg'
Interpolation_Bilinear(file_path, int(183*2), int(275*2)) # 双线性插值法
9、numpy数值计算
import numpy as np
x = np.array([[1,2],[3,4]], dtype=np.float64)
y = np.array([[5,6],[7,8]], dtype=np.float64)
print x + y
print np.add(x, y)
# [[ 6.0 8.0]
# [10.0 12.0]]矩阵加
print x - y
print np.subtract(x, y)
# [[-4.0 -4.0]
# [-4.0 -4.0]]矩阵减
print x * y
print np.multiply(x, y)
# [[ 5.0 12.0]
# [21.0 32.0]]矩阵乘
print x / y
print np.divide(x, y)
# [[ 0.2 0.33333333]
# [ 0.42857143 0.5 ]]矩阵除
print np.sqrt(x)
# [[ 1. 1.41421356]
# [ 1.73205081 2. ]]矩阵开方
#和MATLAB不同，*是元素逐个相乘，而不是矩阵乘法。在Numpy中使用dot来进行矩阵乘法：
import numpy as np
x = np.array([[1,2],[3,4]])
y = np.array([[5,6],[7,8]])
v = np.array([9,10])
w = np.array([11, 12])
# 矩阵乘法，输出219
print v.dot(w)
print np.dot(v, w)
# 矩阵乘法，输出[29 67]
print x.dot(v)
print np.dot(x, v)
# 矩阵乘法，输出
# [[19 22]
# [43 50]]
print x.dot(y)
print np.dot(x, y)
#Numpy提供了很多计算数组的函数，其中最常用的一个是sum：
import numpy as np
x = np.array([[1,2],[3,4]])
print np.sum(x) # 计算所有元素和; prints "10"
print np.sum(x, axis=0) # 计算列元素和; prints "[4 6]"
print np.sum(x, axis=1) # 计算行元素和; prints "[3 7]"
#除了计算，我们还常常改变数组或者操作其中的元素。其中将矩阵转置是常用的一个，在Numpy中，使用T来转置矩阵：
import numpy as np
x = np.array([[1,2], [3,4]])
print x # Prints "[[1 2]
# [3 4]]"
print x.T # Prints "[[1 3]
# [2 4]]"
10、边缘检测
if values['canny']:
frame = cv2.Canny(frame, values['canny_slider_a'], values['canny_slider_b'])
11、轮廓检测
if values['contour']:
hue = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
hue = cv2.GaussianBlur(hue, (21, 21), 1)
hue = cv2.inRange(hue, np.array([values['contour_slider'], values['base_slider'], 40]),
np.array([values['contour_slider'] + 30, 255, 220]))
cnts= cv2.findContours(hue, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)[0]
cv2.drawContours(frame, cnts, -1, (0, 0, 255), 2)
12、调节对比度
if values['enhance']:
enh_val = values['enhance_slider'] / 40
clahe = cv2.createCLAHE(clipLimit=enh_val, tileGridSize=(8, 8))
lab = cv2.cvtColor(frame, cv2.COLOR_BGR2LAB)
lab[:, :, 0] = clahe.apply(lab[:, :, 0])
frame = cv2.cvtColor(lab, cv2.COLOR_LAB2BGR)
13、阈值二值化
if values['thresh']:
frame = cv2.cvtColor(frame, cv2.COLOR_BGR2LAB)[:, :, 0]
frame = cv2.threshold(frame, values['thresh_slider'], 255, cv2.THRESH_BINARY)[1]
14、预处理合集
#图像剪裁与预处理
#图片的缩放
import cv2
import numpy as np
img=cv2.imread('02.jpg',1)
#shape中包含三个信息
# [0]:行数
# [1]:列数
# [2]:通道数
imgInfo=img.shape
height=imgInfo[0]
width=imgInfo[1]
deep=imgInfo[2]
print(deep)
#【1】放大 缩小
a=1.2
dstHeight=int(height*a)
dstWidth=int(width*a)
#放大缩小 涉及到图像的插值
#常见的插值 最近邻域差插值 双线性插值 立方插值
dst=cv2.resize(img,(dstWidth,dstHeight))
cv2.imshow('resizeimage',dst)
#【2】图像剪切
#[2.1]中间位置剪切
x_center=int((width/2))
y_center=int((height/2))
cutimg_median=img[x_center-200:x_center+200,y_center-200:y_center+200]
cv2.imshow('cut_media',cutimg_median)
#[2.2]任意位置剪切
cutimg_every=img[200:600,200:400]
cv2.imshow('cut_every',cutimg_every)
#【3】 图片位移
#设定平移矩阵
#矩阵讲解
#matshift拆成22的矩阵和21的矩阵
#A=[[1,0],[0,1]],B=[[100],[200]] C=[X,Y]
#A*C=B
#最终效果图像沿着x轴移动100，沿着y轴移动200；
matshift=np.float32([[1,0,200],[0,1,200]])
shiftimg=cv2.warpAffine(img,matshift,(height,width))
cv2.imshow('shiftimg',shiftimg)
#【4】图像镜像
#上下镜像
#空白图像的尺寸大小
newImgInfo=(height*2,width,deep)
#创建空白图片
samedst=np.zeros(newImgInfo,np.uint8)
#开始镜像
for i in range(0,height):
for j in range(0,width):
#samedst上部分像素写入
samedst[i,j]=img[i,j]
#samedst下部分像素写入
samedst[height*2-i-1,j]=img[i,j]
#绘制镜像线
for i in range(0,width):
samedst[height,i]=(0,0,255)
cv2.imshow('samedst',samedst)
#【5】仿射变化
#设定变化矩阵 根据三组对应点寻找仿射变化的矩阵
matSrc=np.float32([[0,0],[0,height-1],[width-1,0]])
matDst=np.float32([[50,50],[300,height-200],[width-300,100]])
#寻找仿射变化矩阵
matAffine=cv2.getAffineTransform(matSrc,matDst)
affineImg=cv2.warpAffine(img,matAffine,(width,height))
cv2.imshow('affineImg',affineImg)
#【6】旋转变化
matRotate=cv2.getRotationMatrix2D((height*0.5,width*0.5),45,0.5)
rotateImg=cv2.warpAffine(img,matRotate,(height,width))
cv2.imshow('rotateImg',rotateImg)
cv2.waitKey(0)
（未完）