请问flare图像如何与背景图像合并成一个人造的退化图像？

[HDUyiming]

您能提供一下给定一个F和B，如何变为Isys的代码片段吗？我想给人造夜晚图像添加上flare使其更加逼真，但是我不知道具体的代码实现，希望您可以帮助到我😀谢谢！

[qulishen]

import torch
import torch.utils.data as data
import torchvision.transforms as transforms

import numpy as np
from PIL import Image
import glob
import random

import torchvision.transforms.functional as TF
from torch.distributions import Normal
import torch
import numpy as np
import torch
import torchvision.transforms as transforms
from matplotlib import pyplot as plt

class RandomGammaCorrection(object):
	def __init__(self, gamma = None):
		self.gamma = gamma
	def __call__(self,image):
		if self.gamma == None:
			# more chances of selecting 0 (original image)
			gammas = [0.5,1,2]
			self.gamma = random.choice(gammas)
			return TF.adjust_gamma(image, self.gamma, gain=1)
		elif isinstance(self.gamma,tuple):
			gamma=random.uniform(*self.gamma)
			return TF.adjust_gamma(image, gamma, gain=1)
		elif self.gamma == 0:
			return image
		else:
			return TF.adjust_gamma(image,self.gamma,gain=1)

def remove_background(image):
	#the input of the image is PIL.Image form with [H,W,C]
	image=np.float32(np.array(image))
	_EPS=1e-7
	rgb_max=np.max(image,(0,1))
	rgb_min=np.min(image,(0,1))
	image=(image-rgb_min)*rgb_max/(rgb_max-rgb_min+_EPS)
	image=torch.from_numpy(image)
	return image

class Flare_Image_Loader(data.Dataset):
	def __init__(self, image_path ,transform_base=None,transform_flare=None,mask_type=None):
		self.ext = ['png','jpeg','jpg','bmp','tif']
		self.data_list=[]
		[self.data_list.extend(glob.glob(image_path + '/*.' + e)) for e in self.ext]
		self.flare_dict={}
		self.flare_list=[]
		self.flare_name_list=[]

		self.reflective_flag=False
		self.reflective_dict={}
		self.reflective_list=[]
		self.reflective_name_list=[]

		self.mask_type=mask_type #It is a str which may be None,"luminance" or "color"

		self.transform_base=transform_base
		self.transform_flare=transform_flare

		print("Base Image Loaded with examples:", len(self.data_list))

	def __getitem__(self, index):
		# load base image
		img_path=self.data_list[index]
		base_img= Image.open(img_path)
		
		gamma=np.random.uniform(1.8,2.2)
		to_tensor=transforms.ToTensor()
		adjust_gamma=RandomGammaCorrection(gamma)
		adjust_gamma_reverse=RandomGammaCorrection(1/gamma)
		color_jitter=transforms.ColorJitter(brightness=(0.8,3),hue=0.0)
		if self.transform_base is not None:
			base_img=to_tensor(base_img)
			base_img=adjust_gamma(base_img)
			base_img=self.transform_base(base_img)
		else:
			base_img=to_tensor(base_img)
			base_img=adjust_gamma(base_img)
		sigma_chi=0.01*np.random.chisquare(df=1)
		base_img=Normal(base_img,sigma_chi).sample()
		gain=np.random.uniform(0.5,1.2)
		flare_DC_offset=np.random.uniform(-0.02,0.02)
		base_img=gain*base_img
		base_img=torch.clamp(base_img,min=0,max=1)

		#load flare image
		flare_path=random.choice(self.flare_list)
		flare_img =Image.open(flare_path)
		if self.reflective_flag:
			reflective_path=random.choice(self.reflective_list)
			reflective_img =Image.open(reflective_path)


		flare_img=to_tensor(flare_img)
		flare_img=adjust_gamma(flare_img)
		
		if self.reflective_flag:
			reflective_img=to_tensor(reflective_img)
			reflective_img=adjust_gamma(reflective_img)
			flare_img = torch.clamp(flare_img+reflective_img,min=0,max=1)

		flare_img=remove_background(flare_img)

		if self.transform_flare is not None:
			flare_img=self.transform_flare(flare_img)
		
		#change color
		flare_img=color_jitter(flare_img)

		#flare blur
		blur_transform=transforms.GaussianBlur(21,sigma=(0.1,3.0))
		flare_img=blur_transform(flare_img)
		flare_img=flare_img+flare_DC_offset
		flare_img=torch.clamp(flare_img,min=0,max=1)

		#merge image	
		merge_img=flare_img+base_img
		merge_img=torch.clamp(merge_img,min=0,max=1)

		if self.mask_type==None:
			return adjust_gamma_reverse(base_img),adjust_gamma_reverse(flare_img),adjust_gamma_reverse(merge_img),gamma
		elif self.mask_type=="luminance":
			#calculate mask (the mask is 3 channel)
			one = torch.ones_like(base_img)
			zero = torch.zeros_like(base_img)

			luminance=0.3*flare_img[0]+0.59*flare_img[1]+0.11*flare_img[2]
			threshold_value=0.99**gamma
			flare_mask=torch.where(luminance >threshold_value, one, zero)

		elif self.mask_type=="color":
			one = torch.ones_like(base_img)
			zero = torch.zeros_like(base_img)

			threshold_value=0.99**gamma
			flare_mask=torch.where(merge_img >threshold_value, one, zero)

		return adjust_gamma_reverse(base_img),adjust_gamma_reverse(flare_img),adjust_gamma_reverse(merge_img),flare_mask,gamma

	def __len__(self):
		return len(self.data_list)
	
	def load_scattering_flare(self,flare_name,flare_path):
		flare_list=[]
		[flare_list.extend(glob.glob(flare_path + '/*.' + e)) for e in self.ext]
		self.flare_name_list.append(flare_name)
		self.flare_dict[flare_name]=flare_list
		self.flare_list.extend(flare_list)
		len_flare_list=len(self.flare_dict[flare_name])
		if len_flare_list == 0:
			print("ERROR: scattering flare images are not loaded properly")
		else:
			print("Scattering Flare Image:",flare_name, " is loaded successfully with examples", str(len_flare_list))
		print("Now we have",len(self.flare_list),'scattering flare images')

	def load_reflective_flare(self,reflective_name,reflective_path):
		self.reflective_flag=True
		reflective_list=[]
		[reflective_list.extend(glob.glob(reflective_path + '/*.' + e)) for e in self.ext]
		self.reflective_name_list.append(reflective_name)
		self.reflective_dict[reflective_name]=reflective_list
		self.reflective_list.extend(reflective_list)
		len_reflective_list=len(self.reflective_dict[reflective_name])
		if len_reflective_list == 0:
			print("ERROR: reflective flare images are not loaded properly")
		else:
			print("Reflective Flare Image:",reflective_name, " is loaded successfully with examples", str(len_reflective_list))
		print("Now we have",len(self.reflective_list),'refelctive flare images')


transform_base=transforms.Compose([transforms.RandomCrop((512,512),pad_if_needed=True,padding_mode='reflect'),
							  transforms.RandomHorizontalFlip(),
							  transforms.RandomVerticalFlip()
                              ])

transform_flare=transforms.Compose([transforms.RandomAffine(degrees=(0,360),scale=(0.8,1.5),translate=(300/1440,300/1440),shear=(-20,20)),
                              transforms.CenterCrop((512,512)),
							  transforms.RandomHorizontalFlip(),
							  transforms.RandomVerticalFlip()
                              ])
# 加载数据集
flare_image_loader=Flare_Image_Loader('Flickr24K',transform_base,transform_flare)
flare_image_loader.load_scattering_flare('Flare7K','Flare7k/Scattering_Flare/Compound_Flare')
flare_image_loader.load_reflective_flare('Flare7K','Flare7k/Reflective_Flare')

# 选取背景图像和耀斑图像
img_index=10
test_base_img,test_flare_img,test_merge_img,flare_mask=flare_image_loader[img_index]

#展示结果
plt.imshow(test_flare_img.permute(1,2,0))
plt.show()
plt.imshow(test_merge_img.permute(1,2,0))
plt.show()
plt.imshow(test_base_img.permute(1,2,0))
plt.show()

[HDUyiming]

谢谢学长！万分感谢！！！小星星献上！！！