atomicoo / enhanceimg

Image-enhancement algorithms: low-light enhancement, image restoration, super-resolution reconstruction. 图像增强算法探索：低光增强、图像修复、超分辨率重建 ……

EnhanceIMG

[TOC]

此代码库用于图像增强算法的探索，主要包括：低光增强、图像修复、超分辨率重建 ……

目录结构

.
|--- asserts/
|--- awegan/      # GAN相关算法
     |--- datasets/
     |--- models/         # CycleGAN/Pix2Pix/SelfGAN
     |--- options/
     |--- util/
     |--- __init__.py
     |--- train.py
     |--- ...
|--- colorspace/  # 色彩空间转换
|--- edges/       # 边缘检测算法
|--- filters/     # 各种滤波器
|--- histeq/      # 直方图均衡算法
|--- noises/      # 噪声
|--- priors/      # 自然图像先验信息
     |--- __init__.py
     |--- denoising.py
     |--- inpainting.py
     |--- networks.py     # ResNet/SkipNet/UNet
     |--- restoration.py
     |--- ...
|--- retinex/     # Retinex系列算法
     |--- __init__.py
     |--- enhancer.py
     |--- retinex_net.py  # RetinexNet
     |--- ...
|--- utils/       # 一些方法
|--- .gitignore
|--- demo.py
|--- LICENSE
|--- Madison.png
|--- README.md    # 说明文档
|--- requirements.txt     # 依赖文件

简单示例

添加噪声

噪声（原图|椒盐噪声|高斯噪声）

各种滤波器

滤波器（椒盐噪声|均值滤波|中值滤波）

滤波器（高斯噪声|高斯滤波|双边滤波|联合双边滤波）

滤波器（高斯噪声|引导滤波）

边缘检测

检测算子（灰度图|Laplacian|Sobel|Scharr）

检测算子（灰度图|LoG|DoG|Gabor）

其他算法（灰度图|结构森林|HED|HED-feats-5）

传统增强算法

直方图均衡（原图|HE|AHE|CLAHE）

Gamma 校正（原图|Gamma|Gamma+MSS）

Retinex（原图|MSRCR|AMSRCR|MSRCP）

Retinex 增强（原图|AttnMSR|AttnMSR+MSS）（Mine）

自然图像先验

降噪（噪声图|降噪1|降噪2）

神经网络

RetinexNet（原图|RetinexNet）

生成对抗网络

Pix2Pix

（边缘 <=> 图像）

（低光 <=> 正常）

CycleGAN

（夏天 <=> 冬天）

（低光 <=> 正常）

参考资料

• Retinex
  • Multiscale Retinex
  • An automated multi Scale Retinex with Color Restoration for image enhancement
  • A multiscale retinex for bridging the gap between color images and the human observation of scenes
  • Deep Retinex Decomposition for Low-Light Enhancement
• Image Prior
  • Deep Image Prior
• GAN network
  • Image-to-Image Translation with Conditional Adversarial Networks
  • Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
  • High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
  • Toward Multimodal Image-to-Image Translation
  • Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
  • Deep Residual Learning for Image Recognition
  • U-Net: Convolutional Networks for Biomedical Image Segmentation
  • Fully Convolutional Networks for Semantic Segmentation

TODO

• AttnMSR 图像增强算法（Mine）
• RetinexNet 低光增强模型
• ResNet / SkipNet / UNet
• Deep Image Prior（自然图像先验信息）
• Pix2Pix 模型用于图像增强
• CycleGan 模型用于图像增强
• SelfGAN 图像增强模型（Mine，完善中）

欢迎交流

• 微信号：YcZhouZy

• 企鹅号：793071559

enhanceimg's Issues

performance issue

hello author,

i have implemented similar fcn vgg like yours. But my fcn32 gives better results than fcn 16 and 8. why is this. Can you please check my code.

import torch
import torch.nn as nn
import torchvision.models as models
from pytorch_model_summary import summary

vgg16 = models.vgg16(pretrained=True)
for param in vgg16.features.parameters():
param.requires_grad = False
#False Total params: 185,771,904 Trainable params: 171,057,216 Non-trainable params: 14,714,688
#true Total params: 185,771,904 Trainable params: 185,771,904 Non-trainable params: 0

class fcn(nn.Module):
def init(self):
super(fcn, self).init()
self.features = vgg16.features
self.classifier = nn.Sequential(
nn.Conv2d(512, 4096, 7),
nn.ReLU(inplace=True),
#nn.Dropout2d(),
nn.Conv2d(4096, 4096, 1),
nn.ReLU(inplace=True),
#nn.Dropout2d(),
nn.Conv2d(4096, 32, 1),
nn.ConvTranspose2d(32, 32, 224, stride=32)
)

def forward(self, x):
x = self.features(x)#/32
x = self.classifier(x)
#print(x.shape)
return x

class fcn16(nn.Module):
def init(self):
super(fcn16, self).init()
self.features = vgg16.features
self.classifier = nn.Sequential(
nn.Conv2d(512, 4096, 7),
nn.ReLU(inplace=True),
nn.Conv2d(4096, 4096, 1),
nn.ReLU(inplace=True),
nn.Conv2d(4096, 32, 1)
)
self.score_pool4 = nn.Conv2d(512, 32, 1)
self.upscore2 = nn.ConvTranspose2d(32, 32, 14, stride=2, bias=False)
self.upscore16 = nn.ConvTranspose2d(32, 32, 16, stride=16, bias=False)

def forward(self, x):
pool4 = self.features:-7#512 features /16
pool5 = self.features-7:#512 features /16/2=/32
pool5_upscored = self.upscore2(self.classifier(pool5))#32 class features stride2 /32*2=/16
pool4_scored = self.score_pool4(pool4)#32 features /16
combined = pool4_scored + pool5_upscored
#combined = torch.cat([pool4_scored, pool5_upscored])
res = self.upscore16(combined)# /1
return res

class fcn8(nn.Module):
def init(self):
super(fcn8, self).init()
self.features = vgg16.features
self.classifier = nn.Sequential(
nn.Conv2d(512, 4096, 7),
nn.ReLU(inplace=True),
nn.Conv2d(4096, 4096, 1),
nn.ReLU(inplace=True),
nn.Conv2d(4096, 32, 1)
)
self.score_pool4 = nn.Conv2d(512, 32, 1)
self.score_pool3 = nn.Conv2d(256, 32, 1)
self.upscore2 = nn.ConvTranspose2d(32, 32, 14, stride=2, bias=False)
self.upscore3 = nn.ConvTranspose2d(32, 32, 2, stride=2, bias=False)
#self.upscore16 = nn.ConvTranspose2d(32, 32, 16, stride=16, bias=False)
self.upscore8 = nn.ConvTranspose2d(32, 32, 8, stride=8, bias=False)

def forward(self, x):
pool3 = self.features:-14#256 features /8
pool4 = self.features-14:-7#512 features /8/2=16
pool5 = self.features-7:#512 features /16/2=/32
pool5_upscored = self.upscore2(self.classifier(pool5))#32 class features stride2 /322=/16
pool4_scored = self.score_pool4(pool4)#32 class features /16
pool3_scored = self.score_pool3(pool3)#32 class features /8
combined = pool4_scored + pool5_upscored #/16
#print(combined.shape)
combined_upscored = self.upscore3(combined)#32 class features stride2 /162=/8
#print(combined_upscored.shape)
combined2 = pool3_scored + combined_upscored
#print(combined2.shape)
#res = self.upscore16(combined)#/1
res = self.upscore8(combined2)#/1
#print(res.shape)
return res

gan模型权重

感谢大佬的分享 请问下gan模型那边权重有下载链接吗

如何运行？