搜索资源列表
POCS
- 提供了基于POCS算法的超分辨率图像重建的源程序,效果较好-POCS algorithm based on super resolution image reconstruction of the source, the effect is good
sr_by_example-master
- 图像超分辨率重建,基于例子的单幅图像处理,对于图像处理初学者有很大帮助-image supreme-resolution,Based on the example of a single image processing, are of great help for beginners to image processing
waifu2x-master
- 又waifu2x提供,可以去git上找源码,这里方便大家使用。主要是通过卷积神经网,进行图片的放大达到超分辨率作用。-And waifu2x offer, you can go on to find the git source code, to facilitate the use here. Mainly through the convolution neural network, amplified to achieve super-resolution image of the rol
MOVIE
- C++编写,用于视频超分辨率中评价视频连续性,需要输入原始视频和重建后的视频进行对比,视频格式为YUV420-C++ Prepared for video super-resolution uation of video continuity, the need to import the original video and reconstructed video for comparison, the video format YUV420
SRCNN
- 通过卷积神经网络CNN实现超分辨率重建,利用训练模型实现参数和权重偏移的训练,达到输入低分辨率图像,输出高分辨率图像,测试例程附录。-Convolution neural network CNN to achieve super-resolution reconstruction using a training model parameters and to achieve weight shift training to achieve a low-resolution image inpu
SR
- 使用IBP算法完成四张相邻帧图像的超分辨率合成,效果非常好,代码有详细的注释。-Use IBP super-resolution algorithm to complete the synthesis of four adjacent frames of the image, the effect is very good, the code has detailed notes.
VbGm
- 实现论文 基于稀疏表示的图像超分辨率《Image Super-Resolution Via Sparse Representation》中的Figure2,通过对100000个高分辨率和低分辨率图像块训练得到的高分辨率图像块字典,字典原子总数为512,图像块尺寸大小为9X9.-The Figure2 in 《Image Super-Resolution Via Sparse Representation》
LMS
- 该算法同于计算最小平法差,在图像复原和超分辨率重建中是一个较好的评价参数。-The algorithm is the same as calculating the least square difference, which is a good uation parameter in image restoration and super-resolution reconstruction.
SCN_Matlab-master
- 图像超分辨率重建——CsN_Matlab实现程序,仅作为学习参考之用。-Image super- resolution reconstruction
web-images
- 本程序是基于web图像的超分辨率重建,借鉴参考了其他人的算法-This procedure is based on the super image resolution of web image reconstruction, reference to other people s algorithm
SR
- 图像单帧超分辨率重建主要是将已知的先验知识融合进入图像之中,使图像具有更好的细节和视觉效果。- 46/5000 Túxiàng dān zhèng chāo fēnbiàn lǜ chóngjiàn zhǔyào shi jiāng yǐ zhī de xiān yàn zhīshì rónghé jìnrù túxiàng zhī zhōng, shǐ túxiàng jùyǒu gèng hǎo de xìjié hé shìjué xiàoguǒ. Image single-f
drcn
- 用递归的卷积神经网络方法实现图像的超分辨率重建(深度学习)-Deeply-Recursive Convolutional Network for Image Super-Resolution
77bea7e34bd8
- 利用小波双线性超分辨率重建算法得到的重建图像会出现低、高频系数不匹配的现象,从而使得到的高分辨率图像灰度偏移。本文对该方法进行了改进,并引入局部适应插值得到更为理想的重建算法。即小波与局部适应插值结合算法。-The reconstructed image obtained by wavelet bilinear super- resolution reconstruction algorithm will appear the phenomenon of low and high frequen
IBPsuanfa
- 迭代反向投影法,适合用图像超分辨率复原。内附图片示例-Iterative reverse projection method, suitable for image super resolution resolution. An example of a picture
EFSCG67
- 超分辨率源代码,用它可以实现超分辨率,提高影像的分辨率-Super-resolution source code, using it can realize the super-resolution, improve the resolution of the image
基于单幅图像景深和大气散射模型的去雾方法
- 单景图像超分辨率技术,创新算法,值得研究学习。(Single scene image super-resolution technology, innovative algorithms, it is worth studying.)
SRCNN_v1
- 主要用于图像增强,眼白血管纹理增强,运用的是深度学习算法,超分辨率(For image enhancement, white vascular texture enhancement is the use of deep learning algorithms, super-resolution)
超分辨率学习文档
- 对SRCNN论文的一个翻译学习,其次是在SRCNN的基础上,对训练方法做出了一定改进。(A translation study of SRCNN thesis, and, on the basis of SRCNN, the training methods have been improved.)
Test code for SRCNN
- SRCNN代码实现。该代码使用三层卷积神经网络,进行图像的超分辨率重建,效果比双三次插值好很多(The code uses three layer convolutional neural network for image super-resolution reconstruction, the effect is much better than the double three interpolation.)
DRCN_code
- 使用matlab程序超分辨率重建,图像超分辨率程序(DCRN,Matlab code about super-resolution)