# RAISR
A Python implementation of [RAISR](http://ieeexplore.ieee.org/document/7744595/). Supporting color images.
## How To Use
### Prerequisites
You can install most of the following packages using [pip](https://pypi.python.org/pypi/pip).
* [OpenCV-Python](https://pypi.python.org/pypi/opencv-python)
* [NumPy](http://www.numpy.org/)
* [SciPy](https://www.scipy.org/)
* [Python Imaging Library (PIL)](http://www.pythonware.com/products/pil/)
* [Matplotlib](https://matplotlib.org/)
### Training
Put your training images in the `train` directory, and run the following command to start training.
```
python train.py
```
The learned filters will be saved in the root directory of the project.
### Testing
Put your testing images in the `test` directory, and run the following command to start generating high resolution images using RAISR.
```
python test.py
```
The result will be saved in the `results` directory.
## Visualization
### Visualing the learned filters
Uncomment the following line in `train.py`.
```
# filterplot(h, R, Qangle, Qstrength, Qcoherence, patchsize)
```
### Visualing the process of RAISR image upscaling
Uncomment the following line in `test.py`.
```
# plt.show()
```
## Testing Results
Comparing between original image, bilinear interpolation and RAISR:
| Origin | Bilinear Interpolation | RAISR |
|:----------------------:|:----------------------:|:----------------------:|
|![origin_gray_crop bmp](https://user-images.githubusercontent.com/12198424/27002908-28a69cf2-4e1f-11e7-954d-1135950cd760.png)|![cheap_crop bmp](https://user-images.githubusercontent.com/12198424/27002909-28a7834c-4e1f-11e7-875e-30bb4eaaa0d3.png)|![raisr_gray_crop bmp](https://user-images.githubusercontent.com/12198424/27002910-28ae90f6-4e1f-11e7-83e5-3e63ca07b308.png)|
Other results using images taken from [BSDS500 database](https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/resources.html) and [ArTe-Lab 1D Medium Barcode Dataset](http://artelab.dista.uninsubria.it/downloads/datasets/barcode/medium_barcode_1d/medium_barcode_1d.html):
| Origin | RAISR |
|:-------------:|:-------------:|
|![origin_crop bmp](https://user-images.githubusercontent.com/12198424/27002925-81cf7f88-4e1f-11e7-8a75-4975db1d37b8.png)|![raisr_crop bmp](https://user-images.githubusercontent.com/12198424/27002926-81d126b2-4e1f-11e7-8814-f2fce323caac.png)|
|![origin_crop bmp](https://user-images.githubusercontent.com/12198424/27002932-a39f6cc2-4e1f-11e7-9dea-3aa79a9d9764.png)|![raisr_crop bmp](https://user-images.githubusercontent.com/12198424/27002933-a3a248ac-4e1f-11e7-9c81-807d3a5eac90.png)|
|![origin_crop bmp](https://user-images.githubusercontent.com/12198424/27002942-c9ba697a-4e1f-11e7-8743-d41be65c09d3.png)|![raisr_crop bmp](https://user-images.githubusercontent.com/12198424/27002943-c9bcefd8-4e1f-11e7-9e55-bf5d93f17a9c.png)|
## References
* Y. Romano, J. Isidoro and P. Milanfar, "RAISR: Rapid and Accurate Image Super Resolution" in IEEE Transactions on Computational Imaging, vol. 3, no. 1, pp. 110-125, March 2017.
* P. Arbelaez, M. Maire, C. Fowlkes and J. Malik, "Contour Detection and Hierarchical Image Segmentation", IEEE TPAMI, Vol. 33, No. 5, pp. 898-916, May 2011.
* Alessandro Zamberletti, Ignazio Gallo and Simone Albertini, "Robust Angle Invariant 1D Barcode Detection", Proceedings of the 2nd Asian Conference on Pattern Recognition (ACPR), Okinawa, Japan, 2013
## License
[MIT.](https://github.com/movehand/raisr/blob/master/LICENSE) Copyright (c) 2017 James Chen
没有合适的资源?快使用搜索试试~ 我知道了~
谷歌2016图像超分辨率论文源码
共12个文件
py:6个
md:4个
gitignore:1个
3星 · 超过75%的资源 需积分: 50 83 下载量 169 浏览量
2017-11-09
15:05:54
上传
评论 6
收藏 10KB ZIP 举报
温馨提示
谷歌2016图像超分辨率论文源码,RAISR: Rapid and Accurate Image Super Resolution,宣称是可以在重建质量不差情况下,速度比目前算法如A+之类,能够有10到100倍性能提升,比较有工程意义。
资源推荐
资源详情
资源评论
收起资源包目录
raisr-master.zip (12个子文件)
raisr-master
.gitignore 146B
results
README.md 73B
gaussian2d.py 417B
LICENSE 1KB
filterplot.py 903B
README.md 3KB
test.py 5KB
train
README.md 63B
test
README.md 62B
hashkey.py 1KB
cgls.py 357B
train.py 7KB
共 12 条
- 1
资源评论
- sinat_369950842019-04-22效果还行吧
- nullhty2019-06-18无语,就是github上下载过来的
天泰路小王爷
- 粉丝: 9
- 资源: 41
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功