Skip to content

kcd71461/srcnn-tensorflow

Repository files navigation

srcnn-tensorflow

directory structure

./datasets: training bitmap files
./checkpoint: checkpoint save dir. model will saved in "./checkpoint/srcnn".
./logs: summary logs
./test: test images dir
./result: test ouput dir

script

generate_train_h5: generate "train.h5" file
train.py: train
test.py: generate original, bicubic, srcnn results

How to use

learning

  1. ./datasets 경로에 학습할 이미지(.bmp)를 넣고
  2. "generate_train_h5.py"를 실행시켜 train.h5파일을 생성하고
  3. "train.py" 실행

test

  1. ./test 경로에 테스트할 이미지(.bmp)를 넣고
  2. "test.py --test_img {파일명}" 실행. (./test 경로내의 이미지 파일명)

About

srcnn-tensorflow

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages