The otter image enhancer is a batch processing tool for upscaling folders of images into higher resolution using OpenCV's super resolution and the ESDR models.
It downloads the models at runtime from Saafke/EDSR_Tensorflow
- At least Python 3.8
Firs install requirements.txt
pip install -r requirements.txt
You can the run the script with: python main.py [input folder] [method]
Example: python main.py input/images shrink
The methods available are:
- shrink: It first downscales the images by a factor of 2 then it upscales by 4x using ESDR, this method helps if you have very noisy images
- upscale: It skips the shrink and only upscales the images by a factor of 4
- denoise: Runs Open CV fast denoising algorithm
The script will process images up to 4 sub-folders deep
The processed images will be saved to the output folder in a subfolder named with timestamp of the start of the and the process used.
Example: output/2022-07-22_21-36-00_shrink
It will include a log and a manifest of the images in CSV.