Implementation of the Image Reflection Suppression algorithm from Yang et. al in Python.
Removing unwanted reflections from images recorded through glass is an important task. It can be used to preprocess images for machine learning and pattern recognition applications. Tests on simulated and real-world photos show that our implementation delivers desired results for reflection suppression while drastically cutting execution time.
pip install numpy==1.21.4
pip install scipy==1.8.0
pip install scikit-image==0.21.0
pip install opencv-python==4.7.0.72
image_dereflect.py
- Contains the code to dereflect an imagecreate_synthetic_blend.py
- Contains the code to create a synthetic blend of a reflection and a background image (blends 2 images to form a synthetic image with a reflection)figures\*
- Contains the images showing the results of the blending and dereflection algorithms
python image_dereflect.py [-t GRADIENT_THRESHOLD -e EPSILON] IMAGE_PATH [GROUND_TRUTH_IMAGE_PATH]
python image_dereflect.py -t 0.04 -e 1e-6 reflection_image.jpg ground_truth_image.jpg