When we take photo outdoors, weather is a one of key factors affecting the quality of image. In this project, we try to remove haze of photo to enhance the visibility of image. Due to lack of depth information in images, previous dehazing methods need to include extra information. For example, they evaluate depth by using multiple polarization image or 3D models. Recently, there are some ‘single image dehazing’ methods developed.

In this project, we tried to reproduce one of the methods, which is based on physical model and do well in heavy haze images [1]. Unlike the original paper, we replace the matting laplacian matrix with guided filter to save huge memory size and computation time [2]. We also compare the results generated from matting laplacian matrix and guided filter.

References:

1. R. T. Tan, “Visibility in bad weather from a single image,” in IEEE Conference on Computer Vision and Pattern Recognition, 2008. CVPR 2008, 2008, pp. 1-8.
2. K. He, J. Sun, and X. Tang, “Guided image filtering,” in Proceedings of the 11th European conference on Computer vision: Part I, Berlin, Heidelberg, 2010, pp. 1–14.