Removing Rain from a Single Image via Discriminative Sparse Coding

Abstract

Visual distortions on images caused by bad weather conditions can have a negative impact on the performance of many outdoor vision systems. One often seen bad weather is rain which causes significant yet complex local intensity fluctuations in images. The paper aims at developing an effective algorithm to remove visual effects of rain from a single rain image, i.e. separate the rain layer and the derained image layer from a rain image. Built upon a nonlinear generative model of rain image, namely screen blend model, we propose a dictionary learning based algorithm for single image de-raining. The basic idea is to sparsely approximate the patches of two layers by very high discriminative codes over a learned dictionary with strong mutual exclusivity property. Such discriminative sparse codes lead to accurate separation of two layers from their non-linear composite. The experiments show that the proposed method outperforms the existing single image de-raining methods on tested rain images.

Weighted median guided filtering method for single image rain removal

“…For example, because of the lost of the textures information in the output images, several mountains become one in the first image in the fourth row, and clouds in the third image in the fourth row are also disappeared. For Luo's method [5] and Huang's method [3], it can be seen that there are many rain streaks remained in the output images, as shown in the fifth row and sixth row in Fig. 3.…”

“…Hence, we collected totally 100 natural/synthetic rainy images from the Internet and also from the test image data set released from [10]. We compare the proposed method with several stateof-the-art methods, including Li's method [10], Manu's method [6], Luo's method [5], and Huang's method [3], the results of these methods are generated in MATLAB with suggested parameter setting by the authors.…”

“…This is caused by the inaccurate decomposition of the high-frequency portion into rain components and non-rain components, which failed to recover the non-rain components and faulty incorporation of the rain components into the low-frequency partition. Similar methods have also been proposed in [4,5], Kang et al [4] proposed a method that employing bilateral filter to divide the image with rain into low-frequency portions and highfrequency portions firstly. The rain component is then extracted from the high-frequency portion by using a sparse representation-based dictionary partition in which the dictionary is classified using HOG in each atom where the bilateral filter is used to separate the low-frequency part from its high-frequency part of an input image.…”

“…For Manu's method [10], as shown in the fourth row in Fig. 3, Table 2 Quantitative comparison of different methods on real rainy images Li's method [10] Manu's method [6] Luo's method [5] Huang's method [3] Our method 0.2854 ± 0.01 0.1665 ± 0.01 0.3287 ± 0.01 0.1823 ± 0.01 0.3925 ± 0.01 it works well on rain removal. However, the whole image is over smoothed, the output image looks unnatural, and part texture information is lost.…”

“…Though the decomposition idea is elegant, the selection of dictionaries and parameters are heavily empirical, and the results are sensitive to the choice of dictionaries. Moreover, all the three dictionary learning-based frameworks [3,5] suffer from heavy computation cost. In [6], Manu uses the L0 gradient minimization approach for rain removal.…”