Clearing the Skies: A Deep Network Architecture for Single-Image Rain Removal

Abstract

Abstract-We introduce a deep network architecture called DerainNet for removing rain streaks from an image. Based on the deep convolutional neural network (CNN), we directly learn the mapping relationship between rainy and clean image detail layers from data. Because we do not possess the ground truth corresponding to real-world rainy images, we synthesize images with rain for training. In contrast to other common strategies that increase depth or breadth of the network, we use image processing domain knowledge to modify the objective function and improve deraining with a modestly-sized CNN. Specifically, we train our DerainNet on the detail (high-pass) layer rather than in the image domain. Though DerainNet is trained on synthetic data, we find that the learned network translates very effectively to real-world images for testing. Moreover, we augment the CNN framework with image enhancement to improve the visual results. Compared with state-of-the-art single image deraining methods, our method has improved rain removal and much faster computation time after network training.

Advancing Opportunistic Sensing in Hydrology: A Novel Approach to Measuring Rainfall With Ordinary Surveillance Cameras

“…The rationale behind the four regularization terms is explained by Figure , taking a synthetic image as an example. The synthesis rendered rain‐streak‐like noises into images through graphics‐editing software (i.e., Adobe® Photoshop) following previous studies (Fu et al, ; Kang et al, ).…”

“…The rationale behind the four regularization terms is explained by Figure 2, taking a synthetic image as an example. The synthesis rendered rain-streak-like noises into images through graphics-editing software (i.e., Adobe® Photoshop) following previous studies (Fu et al, 2017;Kang et al, 2012). Figure 2a shows the decomposition of the rain-contained image (Figure 2a-1) into a rain-free background layer (Figure 2a-2) and a rain-streak layer (Figure 2a-3), assuming vertical raindrops in a windless scene.…”

Subjective and Objective De-Raining Quality Assessment Towards Authentic Rain Image

“…8, we show the de-raining results produced from two rain images. In the first row, the deep learning based method, i.e., Fu17 [14], performs well in removing the directional rain streaks and preserving the contours of window and balcony, whose MOS could reach 97.97. By contrast, the dictionary learning based method, such as, Kang12 [10], clearly over smoothes the original image structure, whose MOS is only 37.63.…”

“…By contrast, the dictionary learning based method, such as, Kang12 [10], clearly over smoothes the original image structure, whose MOS is only 37.63. When we change the Ding16 [6] Kang12 [9] Luo15 [10] Li16 [11] Deng17 [20] Fu17 [13] rain image to the second row, it is seen that Fu17 [14] almost does nothing for the dot-like raindrops, whose MOS drops to 40.95. While, Kang12 [10] could perfectly remove these small raindrops without obvious damage to the contour of the player, whose MOS rises to 99.68.…”

“…It works efficiently in estimating rain streaks, but also easily mistakes striped background, whose texture is similar to the rain streaks. To better distinguish rain streaks from background, many researchers propose to simultaneously learn the priors for rain and background layers via dictionary learning, Gaussian mixture model, deep neural network and so on [10]- [14]. These data-driven methods deliver great perform for the rain images whose appearance features are covered by the training samples.…”