IEEE Transactions on Pattern Analysis and Machine Intelligence

Single image haze removal is a challenging ill-posed problem. Existing methods use various constraints/priors to get plausible dehazing solutions. The key to achieve haze removal is to estimate a medium transmission map for an input hazy image. In this paper, we propose a trainable end-to-end system called DehazeNet, for medium transmission estimation. DehazeNet takes a hazy image as input, and outputs its medium transmission map that is subsequently used to recover a haze-free image via atmospheric scattering model. DehazeNet adopts convolutional neural network-based deep architecture, whose layers are specially designed to embody the established assumptions/priors in image dehazing. Specifically, the layers of Maxout units are used for feature extraction, which can generate almost all haze-relevant features. We also propose a novel nonlinear activation function in DehazeNet, called bilateral rectified linear unit, which is able to improve the quality of recovered haze-free image. We establish connections between the components of the proposed DehazeNet and those used in existing methods. Experiments on benchmark images show that DehazeNet achieves superior performance over existing methods, yet keeps efficient and easy to use.

/pdf/dehazenet-an-end-to-end-system-for-single-image-haze-removal-2xm00icsmx.pdf

DehazeNet: An End-to-End System for Single Image Haze Removal

Single image haze removal has been a challenging problem due to its ill-posed nature. In this paper, we propose a simple but powerful color attenuation prior for haze removal from a single input hazy image. By creating a linear model for modeling the scene depth of the hazy image under this novel prior and learning the parameters of the model with a supervised learning method, the depth information can be well recovered. With the depth map of the hazy image, we can easily estimate the transmission and restore the scene radiance via the atmospheric scattering model, and thus effectively remove the haze from a single image. Experimental results show that the proposed approach outperforms state-of-the-art haze removal algorithms in terms of both efficiency and the dehazing effect.

/pdf/a-fast-single-image-haze-removal-algorithm-using-color-4bbfozoymz.pdf

A Fast Single Image Haze Removal Algorithm Using Color Attenuation Prior

Medical Image Computing and Computer-Assisted Intervention

The performance of existing image dehazing methods is limited by hand-designed features, such as the dark channel, color disparity and maximum contrast, with complex fusion schemes. In this paper, we propose a multi-scale deep neural network for single-image dehazing by learning the mapping between hazy images and their corresponding transmission maps. The proposed algorithm consists of a coarse-scale net which predicts a holistic transmission map based on the entire image, and a fine-scale net which refines results locally. To train the multi-scale deep network, we synthesize a dataset comprised of hazy images and corresponding transmission maps based on the NYU Depth dataset. Extensive experiments demonstrate that the proposed algorithm performs favorably against the state-of-the-art methods on both synthetic and real-world images in terms of quality and speed.

/pdf/single-image-dehazing-via-multi-scale-convolutional-neural-3bsvlws9do.pdf

Single Image Dehazing via Multi-scale Convolutional Neural Networks

Respiratory monitoring is widely used in clinical and healthcare practice to detect abnormal cardiopulmonary function during ordinary and routine activities. There are several approaches to estimate respiratory rate, including accelerometer(s) worn on the torso that are capable of sensing the inclination changes due to breathing. In this article, we present an adaptive band-pass filtering method combined with principal component analysis to derive the respiratory rate from three-dimensional acceleration data, using a body sensor network platform previously developed by us. In situ experiments with 12 subjects indicated that our method was capable of offering dynamic respiration rate estimation during various body activities such as sitting, walking, running, and sleeping. The experimental studies also suggested that our frequency spectrum-based method was more robust, resilient to motion artifact, and therefore outperformed those algorithms primarily based on spatial acceleration information.

Estimation of Respiration Rate from Three-Dimensional Acceleration Data Based on Body Sensor Network

In this paper, we propose a simple but powerful prior, color attenuation prior, for haze removal from a single input hazy image By creating a linear model for modelling the scene depth of the hazy image under this novel prior and learning the parameters of the model by using a supervised learning method, the depth information can be well recovered With the depth map of the hazy image, we can easily remove haze from a single image Experimental results show that the proposed approach is highly efficient and it outperforms state-of-the-art haze removal algorithms in terms of the dehazing effect as well

/pdf/single-image-dehazing-using-color-attenuation-prior-4ywy7ndxqk.pdf

Single image dehazing using color attenuation prior

At present, ultrasound is one of the essential tools for noninvasive medical diagnosis. However, speckle noise is inherent in medical ultrasound images and it is the cause for decreased resolution and contrast-to-noise ratio. Low image quality is an obstacle for effective feature extraction, recognition, analysis, and edge detection; it also affects image interpretation by doctor and the accuracy of computer-assisted diagnostic techniques. Thus, speckle reduction is significant and critical step in pre-processing of ultrasound images. Many speckle reduction techniques have been studied by researchers, but to date there is no comprehensive method that takes all the constraints into consideration. In this paper we discuss seven filters, namely Lee, Frost, Median, Speckle Reduction Anisotropic Diffusion (SRAD), Perona-Malik's Anisotropic Diffusion (PMAD) filter, Speckle Reduction Bilateral Filter (SRBF) and Speckle Reduction filter based on soft thresholding in the Wavelet transform. A comparative study of these filters has been made in terms of preserving the features and edges as well as effectiveness of de-noising.We computed five established evaluation metrics in order to determine which despeckling algorithm is most effective and optimal for real-time implementation. In addition, the experimental results have been demonstrated by filtered images and statistical data table.

Evaluation of various speckle reduction filters on medical ultrasound images

With the development of digital multimedia technologies, image matting has gained increasing interests from both academic and industrial communities. The purpose of image matting is to precisely extract the foreground objects with arbitrary shapes from an image or a video frame for further editing. It is generally known that image matting is inherently an ill-posed problem because we need to output three images out of only one input image. In this paper, we provide a comprehensive survey of the existing image matting algorithms and evaluate their performance. In addition to the blue screen matting, we systematically divide all existing natural image matting methods into four categories: 1) color sampling-based; 2) propagation-based; 3) combination of sampling-based and propagation-based; and 4) learning-based approaches. Sampling-based methods assume that the foreground and background colors of an unknown pixel can be explicitly estimated by examining nearby pixels. Propagation-based methods are instead based on the assumption that foreground and background colors are locally smooth. Learning-based methods treat the matting process as a supervised or semisupervised learning problem. Via the learning process, users can construct a linear or nonlinear model between the alpha mattes and the image colors using a training set to estimate the alpha matte of an unknown pixel without any assumption about the characteristics of the testing image. With three benchmark data sets, the various matting algorithms are evaluated and compared using several metrics to demonstrate the strengths and weaknesses of each method both quantitatively and qualitatively. Finally, we conclude this paper by outlining the research trends and suggesting a number of promising directions for future development.

Qingsong Zhu

Papers

A Fast Single Image Haze Removal Algorithm Using Color Attenuation Prior

Estimation of Respiration Rate from Three-Dimensional Acceleration Data Based on Body Sensor Network

Single image dehazing using color attenuation prior

Evaluation of various speckle reduction filters on medical ultrasound images

Targeting Accurate Object Extraction From an Image: A Comprehensive Study of Natural Image Matting