Next Article in Journal
Clip-On IMU System for Assessing Age-Related Changes in Hand Functions
Next Article in Special Issue
Automatic vs. Human Recognition of Pain Intensity from Facial Expression on the X-ITE Pain Database
Previous Article in Journal
Calculation of the Overlap Function and Associated Error of an Elastic Lidar or a Ceilometer: Cross-Comparison with a Cooperative Overlap-Corrected System
Previous Article in Special Issue
Irradiance Restoration Based Shadow Compensation Approach for High Resolution Multispectral Satellite Remote Sensing Images
Letter

Gated Dehazing Network via Least Square Adversarial Learning

Department of Image, Chung-Ang University, Seoul 06974, Korea
*
Author to whom correspondence should be addressed.
Sensors 2020, 20(21), 6311; https://doi.org/10.3390/s20216311
Received: 28 September 2020 / Revised: 30 October 2020 / Accepted: 3 November 2020 / Published: 5 November 2020
(This article belongs to the Special Issue Sensors and Deep Learning for Digital Image Processing)
In a hazy environment, visibility is reduced and objects are difficult to identify. For this reason, many dehazing techniques have been proposed to remove the haze. Especially, in the case of the atmospheric scattering model estimation-based method, there is a problem of distortion when inaccurate models are estimated. We present a novel residual-based dehazing network model to overcome the performance limitation in an atmospheric scattering model-based method. More specifically, the proposed model adopted the gate fusion network that generates the dehazed results using a residual operator. To further reduce the divergence between the clean and dehazed images, the proposed discriminator distinguishes dehazed results and clean images, and then reduces the statistical difference via adversarial learning. To verify each element of the proposed model, we hierarchically performed the haze removal process in an ablation study. Experimental results show that the proposed method outperformed state-of-the-art approaches in terms of peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), international commission on illumination cie delta e 2000 (CIEDE2000), and mean squared error (MSE). It also gives subjectively high-quality images without color distortion or undesired artifacts for both synthetic and real-world hazy images. View Full-Text
Keywords: haze removal; generative adversarial network; gated structure haze removal; generative adversarial network; gated structure
Show Figures

Figure 1

MDPI and ACS Style

Ha, E.; Shin, J.; Paik, J. Gated Dehazing Network via Least Square Adversarial Learning. Sensors 2020, 20, 6311. https://doi.org/10.3390/s20216311

AMA Style

Ha E, Shin J, Paik J. Gated Dehazing Network via Least Square Adversarial Learning. Sensors. 2020; 20(21):6311. https://doi.org/10.3390/s20216311

Chicago/Turabian Style

Ha, Eunjae, Joongchol Shin, and Joonki Paik. 2020. "Gated Dehazing Network via Least Square Adversarial Learning" Sensors 20, no. 21: 6311. https://doi.org/10.3390/s20216311

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop