A Haze Removal Method via The Fusion of Gaussian Low-Frequency Multi-Scale and Median Rank Detail Perspective Network

다국어 초록 (Multilingual Abstract)

Images captured outside in bad weather are polluted by colloidal particles and atmospheric droplets. These acquired pictures are the source of major mistakes in digital image vision systems because they are prone to low contrast, poor visibility, and color distortion. As a result, defogging research is important for real-world applications. This study uses a fog image degradation model to define picture defogging as a mathematical inversion and image restoration procedure. The global atmospheric light A and transmittance can be precisely estimated by combining a Gaussian low-frequency multi-scale convolutional network with a median rank detail perspective network (GLFM-MRDP Net). A Gaussian low-frequency multi-scale convolutional network is first used to obtain the low-frequency part of the image, and then the block convolution model is adopted to acquire accurate A. Then the feature extraction subnet and median rank detail optimized network is adopted to obtain transmittance, which can suppress image noise while retaining details as much as possible and extracting features through convolution. Comparative experimental findings demonstrate that our approach is successful in handling dense fog, complicated scenes, and multi-scale pictures. Besides, the no-reference and full-reference evaluation metrics of our method are superior to methods[8, 11, 17, 22]. As a result, our technology outperforms four other cutting-edge defogging techniques in terms of aesthetic effect, applicability, and running speed.

참고문헌 (Reference)

1 V. Stipetic, "Variational Formulation of Dark Channel Prior for Single Image Dehazing" 64 (64): 845-854, 2022

2 F. Chen, "Two-step modulus-based matrix splitting iteration methods for retinex problem" 88 (88): 1989-2005, 2021

3 K. He, "Single image haze removal using dark channel prior" 1956-1963, 2009

4 T. H. Yu, "Single image dehazing based on multi-scale segmentation and deep learning" 33 (33): 2022

5 B. Ganguly, "Single Image Haze Removal With Haze Map Optimization for Various Haze Concentrations" 32 (32): 286-301, 2022

6 S. -S. Lin, "Separation and contrast enhancement of overlapping cast shadow components using polarization" 14 (14): 7099-7108, 2006

7 C. Hillar, "Revisiting Perceptual Distortion for Natural Images: Mean Discrete Structural Similarity Index" 241-249, 2017

8 M. -J. Seow, "Ratio rule and homomorphic filter for enhancement of digital colour image" 69 (69): 954-958, 2006

9 Y. Y. Schechner, "Polarization-based vision through haze" 42 (42): 511-525, 2003

10 J. Liang, "Polarimetric dehazing method for dense haze removal based on distribution analysis of angle of polarization" 23 (23): 26146-26157, 2015

11 H. Fu, "Perception Oriented Haze Image Definition Restoration by Basing on Physical Optics Model" 10 (10): 1-16, 2018

12 R. Roy, "Optimization of Stego Image Retaining Secret Information Using Genetic Algorithm with 8-connected PSNR" 60 (60): 468-477, 2015

13 B. Jiang, "Novel multi-scale retinex with color restoration on graphics processing unit" 10 (10): 239-253, 2015

14 D. Berman, "Non-local Image Dehazing" 1674-1682, 2016

15 S. Dippel, "Multiscale contrast enhancement for radiographies : Laplacian pyramid versus fast wavelet transform" 21 (21): 343-353, 2002

16 Y. F. Liu, "MFID-Net: Multi-scaled feature-fused image dehazing via dynamic weights" 78 : 2023

17 X. Luo, "LCDA-Net: Efficient Image Dehazing with Contrast-Regularized and Dilated Attention" 55 (55): 11467-11488, 2023

18 S. G. Narasimhan, "Interactive (De)Weathering of an Image using Physical Models" 1-8, 2003

19 L. Liu, "Image segmentation based on gray stretch and threshold algorithm" 126 (126): 626-629, 2015

20 D. Sandic-Stankovic, "Image quality assessment based on pyramid decomposition and mean squared error" 740-743, 2015

21 X. J. Guo, "Image dehazing via enhancement, restoration, and fusion : A survey" 86-87 : 146-170, 2022

22 S. Fang, "Image dehazing using polarization effects of objects and airlight" 22 (22): 19523-19537, 2014

23 A. Horé, "Image Quality Metrics: PSNR vs. SSIM" 2366-2369, 2010

24 K. He, "Guided Image Filtering" 35 (35): 1397-1409, 2013

25 A. Morgand, "Generic and real-time detection of specular reflections in images" 274-282, 2014

26 J.-P. Tarel, "Fast visibility restoration from a single color or gray level image" 2201-2208, 2009

27 W. Zhang, "Fast polarimetric dehazing method for visibility enhancement in HSI colour space" 19 (19): 2017

28 Y. Park, "Fast Execution Schemes for Dark-Channel-Prior-Based Outdoor Video Dehazing" 6 : 10003-10014, 2018

29 U. Hari, "Dark and Bright Channel Priors for Haze Removal in Day and Night Images" 34 (34): 957-967, 2022

30 S. Y. Huang, "An end-to-end dehazing network with transitional convolution layer" 31 (31): 1603-1623, 2020

31 L. Mutimbu, "A relaxed factorial Markov random field for colour and depth estimation from a single foggy image" 355-359, 2013

32 X. Zhao, "A modified prior-based single-image dehazing method" 16 (16): 1481-1488, 2022

33 A. F. M. Raffei, "A low lighting or contrast ratio visible iris recognition using iso-contrast limited adaptive histogram equalization" 74 (74): 40-48, 2015

34 D. -A. Van, "A Review of Characteristics, Causes, and Formation Mechanisms of Haze in Southeast Asia" 8 (8): 201-220, 2022

35 X. R. Liu, "A Hybrid Retinex-Based Algorithm for UAV-Taken Image Enhancement" E104.D (E104.D): 2024-2027, 2021