arrow
返回

An adaptive optic-physic based dust removal method using optimized air-light and transfer function

delete2022-04-21
delete8
PRE
AI
A
Ako Bartani
A
Alireza Abdollahpouri
M
Mohsen Ramezani
F
Fardin Akhlaghian Tab *
DOI:10.1007/s11042-022-13109-2delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
In recent years, computer vision is used in different applications. In these applications, images may be contaminated by different factors such as clouds, shadow, haze, fog, and dust phenomena. The dust phenomenon caused environmental problems in certain regions of the world. This phenomenon, which reduced visibility, acts as a barrier against the object's light reflection to the eyes or camera lens. There are few studies relating to removing dust from images, while haze removal has received more attention. The haze removal methods have been examined for removing dust that haven't proper performance. Thus, we propose a new physic-optic based method to remove dust from images properly. Here, the air-light estimation on the dark channel histogram is modified and a new method is used to estimate the weight of the dust-based transmission function per pixel in the R, G, and B channels, separately. Experimental results indicate that the proposed method appropriately removes dust from the images which contain different density of dust. It should be noted that output images have high contrast, good brightness level, and natural colors. Moreover, the proposed method can be used for removing haze from hazy images as a suitable method that achieves better or acceptable results than popular and recent haze removal methods. Experiments indicate that the PCQI value of the proposed method for two dusty datasets is about 1.24 on average, while this value for the best state-of-the-art method is about 1.15 on average. These values for hazy images are 1.33, and 1.15, respectively.
Keyword:
Image enhancement
Dust removal
Physical-optic model
Transmission function
Air light estimation

期刊

Multimedia Tools and Applications 封面图
Multimedia Tools and Applications
IF:
3
论文数:
2.0W
被引数:
3.2W

机构

U
University of Kurdistan
学者数:
2.1K
论文数: 2.1K
被引数: 2.5K
引用论文

引用论文

Mycoplasma-associated Polyarthritis in a Reticulated Giraffe
err2003-01-01
err0
errOAAI
errElizabeth E. Hammond; Craig A. Miller; Loyd Sneed; Robin W. Radcliffe
err分享
err收藏
A Patch-Structure Representation Method for Quality Assessment of Contrast Changed Images
err2015-12-01
err303
PREAI
errWang, Shiqi; Ma, Kede; Yeganeh, Hojatollah; Wang, Zhou; Lin, Weisi
err分享
err收藏
An airlight estimation method for image dehazing based on gray projection
err2020-07-23
err5
PREAI
errWang, Wencheng; Yuan, Xiaohui; Wu, Xiaojin; Dong, Yihua
err分享
err收藏
Haze removal for a single visible remote sensing image
err2017-08-01
err76
PREAI
errLiu, Qi; Gao, Xinbo; He, Lihuo; Lu, Wen
err分享
err收藏
The formation of the building blocks of peptides on interstellar dust grains
err2020-10-12
err0
errOAAI
errN. F. W. Ligterink; J. Terwisscha van Scheltinga; V. Kofman; V. Taquet; S. Cazaux; J. K. Jørgensen; E. F. van Dishoeck; H. Linnartz
err分享
err收藏
学者 查看更多内容