Fast single-image dehazing based filtering and morphology

Xiaozhou Lei; Zixiang Fei; Minrui Fei

doi:10.1117/12.3032627

13 September 2024 Fast single-image dehazing based filtering and morphology

Xiaozhou Lei, Zixiang Fei, Minrui Fei

Proceedings Volume 13178, Eleventh International Symposium on Precision Mechanical Measurements; 1317816 (2024) https://doi.org/10.1117/12.3032627
Event: Eleventh International Symposium on Precision Mechanical Measurements, 2023, Guangzhou, China

Abstract

In the process of industrial measurement, the phenomenon of water mist appearing in front of the camera lens due to cold environment can cause fogging of the captured image and greatly interfere with the measurement process, affecting the accuracy of measurement. Correspondingly, solving the problem of image fogging has become particularly important. Therefore, based on the atmospheric scattering model, this paper proposes a fast single-image dehazing method. Specifically, for the estimation of atmospheric light, considering that the distribution of atmospheric light in the image scene is inconsistent, a fast estimation algorithm of atmospheric light map in the scene is constructed by using mean filtering and fast guided filtering methods. Then, dark channels, morphological erosion algorithm and weighted fusion are used to construct an estimation algorithm for the transmission map. Finally, we use linear normalization to readjust the numerical range of the dehazed image to improve the visual effect. Experimental results show that the proposed method successfully restored the image scene after dehazing, and has excellent visual effects. In addition, the algorithm runs very fast and has very low time consumption.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Xiaozhou Lei, Zixiang Fei, and Minrui Fei "Fast single-image dehazing based filtering and morphology", Proc. SPIE 13178, Eleventh International Symposium on Precision Mechanical Measurements, 1317816 (13 September 2024); https://doi.org/10.1117/12.3032627

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available