Heterogeneous bad weather image enhancement using colour correction

Abstract

Listen

Translate article

A clear and bright atmosphere is much needed when humans use imaging equipment outside. Bad weather conditions due to haze and fog cause the resulting image to fade and blur. This is because the colour in the image has undergone a condition known as colour fading. The method currently used can be divided into 2 main methods, the first is the use of multiple images and the second is the use of a single image. However, the first method requires more than one image to be taken which causes the image improvement process to increase. For the second method, only one image is required. However, the challenge in this method is the determination of the thin layer on the image to determine the element of interference that causes the image to fade and blur. Various methods are used for this single image method. However, there is still an error that arises which is a change in the main structure of the image namely colour and tone. The same goes for the halo effect that arises as a result of this improvement process. In this study, the use of D-Hazy, FRIDA, and FRIDA2 datasets. This study uses the method of interweaving the colour thinning to solve the problem of image interference that causes fading and blurring. This method maintains the main structure of the image and prevents the occurrence of halo effects by evaluating using the Structural Similarity Index (SSI) method. The degree of image recovery was assessed based on enhanced colour series improvement using the Colourfulness Index method. The performance analysis shows that the proposed method can be used to enhance the heterogenous bad weather image.