Loading…
Enhancement and dehazing of images for real time applications
Simple image dehazing techniques based on prior or deep learning techniques may not be effective in certain cases due to the computation challenges of real environments and inherent bugs in the available data. In such cases, a robust combined contrast enhancement with liable fusion framework can imp...
Saved in:
Main Authors: | , , |
---|---|
Format: | Conference Proceeding |
Language: | English |
Subjects: | |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Simple image dehazing techniques based on prior or deep learning techniques may not be effective in certain cases due to the computation challenges of real environments and inherent bugs in the available data. In such cases, a robust combined contrast enhancement with liable fusion framework can improve the quality of an image. This frame-work improves the image visibility and increases the contrast to reduce the haze in an image. To accomplish this, the contrast enhancement stage involves pre-processing steps that utilize Gamma Correction to adjust the local-visibility of a hazy image. Additionally, Color-Retaining with Adaptive Histogram Equalization (CRAHE) is used to improve the global contrast of the input hazy image, addressing the difficulty of individually applying adaptive histogram equalization to each colour channel. The fusion stage entails creating a quick structural patch decomposition-based fusion technique with an adjustable kernel size to fuse the inputs produced by CRAHE and Gamma Correction after the image has been enhanced. These methods are particularly useful for night time images with haze. CRAHE also has potential applications in image retouching photos and improving the lighting. Therefore, dehazing involves restoring a clean image from a degraded input image with foggy scenes. |
---|---|
ISSN: | 0094-243X 1551-7616 |
DOI: | 10.1063/5.0212582 |