2022
DOI: 10.3390/s22145210
|View full text |Cite
|
Sign up to set email alerts
|

Foggy Lane Dataset Synthesized from Monocular Images for Lane Detection Algorithms

Abstract: Accurate lane detection is an essential function of dynamic traffic perception. Though deep learning (DL) based methods have been widely applied to lane detection tasks, such models rarely achieve sufficient accuracy in low-light weather conditions. To improve the model accuracy in foggy conditions, a new approach was proposed based on monocular depth prediction and an atmospheric scattering model to generate fog artificially. We applied our method to the existing CULane dataset collected in clear weather and … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
1
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 7 publications
(3 citation statements)
references
References 51 publications
0
1
0
Order By: Relevance
“…Similar approaches have been used to augment existing datasets for scenes such as rainy [ 13 ] days and nighttime [ 14 ]. For example, the literature [ 15 ] discusses the use of synthetic images for lane line detection, while previous studies have investigated synthetic fog images for automated driving, including FRIDA [ 16 ], FRIDA2 [ 17 ], Foggy Cityscape [ 18 ], and Multifog KITTI [ 19 ]; few studies have quantitatively evaluated artificially generated fog image enhancement datasets to determine their fidelity. Tarel et al constructed digital image-based synthetic outdoor fog datasets, namely FRIDA (Foggy Road Image Database) and FRIDA2, using software, but these datasets contain a limited number of images with low resolution and poor visual quality and are therefore insufficient to support the training of target detection models with higher accuracy and on a larger scale, especially in fog scenes.…”
Section: Introductionmentioning
confidence: 99%
“…Similar approaches have been used to augment existing datasets for scenes such as rainy [ 13 ] days and nighttime [ 14 ]. For example, the literature [ 15 ] discusses the use of synthetic images for lane line detection, while previous studies have investigated synthetic fog images for automated driving, including FRIDA [ 16 ], FRIDA2 [ 17 ], Foggy Cityscape [ 18 ], and Multifog KITTI [ 19 ]; few studies have quantitatively evaluated artificially generated fog image enhancement datasets to determine their fidelity. Tarel et al constructed digital image-based synthetic outdoor fog datasets, namely FRIDA (Foggy Road Image Database) and FRIDA2, using software, but these datasets contain a limited number of images with low resolution and poor visual quality and are therefore insufficient to support the training of target detection models with higher accuracy and on a larger scale, especially in fog scenes.…”
Section: Introductionmentioning
confidence: 99%
“…Additionally, efforts have been made to generate datasets with complex weather conditions, aiming to increase the diversity of available data. For instance, [14] and [15] successfully synthesized foggy and blurry images from data originally collected under clear daytime conditions. Similar attempts have been made to transform daytime images into nighttime representations [16].…”
Section: Introductionmentioning
confidence: 99%
“…Synthetic images can be easily manipulated to simulate various conditions, such as lighting, orientation, background, etc., that might not be readily available in real-world data. This enhances the ability of the model to generalize under different conditions[60][61][62][63]. •…”
mentioning
confidence: 97%