A Comparative Study of Nighttime Object Detection With Datasets From Australia and China

Qadar, Muhammad Ali; Nie, Chang; Tan, Ying; Zhang, Hui; Sun, and Zhifeng

doi:10.1109/cac51589.2020.9327278

Cited by 3 publications

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dataset is comprised of over 100 k videos obtained from vehicle-mounted sensors. The BDD dataset has been used by many researchers [53,[70][71][72] to develop object detection algorithms and image enhancement for night-time and low-light applications, as it is one of the biggest open source datasets that cover diversity in classes, scenes and time of day. A further reason why BDD was chosen for this study is that it contains night-time and dusk/dawn data, whereas many of the datasets mentioned above do not offer this data.…”

Section: Datasets and Data Imbalancementioning

confidence: 99%

A Study on Data Selection for Object Detection in Various Lighting Conditions for Autonomous Vehicles

Lin,

Parsi,

Mullins

et al. 2024

J. Imaging

View full text Add to dashboard Cite

In recent years, significant advances have been made in the development of Advanced Driver Assistance Systems (ADAS) and other technology for autonomous vehicles. Automated object detection is a crucial component of autonomous driving; however, there are still known issues that affect its performance. For automotive applications, object detection algorithms are required to perform at a high standard in all lighting conditions; however, a major problem for object detection is poor performance in low-light conditions due to objects being less visible. This study considers the impact of training data composition on object detection performance in low-light conditions. In particular, this study evaluates the effect of different combinations of images of outdoor scenes, from different times of day, on the performance of deep neural networks, and considers the different challenges encountered during the training of a neural network. Through experiments with a widely used public database, as well as a number of commonly used object detection architectures, we show that more robust performance can be obtained with an appropriate balance of classes and illumination levels in the training data. The results also highlight the potential of adding images obtained in dusk and dawn conditions for improving object detection performance in day and night.

show abstract

Section: Datasets and Data Imbalancementioning

confidence: 99%