TY - GEN
T1 - Low-Light Image Enhancement for Autonomous Driving Systems using DriveRetinex-Net
AU - Pham, Long Hoang
AU - Tran, Duong Nguyen Ngoc
AU - Jeon, Jae Wook
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/11/1
Y1 - 2020/11/1
N2 - Most autonomous driving algorithms are designed for normal-light images. Hence, insufficient lighting during image capture significantly degrades the visibility of images and hurts the performance of many computer vision systems. Retinex theory is an effective tool for enhancing the illumination and detail of images. In this paper, we collected a Low-Light Drive (LOL-Drive) dataset and applied a deep retinex neural network, named DriveRetinex, which was taught using this dataset. The deep Retinex-Net consists of two subnetworks: Decom-Net (decomposes a color image into a reflectance map and an illumination map) and Enhance-Net (enhances the light level in the illumination map). The whole architecture can be trained in an end-to-end fashion. Extensive experiments demonstrate that the proposed method not only achieves visually appealing low-light enhancement, but it also increases the accuracy of object detection in autonomous driving systems.
AB - Most autonomous driving algorithms are designed for normal-light images. Hence, insufficient lighting during image capture significantly degrades the visibility of images and hurts the performance of many computer vision systems. Retinex theory is an effective tool for enhancing the illumination and detail of images. In this paper, we collected a Low-Light Drive (LOL-Drive) dataset and applied a deep retinex neural network, named DriveRetinex, which was taught using this dataset. The deep Retinex-Net consists of two subnetworks: Decom-Net (decomposes a color image into a reflectance map and an illumination map) and Enhance-Net (enhances the light level in the illumination map). The whole architecture can be trained in an end-to-end fashion. Extensive experiments demonstrate that the proposed method not only achieves visually appealing low-light enhancement, but it also increases the accuracy of object detection in autonomous driving systems.
KW - autonomous driving
KW - image processing
KW - low-light image enhancement
KW - retinex theory
UR - https://www.scopus.com/pages/publications/85098880866
U2 - 10.1109/ICCE-Asia49877.2020.9277442
DO - 10.1109/ICCE-Asia49877.2020.9277442
M3 - Conference contribution
AN - SCOPUS:85098880866
T3 - 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020
BT - 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 IEEE International Conference on Consumer Electronics - Asia, ICCE-Asia 2020
Y2 - 1 November 2020 through 3 November 2020
ER -