TY - GEN
T1 - Underwater image enhancement and detection based on convolutional DCP and YOLOv5
AU - Liu, Guodong
AU - Feng, Lihui
AU - Lu, Jihua
AU - Yan, Lei
N1 - Publisher Copyright:
© 2022 Technical Committee on Control Theory, Chinese Association of Automation.
PY - 2022/1/1
Y1 - 2022/1/1
N2 - Underwater image restoration is conducive to better underwater resource detection and information effective trans-mission. However, the light in the complex water body diffusely reflective and the selection absorption of different band light results in blurring and color distortion of underwater image. Therefore, we propose a convolutional Dark Channel Prior (DCP) underwater image recovery algorithm to enhance underwater images. It can do the pre-processing work for the subsequent YOLOv5 object recognition. The enhancement algorithm first performs Commission International Eclairage Lab (CIELAB) equalization of underwater images for color distortion correction. Meanwhile, the underwater image formation parameters are estimated by the minimum convolution region DCP. Then, Contrast Limited Adaptive Histogram Equalization (CLAHE) is per-formed to obtain an enhanced underwater image. Finally, the enhanced underwater image is input to the YOLOv5 model for detection. Experimental results show that the proposed method outperforms state-of-art algorithms in terms of image recovery effect, evaluation quality and detection accuracy.
AB - Underwater image restoration is conducive to better underwater resource detection and information effective trans-mission. However, the light in the complex water body diffusely reflective and the selection absorption of different band light results in blurring and color distortion of underwater image. Therefore, we propose a convolutional Dark Channel Prior (DCP) underwater image recovery algorithm to enhance underwater images. It can do the pre-processing work for the subsequent YOLOv5 object recognition. The enhancement algorithm first performs Commission International Eclairage Lab (CIELAB) equalization of underwater images for color distortion correction. Meanwhile, the underwater image formation parameters are estimated by the minimum convolution region DCP. Then, Contrast Limited Adaptive Histogram Equalization (CLAHE) is per-formed to obtain an enhanced underwater image. Finally, the enhanced underwater image is input to the YOLOv5 model for detection. Experimental results show that the proposed method outperforms state-of-art algorithms in terms of image recovery effect, evaluation quality and detection accuracy.
KW - color equalization
KW - image classification
KW - peak signal-to-noise ratio
KW - underwater color image quality evaluation
UR - http://www.scopus.com/inward/record.url?scp=85140453539&partnerID=8YFLogxK
U2 - 10.23919/CCC55666.2022.9902814
DO - 10.23919/CCC55666.2022.9902814
M3 - Conference contribution
AN - SCOPUS:85140453539
T3 - Chinese Control Conference, CCC
SP - 6765
EP - 6772
BT - Proceedings of the 41st Chinese Control Conference, CCC 2022
A2 - Li, Zhijun
A2 - Sun, Jian
PB - IEEE Computer Society
T2 - 41st Chinese Control Conference, CCC 2022
Y2 - 25 July 2022 through 27 July 2022
ER -