Color-restoring and energy-saving imaging monitoring for intelligent offshore engineering

• Published in: Ocean engineering Vol. 311; p. 118951
• Main Authors: Quan, Xiangqian, Chen, Xiangzi, Wei, Yucong, Li, Zizheng, Li, Yun, Yan, Peng
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2024
• Subjects: Imaging data; Machine intelligence; Maintenance of offshore structures; Unmanned monitoring; Vision system of auto underwater vehicle; Imaging data; Unmanned monitoring; Machine intelligence; Maintenance of offshore structures; Vision system of auto underwater vehicle
• ISSN: 0029-8018
• DOI: 10.1016/j.oceaneng.2024.118951

Images play a crucial role in artificial intelligence (AI) for offshore engineering, which provide a large amount of data for machine intelligence to ensure the safe operation and maintenance of offshore structures. However, due to the attenuation of sunlight in water or monitoring at nighttime, underwater imaging monitoring systems need a light source, which consumes a lot of energy and impacts the load capacity of an offshore monitoring system. In addition, due to the nonlinearity of attenuation at different wavelengths, underwater images generally suffer from color deviation, tending to be blue and green, which decreases the accuracy of machine learning and intelligent recognition in unmanned offshore engineering. In order to reduce energy consumption and enhance color restoration, we proposed and designed a three-detector underwater imaging system (TD&TP-UIS) based on a prism, in which three detectors were used to enhance the utilization efficiency of light, and a gated band coating was designed on the prism to enhance the color reproduction. A series of experiments were carried out in water tanks, pools, Qiandao Lake, and other places to verify the advantage of color restoration and energy saving using the TD&TP-UIS. In this paper, we firstly put forward the color-restoring and energy-saving imaging monitoring system with three detectors based on triple-prism (TD&TP-UIS) in the word, which could enhance the load capacity and machine intelligent accuracy of offshore platforms and auto underwater vehicle in offshore intelligent engineering remarkably. The advantages of this underwater imaging system are shown as below:•Three detectors combined with triple-prism are adopted to improve the light utilization efficiency for long time monitoring.•By designing coatings on the triple-prism, the waveband was gated to improve the color restoration and accuracy of machine intelligent.