Author(s): Hairong Gao; Zihan Liu; Yu Han

Linked Author(s):

Keywords: Eagle-eye bionic vision; Channel flow field identification; Intelligent flow measurement equipment

Abstract: The monitoring of channel flow is a crucial aspect of quantitative water resource management. Based on the analysis of the hydrodynamic mechanism of ecological river channels, this manuscript establishes a cloud-edge integrated platform for intelligent perception of hydrological information. It constructs an Eagle-eye bionic vision intelligent flow measurement model and equipment suitable for multi-level canal networks, achieving real-time perception of water information and quantified representation of the entire flow field. The Eagle-eye bionic vision intelligent flow measurement equipment integrates a monocular camera and a binocular camera. The monocular camera is utilized to capture surface flow images of the channel. Global surface flow velocity is inferred through dense optical flow analysis based on these images. The binocular camera recognizes the waterline based on the deep learning algorithm, constructs the triangular model of water level, measuring point and distance measurement, and senses the depth of parallax in binocular stereo sense, and monitors the water level information of the canal system in real time. Through binocular stereo vision, it perceives the disparity depth, enabling real-time monitoring of water level information in the channel system. Further integration of monocular flow measurement and binocular water depth recognition technologies, combined with logarithmic laws, allows for the derivation of cross-sectional average flow velocity. The constructed Eagle-eye biomimetic visual intelligent flow measurement equipment technology can promote the deep integration of water conservancy business and information technology, provide reference for optimizing water resource allocation in irrigation areas, and promote the automation, intelligence, and high-quality development of water measurement equipment in irrigation areas.

DOI: https://doi.org/10.64697/HIC2024_P535

Year: 2024