Adaptive river flow measurement method based on spatiotemporal image velocimetry and optical flow

Abstract

This paper proposes an adaptive river discharge measurement method based on spatiotemporal image velocimetry (STIV) and optical flow to solve the problem of blurred texture features and limited measurement accuracy under complex natural environmental conditions. Optical flow tracking generates spatiotemporal images by following the flow mainstream direction of rivers with both regular and irregular natural banks. A texture similarity function filtering method effectively enhances spatiotemporal texture features. The proposed method is applied to a natural river, with measurement results from a propeller-type current meter used as truth values. It is evaluated and compared with three other methods regarding measurement accuracy, error, and other evaluation indices. The results demonstrate that the method significantly improves spatiotemporal image quality. Its estimation outcomes perform better across all evaluation metrics, enhancing the adaptability and accuracy of the flow measurement method. HIGHLIGHTS Automatically selecting velocity lines based on optical flow information from the river surface.; The STIs generated by the velocity line obtained by the optical flow method have a clearer texture.; Texture filtering algorithm enhances the texture features in the STI effectively.; Eliminating the influence of river surface interference information on STI.;

Keywords

• filtering technology
• image analysis
• optical flow
• real-time measurement
• river flow measurement
• spatiotemporal image velocimetry (stiv)