Abstract: The spatial positioning accuracy of ocean bottom nodes (OBNs) is a critical factor in ensuring the quality of seismic data imaging. However, existing secondary positioning methods suffer from either insufficient robustness due to their reliance on first-break picking accuracy or low efficiency stemming from high computational costs, thus failing to meet the demands of high-precision and high-efficiency seismic exploration. To address these issues, a coordinate scanning secondary positioning method based on the principle of minimizing the standard deviation of cross-correlation time differences was proposed. This method worked by constructing a scanning grid within the neighborhood of the node’s initial coordinates. For each grid point, linear normal moveout (NMO) correction was applied, followed by cross-correlation with a simulated wavelet to extract a sequence of inter-trace relative time differences. The standard deviation of this sequence served as a physical measure of waveform event coherence, with the minimum value indicating the optimal node location. Application to actual OBN data demonstrates that the method effectively circumvents the dependency on first-break picking, exhibiting superior robustness and computational efficiency. After correction, the moveout-corrected gather shows significantly improved flatness, and the stacked section displays enhanced event continuity. This successfully suppresses imaging artifacts caused by positioning errors, validating the method’s effectiveness in improving both positioning accuracy and imaging quality and indicating its broad potential for engineering applications.