Identifying weak geochemical anomalies from background is vital in the exploration of complex metallogenic systems. The window-based local singularity analysis (LSA) method is widely used to quantitatively characterize the local singularity structural properties of nonlinear geological processes. In this study, a batch method for local singularity mapping based on quadtree is proposed to improve the spatial retrieval efficiency in batch process with various parameters. Compared to conventional spatial retrieval methods, the quadtree-based local singularity analysis method can significantly enhance retrieval efficiency and reduce calculation time, especially in the batch process. In the case study, the quadtree-based LSA is further applied to characterize Au mineralization associated with geochemical anomalies at Baishilazi, in the province of Heilongjiang, China. A total of seven factors related to window-based LSA including window shape, orientation, aspect ratio, window initial size, step size, maximum window size and average concentration calculation method are considered, and the optimal parameter combination is determined based on receiver operating characteristic curves (ROC) and area under the ROC curves (AUC). The exploration targets are further delineated using the t test. The results show that the quadtree-based LSA is an efficient tool for optimal parameter determination in the batch process.