Recognizing geochemical spatial patterns using deformable convolutional networks guided with geological knowledge

Abstract. This study tackles the limited quantification of irregular spatial geochemical patterns and weak interpretability in deep learning models in geochemical anomaly recognition. We propose a hybrid approach that that integrates geological knowledge (GK) into deformable convolutional networks (DCN), creating a model termed GK_DCN, with the aim of enhancing both the performance and transparency of geochemical anomaly recognition. This model introduces learnable parameters that allow the convolutional kernels to adaptively adjust their sampling locations, enabling them to more accurately capture complex, irregular geochemical anomaly patterns caused by mineralization. To enhance geological consistency, ore-controlling fault are incorporated as geological knowledge constraints, guiding the network to prioritize spatial correlations between deposits and faults. Experimental results in southern Tianshan Au-Cu polymetallic ore district demonstrate that the GK_DCN significantly enhances the accuracy and reliability of geochemical anomaly recognition verified across multiple evaluation metrics, producing more distinct spatial anomalous patterns and higher consistency with known mineral deposits by adaptively adjusting the receptive field. Visualization of the kernel offsets revealed the model's superior adaptive spatial sampling mechanism. Furthermore, using Grad-CAM to generate feature significance heatmaps highlighted the key features the model focused on during geochemical anomaly recognition, significantly improving interpretability and proving effectiveness in capturing complex geochemical patterns. This work provides an effective intelligent method for geochemical pattern recognition and offers a reference for interpretable deep learning in geochemical exploration through multi-angle visualization.