A triple-branch network for latent fingerprint enhancement guided by orientation fields and minutiae

TL;DR

This work addresses latent fingerprint enhancement by acknowledging region-specific restoration needs and proposing a Triple-Branch Spatial Fusion Network (TBSFNet) to apply tailored enhancement across high-quality, low-quality, and background regions within an encoder–decoder framework. Building on this, the authors add orientation-field and minutiae-guidance modules to create MLFGNet, improving generalization and fixation on critical fingerprint features. Experimental results on the MUST and MOLF datasets show that MLFGNet outperforms existing methods in both visual enhancement and identification accuracy, with notable gains over TBSFNet and OFFIENet. The approach offers a practical path to more reliable latent fingerprint restoration and downstream matching in forensic contexts, by combining region-aware enhancement with global and local fingerprint cues.

Abstract

Latent fingerprint enhancement is a critical step in the process of latent fingerprint identification. Existing deep learning-based enhancement methods still fall short of practical application requirements, particularly in restoring low-quality fingerprint regions. Recognizing that different regions of latent fingerprints require distinct enhancement strategies, we propose a Triple Branch Spatial Fusion Network (TBSFNet), which simultaneously enhances different regions of the image using tailored strategies. Furthermore, to improve the generalization capability of the network, we integrate orientation field and minutiae-related modules into TBSFNet and introduce a Multi-Level Feature Guidance Network (MLFGNet). Experimental results on the MOLF and MUST datasets demonstrate that MLFGNet outperforms existing enhancement algorithms.

Figures (7)

• Figure 1: Architecture of proposed TBSFNet network. The input is a grayscale image with height H and width W.
• Figure 2: Framework of TBSF and DBSA blocks.
• Figure 3: Framework of MLFGNet network.
• Figure 4: Framework of MSFF_O block.
• Figure 5: Framework of TBSF_O block.