An efficient and effective pore matching method using ResCNN descriptor and local outliers

Feng Liu, Qiuheng Wang, Yanfeng Xiao, Linlin Shen

Abstract: Highlights•Efficient pore descriptor generation using ResCNN, capturing both static and dynamic information.•Improved AdaLAM algorithm for fast and accurate outlier removal and matching.•Different strategies for partial and full fingerprints to enhance overall performance.•High matching speed and accuracy, demonstrated through training on an in-house dataset and testing on PolyU-HRF.