Signatures verification based on PNN classifier optimised by PSO algorithm
作者：
Highlights：
• We propose a new dynamic signature verification method based on composed features.
• Each feature is paired with a similarity measure and form a composed feature.
• Optimal set of composed features is determined in the Hotelling reduction process.
• Chosen composed features are used as input data for a classifier.
• PNN classifier parameters are tuned by means of the Particle Swarm Optimization (PSO) algorithm.
摘要
Highlights•We propose a new dynamic signature verification method based on composed features.•Each feature is paired with a similarity measure and form a composed feature.•Optimal set of composed features is determined in the Hotelling reduction process.•Chosen composed features are used as input data for a classifier.•PNN classifier parameters are tuned by means of the Particle Swarm Optimization (PSO) algorithm.
论文关键词：Hotelling's statistics,Classification,PNN,PSO,Signature verification
论文评审过程：Received 28 February 2015, Revised 22 April 2016, Accepted 30 June 2016, Available online 2 July 2016, Version of Record 30 July 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.06.032
