Novel Multimodal Biometric Feature Extraction for Precise Human Identification

In recent years, biometric sensors are applicable for identifying important individual information and accessing the control using various identifiers by including characteristics like a fingerprint, palm print, iris recognition, so on. However, precise identification of human features is still physically challenging in humans during their lifetime resulting variance appearance or features. response to these challenges, novel Multimodal Biometric Feature Extraction (MBFE) model proposed extract from noisy sensor data modified Ranking-based Deep Convolution Neural Network (RDCNN). The MBFE enables feature extraction different images that includes iris, lip, where preprocessed initially further processing. extracted validated after optimal RDCNN splitting datasets train then testing with sets input images. simulation performed matlab test efficacy modal over multi-modal result shows method achieves increased accuracy, precision, recall, F1 score than existing deep learning methods. performance improvement Algorithm technique terms attained 0.126%, 0.152%, 0.184%, 0.38% Back Propagation (BPNN), Human Identification Using Wavelet Transform (HIUWT), Segmentation Methodology Non-cooperative Recognition (SMNR), Daugman Iris Localization (DILA) techniques respectively.