Novel BSSSO-Based Deep Convolutional Neural Network for Face Recognition with Multiple Disturbing Environments

Face recognition technology is presenting exciting opportunities, but its performance gets degraded because of several factors, like pose variation, partial occlusion, expression, illumination, biased data, etc. This paper proposes a novel bird search-based shuffled shepherd optimization algorithm (BSSSO), meta-heuristic technique motivated by the intuition animals and social behavior birds, for improving face recognition. The main intention behind research to establish an optimization-driven deep learning approach recognizing images with multiple disturbing environments. developed model undergoes three steps, namely, (a) Noise Removal, (b) Feature Extraction, (c) Recognition. For removal noise, type II fuzzy system cuckoo search (T2FCS) used. feature extraction carried out using CNN, landmark enabled 3D morphable (L3DMM) utilized efficiently fit from single uncontrolled image. obtained features are subjected Deep CNN recognition, wherein training performed BSSSO. experimental findings on standard datasets (LFW, UMB-DB, Extended Yale B database) prove ability proposed over existing approaches.