Thermal Signature Recognition Using Spider Monkey-Inspired Optimization Algorithm

Thermal signature recognition is a promising research area with numerous applications in forensics, secure access, and surveillance. Unlike other classification tasks, thermal signature recognition often involves many classes, and conventional subsystems are typically unsuitable for this type of problem due to their lower validation accuracy. We introduce a method that leverages the Spider Monkey-inspired Optimization (SMO) algorithm for clustering and recognizing thermal signatures, employing Advanced Principal Component Analysis (APCA) for feature reduction and achieving an impressive accuracy of 98.91 % in our Python implementation, evaluated through sensitivity, specificity, and accuracy metrics.