A Review of Methods Employed for Forensic Human Identification

Forensic human identification is of utmost importance in various contexts, including criminal investigations and disaster victim identification. This research paper explores the application of biometric identification techniques in forensic human identification, with a particular focus on fingerprint recognition, face recognition, hand geometry, voice recognition, vein recognition, DNA matching, retina recognition, iris recognition, ECG-based identification, and EEG-based identification. The paper provides an overview of the historical background and development of EEG-based identification, highlighting its significance in forensic investigations. Additionally, the study investigates the integration of machine learning approaches, including support vector machines (SVM), K-nearest neighbors (KNN), random forests, and deep learning. The paper also covers RNN and its architectures styles in particularly long short-term memory (LSTM), and gated recurrent unit (GRU) in forensic human identification. The paper discusses the advantages and limitations of these machine-learning techniques in enhancing the accuracy and efficiency of identification systems. Overall, this research paper sheds light on the importance of biometric identification and the potential of machine-learning approaches in forensic human identification, providing valuable insights for forensic experts, researchers, and practitioners in the field.