Spatial normalization of facial thermal images using facial landmarks

Abstract Human–computer interaction (HCI) is an for mutual communication between humans and computers. HCI needs to recognize the human state quantitatively in real-time. Although it possible evaluate condition by measuring biological signals, challenge that often requires physical constraints. There increasing interest a non-contact method of estimating physiological psychological states facial skin temperature using infrared thermography. However, due individual differences face shape, accuracy estimation thermal images was sometimes low. To solve this problem, we hypothesized spatial normalization image (SN-FTI) could reduce effect shape. The objective study develop SN-FTI on states. First, attempted features. results suggested would result same shape among individuals. Since there are distribution, inter-individual correlation coefficient be lower than intra-individual coefficient. Next, modeled estimated drowsiness level SN-FTIs compared with Normal. showed slightly improved discrimination rate level. were structure