Abstract
We present a system that utilizes a range of image processing algorithms to allow fully automated thermal face analysis under both laboratory and real-world conditions. We implement methods for face detection, facial landmark detection, face frontalization and analysis, combining all of these into a fully automated workflow. The system is fully modular and allows implementing own additional algorithms for improved performance or specialized tasks. Our suggested pipeline contains a histogtam of oriented gradients support vector machine (HOG-SVM) based face detector and different landmark detecion methods implemented using feature-based active appearance models, deep alignment networks and a deep shape regression network. Face frontalization is achieved by utilizing piecewise affine transformations. For the final analysis, we present an emotion recognition system that utilizes HOG features and a random forest classifier and a respiratory rate analysis module that computes average temperatures from an automatically detected region of interest. Results show that our combined system achieves a performance which is comparable to current stand-alone state-of-the-art methods for thermal face and landmark datection and a classification accuracy of 65.75% for four basic emotions.
Highlights
The thermal signature of objects yields information that cannot be obtained in the visual spectrum.In recent years, advances in thermal image sensor design have led to a number of commercially available imaging systems that allow analyzing thermal images with resonable effort
As described in the previos chapter, the tasks required for thermal face analysis are face detection, facial landmark and region of interest (ROI) tracking and the analysis itself, we provide dedicated subsections for these tasks
Curve decay does not occur until an Intersection over Union (IoU) of 0.7 which corresponds to a very good overall stability
Summary
The thermal signature of objects yields information that cannot be obtained in the visual spectrum. Advances in thermal image sensor design have led to a number of commercially available imaging systems that allow analyzing thermal images with resonable effort. Thermal or long-wave infrared (LWIR) imaging has gained increasing attention as imaging modality for analysis of both human and non-human recordings. The thermal signature of humans has been subject to numerous scientific studies. The human face has been in the focus of many of these studies. A specific region of interest (ROI) in the face is analyzed. If video sequences are analyzed, it is crucial that this ROI remains stable in order to minimize motion artifacts
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
