A driver's emotional state is integral to road safety. Unhealthy emotions such as anger, sadness, and fear influence a driver's judgment particularly under bad weather conditions. Independent Component Analysis may be analyzed and used to extract features of a driver's emotion from images and a Support Vector Machine is used to classify a driver's unhealthy emotional expressions. Images of a driver's expression under real-time driving conditions are acquired with a Logitech C905 CCD camera attached to a Lenovo ThinkPad T41 computer. Training and testing sample sets of emotional expression images are preprocessed through color image cutting, scale normalization, and grey processing, and standardized gray images of 108 x120 pixels are acquired. Six cross-validation experiments for recognizing a driver's angry emotion are carried out and the average recognition rate is 93.3%. Therefore, theory analysis and the cross-validation results indicate that the Independent Component Analysis and Support Vector Machine are effective in recognizing a driver's unhealthy emotional expression.
Recognition of the Driver's Unhealthy Facial Expression Based on Independent Component Analysis and Support Vector Machine
11th International Conference of Chinese Transportation Professionals (ICCTP) ; 2011 ; Nanjing, China
ICCTP 2011 ; 1745-1750
2011-07-26
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Features classification using support vector machine for a facial expression recognition system
British Library Online Contents | 2012
|Real Time Facial Expression Recognition from Image Sequences using Support Vector Machines
British Library Conference Proceedings | 2005
|