DocumentCode :
3220462
Title :
Morphological drowsy detection
Author :
Tanha, Mehrdad ; Seifoory, Hossein
Author_Institution :
Res. Inst. for Appl. Phys. & Astron., Tabriz Univ., Tabriz, Iran
fYear :
2011
fDate :
16-18 Nov. 2011
Firstpage :
63
Lastpage :
65
Abstract :
Drowsiness detection is vital in preventing accidents. Eye state analysis - Detecting whether the eye is open or closed - is critical step for drowsiness detection. In this article we proposed an innovative algorithm for detecting eyes in drowsiness detection which is based on morphology. Morphological processes are one of the powerful instruments for analysing images even with low quality, so it can be possible to actualize this method even with one Megapixel webcams. This method has best performance in dimness conditions. We analyse our eye state analysis algorithm via 5 sequences of frames in video for 10 different intervals between face and camera and show superior results compared to the common technique based on distance between eyelids that are distance depended.
Keywords :
accident prevention; behavioural sciences; eye; image sequences; object detection; road accidents; road safety; traffic engineering computing; video signal processing; accident prevention; dimness conditions; drowsiness detection; eye state analysis algorithm; eyelids; image quality; megapixel Webcams; morphological drowsy detection; video camera; video frame sequences; Accuracy; Cameras; Face; Monitoring; Sensors; Shape; Vehicles; Morphological eyeball detection; distances independency; drowsiness; facial features;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal and Image Processing Applications (ICSIPA), 2011 IEEE International Conference on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4577-0243-3
Type :
conf
DOI :
10.1109/ICSIPA.2011.6144070
Filename :
6144070
Link To Document :
بازگشت