| A lot of accidents result from tiredness drive, and eye tiredness monitoring can reduce fatal car crashes, the enormous pain and the social economic loss. The system of distinguishing eye tiredness has a number of key factors which include eye tiredness recognition, eye detection and a condition of the eye. Active contour model is a hot research in computer vision, and it is very suited to be used in edge detection . The key technology and the theorys in this area are studied, and a software system about recognizing eye tiredness is developed.A method to detect them using ballon snake model based on pretreatment in a picture with human eyes is presented. Firstly, we can detect edge in differential algorithm and mark objects and find out location of eyes according to geometric properties and construct eyes area , and the initial contour of eyes can be obtained automatically . Then, we can converge contour of eyes using ballon snake model based on pretreatment. Finally, combined with mathematical morphology, the contour of eyeballs can be obtained using snake model again. Experiments showed that the method has advantages of robust and small impact about light effect .This paper puts forward a method combinated integral projection and snake model to identify human eye tiredness. It uses snake model to detect the eyes , computes the contour of a person's face to locate the position of the eyes and squares the position of the face to level the eyes. Then, vertical gradation integral projection of the region around eye is calculated, and the result is a binary waveform picture of vertical gradation integral projection. The last , the values of the waveform's three characteristics are calculated and they cab be used to recognize if the eye is in an open or closed state. Based on this, this paper also presents a scrolling window method to calculate the frame rate of the eye blinking which indentifies the eyes tiredness status. This method is not only simple, effective and fast, but at the same time, it can also be applied in the inclination of the human eye.Finally, this paper provides the architecture of eye tiredness system and the modules about the system are realized and a software system about how to indentify the eyes tiredness status is developed. Experiments showed that all of the methods are effective. |
