: With increasing technology to improve driving security, surrounding camera is increasing popular among recent models of family using vehicles. With abundant information collected by these cameras, there are few existing techniques that can automatically analyze and understand the content of recording and providing valuable information about behavior of driver and driving conditions. This paper explores the possibility of using combination of image processing and computer vision techniques to extract information about the driving behavior. The paper is collaborated with computer vision research. The data used in this paper is collected in urban streets of the cities. Along with the videos, there are also sensor data for acceleration of x,y,z directions for training of the model. Keywords:Detection,Driving Conditions,Image Recognition..