Research On Visual-Based Road And Lane Identifying Of Intelligent Vehicle Navigation

Intelligent vehicle(IV) is a mobile robot, which can run autonomously and continuously in real-time indoor and outdoor environment. Its research has become the tactic research object of high technology of all countries. With the promotion of urgent demand on military, civilian and research fields, intelligent vehicle are changing rapidly. Of all the IV key technologies, vision-based navigation is a difficult and comprehensive subject that involves almost every aspect of computer vision researches. The fundamental tasks of visual-based navigation are identifying the road region and the lane correctly according to the image information. Due to the complexity and inconstancy of outdoor environment and huge information of colorful image, it is difficult to satisfy the real-time requirement and robustness for vision-based system. In order to overcome these shortages, two pivotal technologies, such as road identifying, lane identifying, are researched thoroughly and some novel and effective algorithms are proposed in this dissertation.Firstly, a real-time and robust road-identifying algorithm is presented, which is based on vision character Stylebook. Considering the complexity of the outdoor road environment, it is very difficult to develop a versatile algorithm that can handle every case. The algorithm begins the work from the road character analysis, and makes a principle for road character selecting. By analyzing different color models, it chooses normalized RGB value to present the road color. In this dissertation, eight parameters ,normalized RGB values r|-,g|-,b|-, their standard deviation respectivelyδ_r,δ_g,δ_b, and the average gray value and gray standard deviationδ₁, is presented as the road's character parameter. These parameters can efficiently eliminate the color changes which are caused by the light intensity changes outdoors. Thus, the number of stylebook is reduced. On the road region identification, based on stylebook characters, the algorithm uses the regional growth methods to identify the road region. It can identify the road region effectively and correctly. Also, in dissertation the definition of road is studied from the perspective of computer vision research.Secondly, a new algorithm based on character stylebook are presented aimed at the problem of the current complicated algorithm and vast calculate. The new algorithm extracts lane characters, improves the region identifying algorithm, realizes that road and lane identifying was completed in same process, reduces the steps in lane identifying, and improves the system efficiency. After lane identification, the algorithm could achieve the lane vector quantization, by using skeleton line extraction, pre- vector quantization and sub-straight-line method, etc. The algorithm is simple, piratical and efficient.The algorithm has advantages as follows, which is different from other identifying algorithms: It has fewer processing steps. Computing is simple. And it has a wider versatility. The algorithm satisfies the vision-based navigation system's requirements on versatility, real-time and robust. The algorithm has been verified during research.