Ge, Hancheng (2011-08). Determination of the Presence Conditions of Pavement Markings using Image Processing. Master's Thesis. Thesis uri icon

abstract

  • Pavement markings, as a form of traffic control devices, play a crucial role in safely guiding drivers. Restriping pavement markings is an important task in the maintenance of traffic control devices. Every year state agencies spend a lot of money in maintaining pavement markings as the retroreflectivity or durability values of the markings fall below a minimum level. Currently, the most widely adopted method used to determine the presence conditions of pavement markings is by expert observation, a subjective technique that may not provide consistent and convinced results for agencies. Hence, a fast and accurate way to determine the presence conditions of pavement markings can lead to significant cost savings while ensuring driving safety. In this study, a systematic approach that can automatically determine the presence conditions of pavement markings using digital image processing techniques is presented. These techniques are used to correct the geometric deformity, detect colors of pavement markings, segment images, enhance images, detect edge lines of ideal pavement markings, and recognize the features of pavement markings appearing in the photographs. To better implement the aforementioned techniques, a software package has been developed by Graphic User Interface (GUI) as a platform to simultaneously evaluate the presence conditions of single or multiple pavement markings. The developed software package is able to do operations such as open files, calibrate camera calibration, clip, rotation, histogram display, and detection of edge lines of ideal pavement markings. The above system was tested and evaluated with the photograph datasets provided by the NTPEP Mississippi test deck. The empirical results (when compared with the manual method and expert observation) show that the developed system in this study is accurate and reliable. Additionally, the interactivity of the developed software package is satisfactory due to the feedback from ten volunteers. It is also concluded that the developed system, as an important reference, potentially helps agencies make a better decision in the maintenance of pavement markings with more accurate and speedy evaluation of the presence conditions of pavement markings.
  • Pavement markings, as a form of traffic control devices, play a crucial role in safely guiding drivers. Restriping pavement markings is an important task in the maintenance of traffic control devices. Every year state agencies spend a lot of money in maintaining pavement markings as the retroreflectivity or durability values of the markings fall below a minimum level. Currently, the most widely adopted method used to determine the presence conditions of pavement markings is by expert observation, a subjective technique that may not provide consistent and convinced results for agencies. Hence, a fast and accurate way to determine the presence conditions of pavement markings can lead to significant cost savings while ensuring driving safety.

    In this study, a systematic approach that can automatically determine the presence conditions of pavement markings using digital image processing techniques is presented. These techniques are used to correct the geometric deformity, detect colors of pavement markings, segment images, enhance images, detect edge lines of ideal pavement markings, and recognize the features of pavement markings appearing in the photographs. To better implement the aforementioned techniques, a software package has been developed by Graphic User Interface (GUI) as a platform to simultaneously evaluate the presence conditions of single or multiple pavement markings. The developed software package is able to do operations such as open files, calibrate camera calibration, clip, rotation, histogram display, and detection of edge lines of ideal pavement markings. The above system was tested and evaluated with the photograph datasets provided by the NTPEP Mississippi test deck. The empirical results (when compared with the manual method and expert observation) show that the developed system in this study is accurate and reliable. Additionally, the interactivity of the developed software package is satisfactory due to the feedback from ten volunteers. It is also concluded that the developed system, as an important reference, potentially helps agencies make a better decision in the maintenance of pavement markings with more accurate and speedy evaluation of the presence conditions of pavement markings.

publication date

  • August 2011