| Current, complete and accurate roadway asset inventory is vital for data reporting, data analysis and decision support of daily management of road maintenance departments. The acquisition of roadway geometric dimensions,3D coordinates and text information of roadway facilities via the technology of machine vision has some advantages like great safety, high efficiency, less human influences and so on, and thus becomes one of the research focuses in the field of Intelligent Transportation Systems (ITS) in recent years.The hardware design, key algorithms, database management and other relevant techniques of a machine vision-based 3D information acquisition and processing system of roadway facilities are studied intensively in this thesis. The main contents are as follows:(1) A dual-computer-based concurrent data acquisition system is designed, in which four environment area-scan cameras, a pavement surface line-scan camera, a milometer, a GPS receiver, a ground-penetrating radar and other acquisition devices are connected with two industrial control computers. Also, acquisition tasks are reasonably distributed on both the computers, each of which communicates with the other through a 1000 Mbps network.(2) The binocular measuring system of roadway facilities consists of two environment area-scan cameras. A planar calibration object is used to calibrate the left camera as well as the right camera, and then the system parameters are calculated. Finally, with all the parameters of the binocular measuring system, measurement of roadway facilities in the common view of both the cameras is implemented on the basis of the disparity principle.(3) According to the requirement of the comprehensive information data management of roadway facilities, the conceptual data model, physical data model and object-oriented model of the comprehensive information database are designed with the application of standard database design methods, and then the database is implemented on the selected DBMS platform.Tests in the laboratory environment and applications in the field reveal the stability, reliability and convenience of the system, showing that the 3D information measurement precision of roadway facilities can meet the needs of practical engineering applications. |