Real-time adaptive gray threshold measurement in extracting traffic parameters

In order to extract traffic parameters effectively in all day time an adaptive gray threshold traffic parameters measurement (AGTTPM) system is built in this paper. In AGTTPM. a novel real-time adaptive gray threshold (RAGT) measurement method is presented to overcome the over sensitivity of vision based system to illumination changing. RAGT is a method which could highly enhance the detection accuracy by means of adjusting the gray threshold accorthng to the surrounding brightness automatically. A restricted Laplacian sharpening algorithm is used in image preprocessing for image enhancement. Then relying on the RAGT measurement, the traffic parameters can be obtained in real time and the corresponding errors can also he inhibited obviously. The implementation of system hardware and software is introduced by steps and RAGT is introduced in the part of software design. Experimental results show the improvements in detection precision and real-time performance.