基于CCD的待车数传感器研究--等待车辆图像的腐蚀-膨胀识别算法

城市交通系统是一个复杂系统，为实现城市汽车流实时响应控制，交通信息的实时采集和处理显得尤为重要。针对现有车辆检测器需对道路进行破坏性改造等问题，研究一种基于单幅、静态图像处理的CCD路口等待车辆数传感器；提出在对图像连行有效的预处理的基础上，采用腐蚀-膨胀识别算法，使系统实时检测出交通路口的等待车辆数，并以此作为城市交通信号控制系统实时决策的状态信息。通过示例证明了算法的实时性、健壮性及自动化的应用前景。

Study on sensors for quantity of waiting vehicles based on CCD -corrosion-expansion identifying method for image of waiting vehicles

Urban traffic system is a complex system. In order to realize the real time response control of urban automobile flow, real time gathering and processing of traffic information seems to be even more important. Aimed at the problems that the now existing vehicle detector needs a destructiveness reformation of the road, a kind of sensor for the quantity of waiting vehicles at the crossroads based on the CCD of single width and static image processing was being studied. On the basis of carrying out effective pre-processing on the image and of adopting the corrosion-expansion identifying method, to let the system be able to make a real time detection upon the quantity of waiting vehicles at the crossroads, and taking this detection as the status information for the real time decision of the urban traffic signal controlling system. By means of a typical example this paper proved the real time character and robustness of the algorithm and prospects of automatic application.