| With the rapid development of social economy and technology, in recent years the number of domestic vehicle is even more significant growth, the growt h rate is far more than the parking lot construction speed. This also leads to more and more people park their cars on busy or narrow road, brought great inconvenience to people and vehicles in the vicinity, but also more prone to traffic jams or accidents. In order to solve this problem, we design a system of illegal parking detection which is for the purpose to detect illegal parking, based on color image processing.This system is mainly composed by IPC, HD video camera, computer. This system can be delineated illegal area, and monitored the illegal area 24 hours. The camera acquisition image information and the IPC analysis the image, judge the behavior of vehicles. If the vehicles break the traffic rules, the video image information will be saved, and further through the background to identify illegal ve hicle license plate number. Finally through the computer network transmission to the background, as the punishment evidence.The system extracted the background with the improved Surendra algorithm. And applied new segmentation method based on fuzzy color extraction to segmenting and positioning new target. When judging the new target was vehicle, recording the position information of vehicle, according to its location information to judge whether the vehicle was illegal. If the vehicle was illegal, the system use the method of fuzzy color extraction to segmented the plate. Finally enhancing the positioning plate and filtering out irrelevant characteristics with character, the character can accurately identified when matched with template.This article adopts Visual C++ 6.0 to make programming. Experimental results show that the proposed theory and method has higher accuracy and computational efficiency. The system makes a significant improvement on efficiency of traffic management. |
PDF Full Text Request