Video Acquisition Of Transportation And Identity Recognition

Video acquisition of transportation and License Plate Recognition are the vital technology which used in the road management system. With the development of auto industry, more and more researcher devote in improving algorithm for this technology. The main research in this essay can be divided into four parts: video acquisiting, license plate locating, character segmentation, character recognition of vehicle license plate.This system is based on Cortex-A8 embedded processor, and use V4L2 interface to capture video streaming, running the improved algorithm based on OpenCV to complete license plate locating and recognition, the results of the improvement is acceptable.For the algorithm of license plate locating, this essay proposed color statistics with edge detection based on the future color. For the algorithm of character segmentation, a hybrid of connected domain analysis and projection histogram segmentation is proposed. The process first execute the algorithm of connected component labeling in the region of license plate, eliminate those connected-components according to the information of their area, width and height of the minimum bounding rectangle, after noise and obviously impossible regions are removed, projection histogram segmentation can accurately completed. Finally, K Nearest Neighbor algorithm is used for license plate character recognition. The experiment shows high recognition rate.