基于LabVIEW Vision的车牌识别系统设计

文中设计了一种基于LabVIEW Vision的车牌识别系统。该系统根据车牌颜色在HSL色彩空间上的特征,采用色彩阈值分割、腐蚀等图像处理方法将车辆图像转化成二值图,并在二值图中进行边缘检测,再根据检测到的车牌与背景间的边缘线的信息组建方程组,求出车牌边缘特征点的坐标,以确定车牌在图像中的位置。然后,以车牌在图像中的位置是否在指定区域内作为是否进行车牌识别以及当前车辆图像是否保存为车牌识别输入图像的触发条件。如果条件不满足,则系统将重新采集车辆图像,否则会根据车牌位置抓捕车牌,并采用均匀性度量法等方法对抓捕到的车牌进行二次处理,并将其转化成车牌二值图,再由OCR函数对该图进行字符识别。经实验验证,该系统具有较高的识别精确度与稳定性。

Design of License Plate Recognition System Based on LabVIEW Vision

A LabVIEW Vision-based license plate recognition system is designed. The system converts vehicle images into binary maps using image processing methods such as color threshold segmentation and erosion based on the characteristics of license plate color on HSL color space. Edge detection is performed in the binary map, and the coordinates of the edge feature points of the license plate are solved by forming a system of equations based on the information of the detected edge line between the license plate and the background to determine the location of the license plate in the image. Whether the position of the license plate in the image is within the specified position is used as the trigger condition for whether license plate recognition is performed and whether the current vehicle image is saved as the input image for license plate recognition. If the condition is not satisfied, the system will re-capture the vehicle image. On the contrary, the license plate will be captured according to the license plate position and the secondary image processing will be performed on the captured license plate using the uniformity metric method and other methods to transform it into a binary map of the license plate, and then the OCR function will perform character recognition on the map. It is experimentally verified that the system has high recognition accuracy and stability.