【摘要】 复杂场景下的车牌定位算法由于环境的变化导致车牌识别率低、计算效率不高。在对Faster R-CNN算法分析的基础上,结合MEAN-SHIFT聚类算法的特点,提出了一种基于MEAN-SHIFT特征增强的车牌定位方法。该方法采用并行计算的方式,通过增强目标区域的特征,有效提升了复杂场景下车牌定位的效率和准确度。实验表明,该方法能够在多种复杂场景下快速定位车牌照区域,准确率高,具有较好的鲁棒性。更多还原

【Abstract】 The license plate localization algorithm under complex scenarios has a low license plate recognition rate due to changes in the environment, and the calculation efficiency is not high. Based on the analysis of Faster R-CNN algorithm and the characteristics of MEAN-SHIFT clustering algorithm, this paper proposes a method of vehicle license plate positioning based on matures to effectively improve the efficiency of fast and accurate positioning of license plates in complex scenes. MEAN-SHIFT feature enhancement is a method uses parallel computing to enhance the target area. The fear experiments show that this method can quickly locate the license plate area in a variety of complex scenarios, with high accuracy and good robustness.更多还原