利用局部特征的子空间车辆识别算法

Subspace vehicle recognition algorithm using local features

利用改进的主成分分析(Principal Component Analysis,PCA)方法,通过研究不同的车辆特征(如全局特征、各种局部特征)对静态图像车辆识别效果的影响,提出了一种新的静态图像车辆识别算法。该算法可有效降低光照和背景噪声对识别的影响,实现对存在部分遮挡的车辆检测。实验结果表明,该算法具有良好的鲁棒性和车辆识别率。

Utilize the method of principal components analysis to research the influence to the recognition result caused by different vehicle features（such as global feature,various kinds of local features）,a new vehicle recognition algorithm is proposed.The proposed algorithm can reduce the influence of lighting conditions and background noise effectively and detect partially occluded vehicles accurately.Testing results demonstrate that by using the proposed algorithm the vehicle detection can be realized with a strong robusticity and high identification ratio.