基于多特征集成学习的景区人群密度估计

Density Estimation of Scenic Spots Based on Multi Features Ensemble Learning

对景区等公共区域进行人群密度估计对于保障人群安全和社会稳定具有重要意义。在景区中由于光照变化、相机高度角度变化以及行人遮挡的问题,现有的方法难以做出准确的估计。为此提出了一种结合支持向量机回归(SVR)进行集成学习的方法来进行人群密度估计。方法首先以人的头部宽度作为参照,对场景进行多层次的分块;然后采用第一层SVR模型,对从子图像块提取出的三种特征进行粗预测,将预测结果作为新的特征,并对其使用第二层SVR模型进行细预测,将所有子图像的预测结果相加;最后根据不同景区场景设定的人数分级进行密度估计。实验结果表明,方法在景区多个场景分类准确率达到85%以上,是一种有效且在类似场景有较强扩展性的人群密度估计算法。

For the crowd safety and social stability, ifs important to take crowd density estimation on public are-as ,such as scenic spots. Due to illumination changes, different camera height and angles and the pedestrian occlu-sion, it？s difficult to make an accurate estimation with the existing methods. A method based on ensemble learning with support vector regression is proposed to estimate the crowd density. First, the scene is divided to multiple lev-els patches using the head width as a reference, then the first layer support vector regression mode is used to make coarse prediction on three feature descriptors extracted on the image patches. The prediction results are used as a new feature and to make fine prediction with the second layer support vector regression mode. At last, the crowd density is estimated with the sum of all patches prediction results and the classification standard which is set accord-ing to the characteristics of the scene. Experimental results show that the method proposed can achieve the classifi-cation accuracy above 85% on many scenes of the scenic spots, and it＇s an effective and robust crowd density esti-mation algorithm.