摘要
论文中提出了将计算机双目视觉技术应用于人群密度估计方法:首先根据目标在两个摄像平面上的视差的平均值,计算出该位置的矫正参数,然后根据目标在不同位置的矫正参数,拟合出矫正函数。以前景像素、前景边缘像素等图像特征作为人群密度估计的特征,应用矫正参数进行矫正。实验证明,相对于现有方法,该方法可以消除射影畸形的影响,大大提高特征的有效性,从而提高人群密度估计的准确性。
In this paper,a new method is proposed to estimate crowd density based on binocular vision： First,according to the average parallax of the object in the two camera plane,the correction parameters of monitoring location are calculate,then the correction function is fittied based on the result of the correction parameters of the same target objects in different locations.Take foreground pixels,foreground edge pixels of image as feature characteristics of crowd density estimation,application of correction parameters to correct.Experiments show that compared to existing methods,this method can greatly improve the effectiveness of the characteristics,thereby enhancing the accuracy of the estimation of crowd density.
出处
《计算机与数字工程》
2012年第8期116-118,共3页
Computer & Digital Engineering
基金
广东省自然科学基金项目(编号:10151802904000013)
深圳市科技计划项目(编号:JC200903180648A)
深圳市科技计划项目(编号:JC201006020811A)资助
关键词
人群密度估计
双目视觉
射影畸形
crowd density estimation
binocular camera
projective deformity
