论文部分内容阅读
以数学形态学滤波和模糊推理为基础,提出一种基于彩色视频图像处理的小区监视及人数统计方法。首先确定监视区范围,然后使用形态学运算将差减后离散相邻的部分连在一起以确定人数,并在未进行形态学运算的差减图像中找到头顶、头宽,同时确定肤色、头发像素的百分比以及头部中央的灰度平均值,以肤色所占百分比和灰度平均值为模糊推理的前提条件进行模糊决策。实验结果表明,该方法能很准确地完成人数统计。
Based on mathematical morphological filtering and fuzzy inference, this paper proposes a cell monitoring and population counting method based on color video image processing. First, determine the scope of the surveillance area, and then use morphological operations to discretize the adjacent discrete parts together to determine the number of people, and morphological operations in the subtraction image to find the top, head width, while determining the color, hair The percentage of pixels and the average of the head of the gray, the percentage of skin color and the average gray-scale fuzzy reasoning conditions for fuzzy decision-making. Experimental results show that this method can accurately complete the number of people.