论文部分内容阅读
针对矿物浮选过程中泡沫图像分割参数选取问题,提出了形态模式谱计算方法,从统计学角度证明了最优结构元素的存在性.首先在二值图像粒度测定理论基础上,引入形态模式谱概念,获取连通区域的尺寸与形状分布模式谱二维特征矩阵,使得该过程本身不需要选择结构元素.根据形态模式谱矩阵计算其方差值,并求最小值得到最优结构元素.实验结果表明:本方法能充分利用浮选泡沫图像信息,有效地为泡沫图像分割过程提供最优结构元素,提高浮选泡沫图像分割准确性.
Aiming at the selection of foam image segmentation parameters in mineral flotation process, a new method for calculating the morphological spectrum is proposed, and the existence of the optimal structure element is proved from the statistical point of view.Firstly, based on the binary image particle size determination theory, Concept, to obtain the size and shape of the connected region of the two-dimensional feature matrix spectrum distribution model, so that the process itself does not need to select structural elements according to the morphological mode matrix matrix to calculate the variance value and find the minimum optimal structural elements. It shows that this method can make full use of flotation foam image information and effectively provide optimal structural elements for the image segmentation of foam and improve the accuracy of flotation image segmentation.