基于光照建模的改进分水岭锌浮选泡沫图像分割
首发时间:2017-06-02
摘要:浮选泡沫表层的气泡结构特征是表征浮选工况的重要指标,但有效的气泡结构表征依赖于准确的泡沫图像分割。有色金属的浮选泡沫有其特异性,多金属优先浮选的泡沫图像气泡高亮点难以有效提取。本文提出一种基于光照建模的标记分水岭算法,首先估计出气泡顶部图像和光照分布图像;接着以二者的差值为拓扑图像求取标记图像;最后进行分水岭变换。结果表明,该算法能有效解决多金属及矿泥杂质共存的泡沫图像可靠分割问题。
For information in English, please click here
Improved Watershed based on Illuminance Modeling for Zinc Froth Image Segmentation
Abstract:The surface bubble structure is an important indicator for froth flotation performance, but the effective bubble structure characterizing relies on accuracy bubble segmentation. The bubble images in the nonferrous metallurgy flotationhave some specificities, it is hard to extract the highlight region of bubbles in multi-metal flotation. A watershed method based on illuminance modeling is provided in this study. Firstly, the dome of bubble and the distribution of illuminance are estimated by median filtering; Then, the difference image of these two estimated images is set as topological image and markers are extracted; Finally, watershed transform is performed to obtain the segmentation result. The experimental results have demonstrated the effectiveness of the proposed segmentation method in the dealing of froth images captured in the flotation condition full with impuriy particles.
Keywords: froth flotation watershed illuminance modeling median filtering
论文图表:
引用
No.4732695120003514****
同行评议
共计0人参与
勘误表
基于光照建模的改进分水岭锌浮选泡沫图像分割
评论
全部评论0/1000