A Gas Recognition Method based on Hough Transform and K-Means Algorithm

摘要：

In this paper, we apply Hough transform circle detection method to solve the problem of initial points of K-Means clustering algorithm. With the RGB values of the four detected circle centre points and a background point ,all pixels in the image are clustered into 5 groups. In this way, the convergence and accuracy of K-Means algorithm can be greatly improved. Then a 12 dimension vector can be obtained which will be used for the recognition process. In the end, we give an example of acetaldehyde to demonstrate the effectiveness of proposed algorithm.

关键词： electronic nose Hough Transform K-means clustering foreground segmentation

作者: Xiangbin Kong Jin Qi

作者单位: School of Electronic Engineering UESTC Chengdu, China

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Automation Engineering(ICCAE 2011)(2011年第三届IEEE计算机与自动化工程国际会议)

会议地点: 重庆

会议语种:英文

页码: 450-452

在线出版日期: 2011-01-21（万方平台首次上网日期，不代表论文的发表时间）