摘要
针对传统数据挖掘方法存在着挖掘数据效率、精度低等问题,提出一种基于最大间隔准则与最小最大概率机相融合的应急决策系统数据自助挖掘方法。首先利用应急系统数据传输信道对整个系统需求数据进行采集,利用采集数据构建信号模型,通过信号模型提取应急决策系统中需求数据特征,其次运用最大间隔准则算法将应急系统中的需求数据高维特征投影至低维特征空间中,结合最小最大概率机算法对应急系统中需求数数据进行自助挖掘,得到需求数据挖掘结果。通过仿真结果验证了所提挖掘方法在精确度、运行时间等方面具有优越性能。
Due to long detection time and high false detection rate of traditional automatic detection method for building facade damage, this article proposed an automatic detection method for building facade damage based on Gini coefficient. Firstly, our method combined the total variation denoising model with fractional order differential theory to build a new denoising model of building facade image. And then, we obtained the smooth image by denoise the image. Moreover, we used Hear wavelet transform to perform the adaptive threshold segmentation on approximate components at the highest level, so as to obtain the initial damage area of building facade. In other layers, we further segmented the damaged area. Finally, we automatically determine whether the building facade was damaged by Gini coefficient. Simulation results prove that the proposed method can effectively suppress noise interference and reduce detection time. Meanwhile, this method can improve the detection efficiency and obtain the accurate detection result.
作者
彭秦晋
PENG Qin-jing(Jinzhong University, Jinzhong Shanxi 030600, China)
出处
《计算机仿真》
北大核心
2019年第8期329-332,共4页
Computer Simulation
