摘要
针对植物病害图像成分复杂、病斑排列无规则等特点,提出了一种改进型模糊聚类的病斑检测算法。该算法采用Markov随机场与模糊聚类算法耦合策略,能够有效解决植物病斑检测时的模糊性和随机性问题。仿真实验表明,改进后的算法能够实现植物病斑的自适应检测,鲁棒性较好。然而,对于Markov与模糊聚类算法的最佳耦合方式及对于如何减少算法的运算量仍需作深入的研究。
According to the characteristics of complex components of plant disease images and random alignment of disease spot, this paper introduces a new detection algorithm with improved Fuffzzy clustering algorithm.By adopting the coupled method of Markov random field and fuzzy clustering algorithm,this algorithm avoids the fuzziness and randomness during the plant disease detection.Simulation experiment shows that the improved algorithm has better robust and can realize the self-adaptive detection of plant disease spot.However,how to reduce the operating quantum of the method and get the optimal coupling method between Markov random field and fuzzy clustering algorithm need further research.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第24期203-204,245,共3页
Computer Engineering and Applications
基金
农业科技成果转化资金项目(No.05EFN211100002)
北京市科技计划(the Key Technologies R&D Program of Beijing City
China under Grant No.No.Z0005190040831)联合资助
