摘要
为解决采用遥感技术监测红树林群落存在的识别率较低的问题,提出了一种基于协同神经网络算法的红树林图像识别方法.首先,采用协同神经网络算法中的平衡网络参数方法对红树林图像进行识别.其次,利用微粒群算法对平衡参数方法进行改进.实验结果显示,该方法对红树林图像识别效率达到88.0%,显著优于传统的协同神经网络算法的识别率(78.0%),因此该方法具有良好的应用价值.
In order to solve the problem of low recognition rate in mangrove community monitoring by remote sensing technology,a method of mangrove image recognition based on synergetic neural network algorithm was proposed.Firstly,the synergetic neural network algorithm was used to recognize mangrove images by balancing the network parameters.Secondly,the method of particle swarm optimization algorithm was used to improve the balance parameter method.The result shows that the recognition efficiency of the method reaches 88.0%,which is significantly better than the recognition efficiency(78.0%)of the traditional synergetic neural network algorithm.So the method has good application value.
作者
孙静
SUN Jing(College of Intelligent Manufacturing Engineering,Liming Vocational University,Quanzhou 362000,China)
出处
《延边大学学报（自然科学版）》
CAS
2021年第1期64-69,共6页
Journal of Yanbian University（Natural Science Edition）
基金
泉州市科技局科技计划项目(2018C102R)
福建省教育厅中青年教师教育科研项目(JAT191465)
黎明职业大学科研团队项目(LMTD202001)。
关键词
红树林
协同神经网络
平衡参数
微粒群算法
mangrove
synergetic neural network
balancing the parameters
particle swarm optimization
