摘要
植被分类能够对一个地区的植被状况进行分析,在环境保护和农业生产方面具有很重要的意义。本研究针对植被遥感图像分类识别需求,首先通过粒子群算法来实现马尔科夫的寻优过程,然后通过遗传算法进行支持向量机解空间的搜索,从而实现更加准确的图像分割和识别,最后通过实验验证了算法的有效性。
The vegetation condition in a certain area can be analyzed by vegetation classification,which has great significance in environmental protection and agricultural production.According to the requirement of classification and recognition for the remote sensing vegetation images,a new method for image segmentation is presented based on particle swarm optimization.The genetic algorithm is used to search the solution space of support vector machine.Experiment results showed that it is an effective method to the image segmentation and recognition.
出处
《河北农业大学学报》
CAS
CSCD
北大核心
2016年第3期131-135,共5页
Journal of Hebei Agricultural University
基金
基金项目:基于GIS的保定市检察院警用车辆优化调度策略研究(SZ151020)
关键词
遥感影像
特征提取
图像分类
植被覆盖
remote sensing image
feature extraction
image classification
vegetation coverage
作者简介
陶佳(1980-),女,河北省保定人,硕士,实验师,研究方向:农业信息化.