基于多源遥感数据的森林树种准确识别方法研究

Accurate Identification Method of Forest Tree Species Based on Multi-source Remote Sensing Data

针对传统森林树种识别方法存在的远程遥感信息监测和识别能力较低的问题,提出基于遥感成像和大数据分析的森林树种准确识别方法。采用遥感影像技术采集森林树种的多源遥感成像大数据,并通过建立森林树种多源遥感影像分析模型,对采集的森林树种多源遥感数据进行融合处理,然后通过边缘像素特征检测方法分析森林树种多源遥感图像的特征,结合点匹配方法得到森林树种多源遥感图像的子带像素特征点,对图像实施遥感特征匹配和信息增强,并根据信息增强结果以及图像的角点信息和边缘像素点信息检测结果进行模糊度识别和分类,结合支持向量机和极限学习机进行森林树种遥感图像检测和信息识别。仿真结果表明:该方法对森林树种识别分类的准确性较高,且图像辨识度较高,提高了对森林树种的远程遥感信息监测和识别能力。

Aiming at the poor remote sensing information monitoring and identification ability of traditional forest tree species identification methods,an accurate forest tree species identification method based on remote sensing imaging and big data analysis was proposed.With remote sensing image acquisition forest tree species of multi-source remote sensing image data,andthe forest tree species multi-source remote sensing image analysis model,the forest tree species to fusion of multi-source remote sensing data was analyzed.And according to the result of information enhancement and the angle of the image information and edge pixel information detection,the fuzzy degree of recognition and classification for remote sensing image detection and information recognition of forest tree species are carried out.The simulation results show that this method has a higher accuracy in the identification and classification of forest tree species,and the image identification is higher,which improves the remote sensing information monitoring and identification ability of forest tree species.