摘要
在农作物遥感估产研究过程中,如何快速、准确获取当年种植面积是一个关键技术问题。本文重点研究在禹城县冬小麦遥感估产试验中,应用同步TM信息源,根据冬小麦生长发育的特征,选择 TM 的适宜时相,构建多维绿度图,采用模式识别技术,分层自动提取纯麦地、套种麦地信息。这项研究结果与1/5万比例尺 TM 图像目视解译小麦面积相比较,其相对误差甚小,达到了估产实际应用的精度。
When estimating crop production by remote sensing technology,a key problem is how to acquire crop area of the same year quickly and accurately.This paper centers on determining pure wheat area and mixed wheat area separately in the experiment of winter wheat production estimation by remote sensing technique in Yucheng County.The study takes TM imagery as the main information source, selects suitable date of TM images according to the winter wheat growing feature,creates multi—di- mensional greenness maps,and utilizes pattern recognition to extract wheat area and interplanted wheat area separately.The result of this study conforms to wheat area acquired by TM image visual interpretation on the scale of 1:50000.There is only a little relative error between these two re- sults.Accuracy of this study reaches the requirement of practical application.
出处
《遥感技术与应用》
CSCD
1993年第4期28-33,共6页
Remote Sensing Technology and Application
关键词
TM图像
模式识别
冬小麦
播种面积
TM imagery
Multi—dimensional greenness image
Pattern recognition
Discriminative function