WorldView-2影像林地信息提取的研究与实现

林地信息的获取具有重要意义,可广泛应用于城市规划编制、林地资源分析、水土流失治理等领域。本文基于WorldView-2影像,在研究分割与对象特征的基础上,采用面向对象方法,实现了林地信息提取,其用户精度达到94%,制图精度达到97.2%。然后,将像元二分模型应用于林地类对象,计算其植被覆盖度。试验表明,面向对象方法能高效、准确、经济地提取林地信息;基于林地类对象计算的植被覆盖度能较好地表征植被覆盖情况。

The research and realization of the forest information extraction based on the WorldView-2 image

The forest plays an important role in urban planning,forest analysis and soil erosion management.After studying the segmentation and the objects features,the forest information is extracted from the WorldView-2 image based on the object-oriented method in this article.The users' accuracy of the forest information extraction is 94%,and the mapping accuracy is 97.2%.Then,the pixel binary model is applied to the forest objects to calculate their vegetation coverage.There is a conclusion that we can extract the forest information from the high resolution image based on the object-oriented method economically,accurately and efficiently.It is good to apply the pixel binary model to the objects to calculate the vegetation coverage,and the vegetation cover situation can be represented well.