ШУА, ГГХ-ийн эрдэм шинжилгээний бүтээл, МОГЗА, ...

URL: http://portal.igg.ac.mn/dataset/90eeb3c9-9953-428e-b7d4-1e9f5c1a3596/resource/4b00a402-7d79-4e72-b089-2957491a58ba/download/-_2016.pdf

Abstract The purpose of this paper is to conduct the forest mapping using the remotely sensed optical and synthetic aperture radar data sets. For the image fusion the principal component transformation (PCA) has been applied. Advanced satellite images classification represents an accurate and cost effective for land cover mapping at regional scale. The output of each of the feature extraction method is classified using NN (neural net) and SVM (support vector machine) classification methods. The results are analyzed and compared. For the second purpose object oriented segmentation approach has been used for the forest biomass. In this aim eCognition software can be used to detect and discriminate forest biomass by describing NDVI. This software is completely object oriented and uses a patented, multi-scale image segmentation approach. The generated segments act as image objects whose physical and contextual characteristics can be described by means of fuzzy logic. As a test site Bogdkhan mountain has been selected and Sentinel 1A radar and Sentinel 2A satellite optical images have been used.

Энэ нөөцөд зориулж үүсгэсэн харагдац байхгүй байна

Нэмэлт мэдээлэл

Талбар Утга
Өгөгдлийн сүүлийн шинэчлэл 2018 9-р сар 26
Мета өгөгдөлийн сүүлийн шинэчлэл 2018 9-р сар 26
Үүссэн 2018 9-р сар 26
Хэлбэр application/pdf
Ашиглах зөвшөөрөл Бусад (Арилжааны бус)
created2 жилээс өмнө
formatPDF
id4b00a402-7d79-4e72-b089-2957491a58ba
last modified2 жилээс өмнө
mimetypeapplication/pdf
on same domainTrue
package id90eeb3c9-9953-428e-b7d4-1e9f5c1a3596
revision id7485adb6-5eff-41d8-999c-6f087a1772dd
size1.7 мегабайт
stateactive
url typeupload