東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

A fuzzy multicriteria decision proce...

Chiou, Chyi-Rong.

FindBook

Google Book

Amazon

博客來

A fuzzy multicriteria decision process for classification of Landsat TM data of the Rocky Mountain National Park.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	A fuzzy multicriteria decision process for classification of Landsat TM data of the Rocky Mountain National Park./
作者:	Chiou, Chyi-Rong.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 1994,
面頁冊數:	167 p.
附註:	Source: Dissertation Abstracts International, Volume: 55-11, Section: B, page: 4657.
Contained By:	Dissertation Abstracts International55-11B.
標題:	Forestry. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9509628

A fuzzy multicriteria decision process for classification of Landsat TM data of the Rocky Mountain National Park.
Chiou, Chyi-Rong.

A fuzzy multicriteria decision process for classification of Landsat TM data of the Rocky Mountain National Park. - Ann Arbor : ProQuest Dissertations & Theses, 1994 - 167 p.

Source: Dissertation Abstracts International, Volume: 55-11, Section: B, page: 4657.

Thesis (Ph.D.)--Colorado State University, 1994.

Landsat TM can provide important spectral and time specific information about ground cover types. Generally speaking, the classification accuracy of Landsat TM data for Level III in mountainous terrain is relatively low. Therefore, the main purpose of this research was to develop an effective method to integrate spectral and ancillary information in order to increase the classification performance in areas of mountainous terrain.Subjects--Topical Terms:

895157
Forestry.

A fuzzy multicriteria decision process for classification of Landsat TM data of the Rocky Mountain National Park.
LDR:03306nmm a2200289 4500 001 2121881
005 20170821084144.5
008 180830s1994 ||||||||||||||||| ||eng d
035 $a (MiAaPQ)AAI9509628
035 $a AAI9509628
040 $a MiAaPQ $c MiAaPQ
100 1 $a Chiou, Chyi-Rong. $3 3283824
245 1 2 $a A fuzzy multicriteria decision process for classification of Landsat TM data of the Rocky Mountain National Park.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 1994
300 $a 167 p.
500 $a Source: Dissertation Abstracts International, Volume: 55-11, Section: B, page: 4657.
502 $a Thesis (Ph.D.)--Colorado State University, 1994.
520 $a Landsat TM can provide important spectral and time specific information about ground cover types. Generally speaking, the classification accuracy of Landsat TM data for Level III in mountainous terrain is relatively low. Therefore, the main purpose of this research was to develop an effective method to integrate spectral and ancillary information in order to increase the classification performance in areas of mountainous terrain.
520 $a A sequential sampling was employed to select and evaluate training blocks. Texture and aspect information were used to identify spectrally homogeneous areas for developing training areas, and a spatial clumping method was used to clump the training polygons. A quantitative distribution model for describing the cover types was utilized to examine the elevation distribution of the cover types. A fuzzy multicriteria decision process was developed to integrate spectral and elevation information for classifying Landsat TM data. Nine classification algorithms were compared in this study, and four accuracy assessment methods were used to evaluate the classification performances.
520 $a The results showed that twenty-three training blocks were effective for representing the entire Rocky Mountain National Park. Within the training blocks, texture and aspect information were useful to identify spectrally homogeneous areas. A total of 2269 potential training polygons were clumped by the spatial clumping methods. From these 2269 potential training polygons, a total of 685 training polygons were identified. Each polygon was treated as a training class. In addition, in order to evaluate the assumption of normality and the effectiveness of transformed divergence, two additional sets of training data involving 257 and 163 training classes, were developed from these 685 training polygons. For the purpose of evaluation, three levels of classification were used. The results indicated that a two-stage approach which uses the minimum-distance to means to determine the five most possible candidates and the fuzzy multicriteria decision process for the final decision, had the best classification performance for using the spectral and elevation information. This approach resulted in an overall accuracy of 62.2%, for the entire Rocky Mountain National Park based on 17 information categories, and an overall accuracy of 75.3% based on 8 information categories.
590 $a School code: 0053.
650 4 $a Forestry. $3 895157
650 4 $a Remote sensing. $3 535394
690 $a 0478
690 $a 0799
710 2 $a Colorado State University. $3 675646
773 0 $t Dissertation Abstracts International $g 55-11B.
790 $a 0053
791 $a Ph.D.
792 $a 1994
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9509628