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

Evaluation functions in aerial image...

Langford, William Todd.

FindBook

Google Book

Amazon

博客來

Evaluation functions in aerial image segmentation.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Evaluation functions in aerial image segmentation./
作者:	Langford, William Todd.
面頁冊數:	130 p.
附註:	Advisers: Thomas G. Dietterich; Warren Cohen.
Contained By:	Dissertation Abstracts International63-03B.
標題:	Computer Science. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3044338
ISBN:	049358269X

Evaluation functions in aerial image segmentation.
Langford, William Todd.

Evaluation functions in aerial image segmentation. - 130 p.

Advisers: Thomas G. Dietterich; Warren Cohen.

Thesis (Ph.D.)--Oregon State University, 2002.

Finally, we describe a prototype segmentation editing tool we built to simplify generating hand segmentations.

ISBN: 049358269XSubjects--Topical Terms:

626642
Computer Science.

Evaluation functions in aerial image segmentation.
LDR:03663nam 2200373 a 45 001 927371
005 20110425
008 110425s2002 eng d
020 $a 049358269X
035 $a (UnM)AAI3044338
035 $a AAI3044338
040 $a UnM $c UnM
100 1 $a Langford, William Todd. $3 1250929
245 1 0 $a Evaluation functions in aerial image segmentation.
300 $a 130 p.
500 $a Advisers: Thomas G. Dietterich; Warren Cohen.
500 $a Source: Dissertation Abstracts International, Volume: 63-03, Section: B, page: 1429.
502 $a Thesis (Ph.D.)--Oregon State University, 2002.
520 $a Finally, we describe a prototype segmentation editing tool we built to simplify generating hand segmentations.
520 $a This work is of interest to landscape ecologists computing metrics from maps derived from images, to remote sensing scientists evaluating their efforts in generating those maps, and to developers of segmentation algorithms and evaluation functions.
520 $a Remote sensing is the most practical way to acquire large amounts of land cover data for monitoring and understanding environmental change, so it is important to be able to map land cover from imagery. Maps defining land cover patches as polygons rather than pixels greatly improve processing efficiency in models and are often more relevant to the scale of biophysical processes. Defining boundaries around homogeneous patches of an image is called <italic> segmentation</italic>. Assigning a cover type label to a pixel or polygon is called <italic>classification</italic>.
520 $a Accuracy of maps generated from images is generally specified in terms of classification accuracy with no specification of spatial distribution of error. We demonstrate need for spatial accuracy assessment by showing that even small classification error rates can cause large changes in a standard landscape statistic, average patch compaction (APC). Furthermore, the common practice of reducing classification error by forcing a minimum mapping unit size can <italic>increase</italic> error in APC.
520 $a We also investigate the importance of segmentation evaluation functions as the basis for scientific decisions, comparing algorithms, and driving optimizers. We divide these functions into <italic>reference-based</italic> (scores are derived through comparison to a “correct” reference segmentation) and <italic>heuristic</italic> (scores are based on attributes of the image without knowledge of a reference).
520 $a We quantitatively measure performance of several evaluation functions embedded in a region merging algorithm on ecological images where hand segmentations are available. We explain their poor performance by showing that deterministic, irreversible segmentation algorithms like region merging can only represent <italic> n</italic> of 2<italic><super>n</super></italic> possible segmentations for an initial segmentation containing <italic>n</italic> boundaries, suggesting that evaluation functions must be learned. We demonstrate a reference-based evaluation function as a target for learning. It achieves near optimal performance inside a region merging algorithm, showing that poor evaluation functions, not greedy algorithms, limit performance in region merging.
590 $a School code: 0172.
650 4 $a Computer Science. $3 626642
650 4 $a Environmental Sciences. $3 676987
650 4 $a Physical Geography. $3 893400
650 4 $a Remote Sensing. $3 1018559
690 $a 0368
690 $a 0768
690 $a 0799
690 $a 0984
710 2 0 $a Oregon State University. $3 625720
773 0 $t Dissertation Abstracts International $g 63-03B.
790 $a 0172
790 1 0 $a Cohen, Warren, $e advisor
790 1 0 $a Dietterich, Thomas G., $e advisor
791 $a Ph.D.
792 $a 2002
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3044338