本文選題：TM遙感影像 + 最佳波段組合　； 參考：《內(nèi)蒙古農(nóng)業(yè)大學(xué)》2014年碩士論文

【摘要】：本文以烏拉特前旗黃河斷面為研究區(qū)。首先,根據(jù)最佳波段組合的選取原則,利用matlab編程語(yǔ)言實(shí)現(xiàn)單波段信息提取和相關(guān)系數(shù)矩陣、信息熵、最佳指數(shù)的統(tǒng)計(jì)。經(jīng)反復(fù)比對(duì)和分析后,得到烏拉特前旗黃河斷面暢流期TM542和冰封期TM543的最佳波段組合。經(jīng)組合后得到的遙感影像能夠最大程度的反映地物的信息,使各類地物間的差異明顯,為下一步的遙感影像分類奠定基礎(chǔ)。其次,在最佳波段組合影像的基礎(chǔ)上,開始對(duì)遙感圖像進(jìn)行分類操作,經(jīng)對(duì)比分析K-means算法、馬氏距離和最大似然法的分類結(jié)果,得到分類效果最好的人機(jī)交互式最大似然法的分類效果圖。接著制作出地物分類的專題圖,計(jì)算出冰封期各類地物的面積。通過(guò)專題圖,我們還可以觀察各類地物的分布現(xiàn)狀,這些都為防凌工作提供了技術(shù)支持。最后,進(jìn)行遙感圖像拼接的操作,它可以解決在不同時(shí)間不同角度獲取的兩幅或者多幅圖像的匹配處理問(wèn)題。本文在現(xiàn)有的配準(zhǔn)技術(shù)中選取了基于特征點(diǎn)匹配的SIFT算法,并引入可以對(duì)誤匹配點(diǎn)進(jìn)行優(yōu)化的RANSAC算法,以便進(jìn)一步提高匹配的準(zhǔn)確度。匹配結(jié)束后通過(guò)雙向線性內(nèi)插法,對(duì)拼接處的影像進(jìn)行平滑,加強(qiáng)了視覺效果,得到了圖像拼接的全景圖。
[Abstract]:This paper takes the Yellow River section of the former Banner of Wulat as the research area. Firstly, according to the selection principle of optimal band combination, the single band information extraction, correlation coefficient matrix, information entropy and best index statistics are realized by matlab programming language. After repeated comparison and analysis, the optimal band combination of TM542 and TM543 in the Yellow River section of Qianqi, Wulat, was obtained. The combined remote sensing images can reflect the information of the ground objects to the greatest extent, which makes the difference between different ground objects obvious, and lays a foundation for the classification of remote sensing images in the next step. Secondly, on the basis of the best band combination image, the classification operation of remote sensing image is started, and the classification results of K-means algorithm, Mahalanobis distance and maximum likelihood method are compared and analyzed. The classification effect map of the best human-computer interactive maximum likelihood method is obtained. Then make the thematic map of the classification of features and calculate the area of all kinds of objects in the ice period. Through the thematic map, we can also observe the distribution of various features, which provide technical support for anti-ice work. Finally, the operation of remote sensing image mosaic is carried out, which can solve the problem of matching two or more images obtained at different time and different angles. In this paper, the SIFT algorithm based on feature point matching is selected from the existing registration techniques, and the RANSAC algorithm which can optimize the mismatch points is introduced in order to further improve the accuracy of matching. After matching, the image is smoothed by bidirectional linear interpolation, the visual effect is enhanced, and the panoramic image of image stitching is obtained.
【學(xué)位授予單位】：內(nèi)蒙古農(nóng)業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TV875;TV882.1

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 武文波;劉正綱;;一種基于地物波譜特征的最佳波段組合選取方法[J];測(cè)繪工程;2007年06期

2 潘琛;杜培軍;張海榮;;決策樹分類法及其在遙感圖像處理中的應(yīng)用[J];測(cè)繪科學(xué);2008年01期

3 羅音,舒寧;基于信息量確定遙感圖像主要波段的方法[J];城市勘測(cè);2002年04期

4 文雅玫;王連生;李思昆;;一種遙感圖像高精度自動(dòng)拼接算法[J];湘南學(xué)院學(xué)報(bào);2006年05期

5 張韜;呂洪娟;孫美霞;安慧君;;遙感多光譜數(shù)據(jù)在內(nèi)蒙古西部濕地監(jiān)測(cè)中最佳波段選取的應(yīng)用研究——以烏梁素海域?yàn)槔齕J];干旱區(qū)資源與環(huán)境;2007年04期

6 李石華,王金亮,畢艷,陳姚,朱妙園,楊帥,朱佳;遙感圖像分類方法研究綜述[J];國(guó)土資源遙感;2005年02期

7 朱曉榮;張懷清;;西洞庭湖濕地遙感最佳波段選擇研究[J];現(xiàn)代農(nóng)業(yè)科技;2012年15期

8 金寶石;周葆華;;TM影像在湖泊濕地信息提取中的最佳波段組合[J];光譜實(shí)驗(yàn)室;2012年06期

9 李響;;淺談MATLAB在圖像處理中的應(yīng)用[J];產(chǎn)業(yè)與科技論壇;2012年14期

10 卜凡艷;檀結(jié)慶;;利用SIFT算子與圖像插值實(shí)現(xiàn)圖像匹配[J];計(jì)算機(jī)工程與應(yīng)用;2011年16期

，

本文編號(hào)：1899129