基于機(jī)器學(xué)習(xí)的遙感圖像分類研究
本文選題:遙感影像 + 混合核函數(shù); 參考:《江西理工大學(xué)》2017年碩士論文
【摘要】:隨著航空航天科技和傳感器科技的快速更新?lián)Q代,遙感影像的數(shù)據(jù)來源變得多樣化,數(shù)據(jù)集也日趨復(fù)雜,地物復(fù)雜難辨,如何準(zhǔn)確、高效地進(jìn)行遙感圖像分類成為了近年來研究的重要內(nèi)容。由于人工智能科技發(fā)展迅速,機(jī)器學(xué)習(xí)分類方法也逐漸成為一種有效的遙感圖像分類處理方法,為了有效提高影像的分類精度,本文在機(jī)器學(xué)習(xí)的理論上構(gòu)建了一種高效簡單的分類模型以及一種兩級分類模型。論文的主要工作與創(chuàng)新點如下:(1)系統(tǒng)地介紹了遙感技術(shù)發(fā)展的研究現(xiàn)狀,簡要闡述監(jiān)督分類與非監(jiān)督分類常用的一些方法以及現(xiàn)今前沿的分類器;扼要地總結(jié)了本文的研究內(nèi)容、及其組織結(jié)構(gòu)流程。(2)針對遙感影像在獲取過程中易受大氣吸收與散射、傳感器定標(biāo)、地形等因素的影響而造成圖像失真的特點,本文利用二次多項式模型進(jìn)行遙感影像幾何校正,采取雙線性內(nèi)插法進(jìn)行重采樣等技術(shù)對影像進(jìn)行校正預(yù)處理,并進(jìn)行了大氣校正,有效地去除了傳感器等因素的畸變影響,同時也排除了大氣中散射顆粒的影響,為后續(xù)分類奠定基礎(chǔ)。(3)神經(jīng)網(wǎng)絡(luò)模型具有容錯性、學(xué)習(xí)能力強(qiáng)等特點,但要得到較好的分類效果耗時非常大,而極限學(xué)習(xí)機(jī)分類器是一種結(jié)構(gòu)簡單的神經(jīng)網(wǎng)絡(luò)方法,能快速高效的對樣本進(jìn)行識別。本文構(gòu)建了一種混合核極限學(xué)習(xí)機(jī)遙感圖像分類模型,該模型利用混合核函數(shù)的全局與局部特性,結(jié)合遙感圖像的鄰域信息,有效地提高了分類精度。(4)基于遙感影像的光譜和空間信息提出了一種兩級分類器的方法。結(jié)合光譜信息與空間結(jié)構(gòu)信息,首先采用光譜角匹配方法作為前級分類器,提取影像中光譜信息特征明顯且區(qū)別較大的地物;然后利用遙感數(shù)據(jù)的張量空間結(jié)構(gòu)信息,選取支持張量機(jī)作為后級分類器。對選取的感興趣區(qū)域進(jìn)行分類,不僅提高了分類精度,而且分類視覺效果也有了明顯改善。
[Abstract]:With the rapid upgrading of aerospace and sensor technologies, the data sources of remote sensing images become more and more diverse, the data sets become more and more complex, the complexity of ground objects is difficult to distinguish, how to be accurate, Efficient classification of remote sensing images has become an important research content in recent years. Due to the rapid development of artificial intelligence technology, machine learning classification method has gradually become an effective remote sensing image classification processing method, in order to effectively improve the image classification accuracy, In this paper, an efficient and simple classification model and a two-level classification model are constructed in theory of machine learning. The main work and innovation of this paper are as follows: (1) the research status of remote sensing technology is introduced systematically, and some common methods of supervised classification and unsupervised classification are briefly described, as well as the current frontier classifiers. This paper briefly summarizes the research contents of this paper, and its organization and structure. 2) aiming at the characteristics of image distortion caused by the influence of atmospheric absorption and scattering, sensor calibration, topography and so on, in the process of remote sensing image acquisition. In this paper, the quadratic polynomial model is used for the geometric correction of remote sensing image, the bilinear interpolation method is used to resample the image and the atmospheric correction is carried out, which effectively removes the distortion effect of the sensor and other factors. At the same time, the influence of scattering particles in the atmosphere is excluded, which lays the foundation for the following classification. The neural network model has the characteristics of fault tolerance and strong learning ability, but it takes a lot of time to obtain better classification effect. The extreme learning machine classifier is a simple neural network method, which can identify samples quickly and efficiently. In this paper, a hybrid kernel extreme learning machine remote sensing image classification model is constructed. The model combines the global and local characteristics of the hybrid kernel function and the neighborhood information of the remote sensing image. A two-level classifier based on spectral and spatial information of remote sensing image is proposed. Combining the spectral information with the spatial structure information, the spectral angle matching method is first used as the front classifier to extract the features of the spectral information in the image, and then the spatial structure information of Zhang Liang from the remote sensing data is used. Zhang Liang machine is selected as the posterior classifier. The classification of selected regions of interest not only improves the classification accuracy, but also improves the visual effect of classification.
【學(xué)位授予單位】:江西理工大學(xué)
【學(xué)位級別】:碩士
【學(xué)位授予年份】:2017
【分類號】:TP751;TP181
【參考文獻(xiàn)】
相關(guān)期刊論文 前10條
1 曹兆偉;林寧;徐文斌;盧文虎;張孝龍;;基于BP神經(jīng)網(wǎng)絡(luò)的東嶼島遙感影像分類[J];海洋通報;2016年05期
2 曹林林;李海濤;韓顏順;余凡;顧海燕;;卷積神經(jīng)網(wǎng)絡(luò)在高分遙感影像分類中的應(yīng)用[J];測繪科學(xué);2016年09期
3 周宇谷;王平;高穎慧;;基于視覺詞袋模型的遙感圖像分類方法[J];重慶理工大學(xué)學(xué)報(自然科學(xué));2015年05期
4 劉萬軍;楊秀紅;曲海成;孟煜;;基于光譜信息散度與光譜角匹配的高光譜解混算法[J];計算機(jī)應(yīng)用;2015年03期
5 張樂飛;;遙感影像的張量表達(dá)與流形學(xué)習(xí)方法研究[J];測繪學(xué)報;2013年05期
6 陳善靜;胡以華;石亮;王磊;孫杜娟;徐世龍;;空-譜二維蟻群組合優(yōu)化SVM的高光譜圖像分類[J];光譜學(xué)與光譜分析;2013年08期
7 閻繼寧;周可法;王金林;王珊珊;汪瑋;李東;;基于SAM與SVM的高光譜遙感蝕變信息提取[J];計算機(jī)工程與應(yīng)用;2013年19期
8 周蓉;楊曉偉;吳廣潮;;在線支持張量機(jī)[J];計算機(jī)科學(xué)與探索;2013年07期
9 萬曙靜;張承明;劉俊華;;基于自適應(yīng)最小距離調(diào)整的多光譜遙感圖像分類方法[J];測繪通報;2012年S1期
10 張樂飛;黃昕;張良培;;高分辨率遙感影像的支持張量機(jī)分類方法[J];武漢大學(xué)學(xué)報(信息科學(xué)版);2012年03期
相關(guān)博士學(xué)位論文 前3條
1 何同弟;高光譜圖像的分類技術(shù)研究[D];重慶大學(xué);2014年
2 楊國鵬;基于機(jī)器學(xué)習(xí)方法的高光譜影像分類研究[D];解放軍信息工程大學(xué);2010年
3 王煜;基于決策樹和K最近鄰算法的文本分類研究[D];天津大學(xué);2006年
相關(guān)碩士學(xué)位論文 前2條
1 唐雪飛;基于案例推理的高光譜圖像分類研究[D];哈爾濱工業(yè)大學(xué);2010年
2 吳靜;基于支持向量機(jī)與圖斑的高光譜分類方法研究[D];長安大學(xué);2010年
,本文編號:1794853
本文鏈接:http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/1794853.html