基于CUDA的影像配準(zhǔn)與拼接方法研究
發(fā)布時(shí)間：2018-07-30 08:33
【摘要】：圖像作為客觀世界能量或狀態(tài)的可視化形式,為人們認(rèn)知和改造客觀世界提供了豐富的信息。而圖像配準(zhǔn)和圖像拼接作為圖像處理的基本問題之一,在虛擬現(xiàn)實(shí)、地質(zhì)勘測(cè)、醫(yī)學(xué)影像、氣象預(yù)測(cè)、應(yīng)急響應(yīng)等領(lǐng)域都有著廣泛的應(yīng)用。 圖形處理器GPU將更多的晶體管用作執(zhí)行單元,計(jì)算能力遠(yuǎn)遠(yuǎn)超過傳統(tǒng)的中央處理器CPU。目前,GPU技術(shù)已經(jīng)廣泛用于數(shù)據(jù)挖掘、數(shù)理統(tǒng)計(jì)、圖像語音識(shí)別、基因工程、全球氣候準(zhǔn)確預(yù)報(bào)等領(lǐng)域,同時(shí)也為遙感影像的快速處理提供了一種新的解決方案。 針對(duì)當(dāng)前影像配準(zhǔn)和拼接計(jì)算中存在問題,本文在分析當(dāng)前影像配準(zhǔn)和拼接技術(shù)的基礎(chǔ)上,結(jié)合CUDA并行計(jì)算技術(shù),重點(diǎn)研究了大尺寸高分辨率遙感影像配準(zhǔn)方法以及航空影像的在線實(shí)時(shí)拼接方法。本文的主要研究?jī)?nèi)容包括：(1)大尺寸遙感影像配準(zhǔn)。針對(duì)傳統(tǒng)遙感影像配準(zhǔn)方法難以適應(yīng)于大尺寸高分辨率遙感影像配準(zhǔn)的問題,本文提出一種由粗到精的配準(zhǔn)控制點(diǎn)匹配方法；在此基礎(chǔ)上,采用小面元微分糾正方法實(shí)現(xiàn)大尺寸影像的高精度糾正,并采用一種自適應(yīng)的掃描線填充算法來計(jì)算每個(gè)糾正像素所在的三角形；本文通過分析配準(zhǔn)中各個(gè)步驟的計(jì)算瓶頸問題,利用CUDA并行計(jì)算技術(shù)對(duì)控制點(diǎn)匹配和影像糾正兩個(gè)階段進(jìn)行了加速。通過IKONOS全色影像、Geoeye全色影像和多光譜影像、ZY-3衛(wèi)星影像等實(shí)驗(yàn)表明,本文方法可以取得較高的配準(zhǔn)精度,且利用GPU加速算法獲得了較高的加速比。(2)航空影像實(shí)時(shí)在線拼接。針對(duì)航空影像在線實(shí)時(shí)拼接的計(jì)算瓶頸,本文提出了一種基于CPU與GPU協(xié)同處理的在線拼接方法�；赑OS數(shù)據(jù),在CPU端計(jì)算原始影像與糾正影像之間的單應(yīng)變換關(guān)系,然后利用GPU并行計(jì)算實(shí)現(xiàn)影像的糾正過程；由于航空影像間具有較大重疊度,因此本文提出一種自適應(yīng)的拼接方法,即通過計(jì)算后續(xù)影像的重疊度以判斷當(dāng)前影像是否需要拼接,大大了減少冗余計(jì)算；同時(shí)本文利用兩臺(tái)計(jì)算機(jī)進(jìn)行了模擬實(shí)驗(yàn),其中一臺(tái)作為模擬相機(jī)拍攝影像,往處理機(jī)器上傳輸影像。實(shí)驗(yàn)結(jié)果表明,該方法基本實(shí)現(xiàn)了航空影像的在線實(shí)時(shí)拼接,且拼接結(jié)果滿足為災(zāi)害、突發(fā)情況的實(shí)時(shí)救援提供決策信息需求。 本文研究所提出的配準(zhǔn)和拼接方法采用Visual C++進(jìn)行了實(shí)現(xiàn),并開發(fā)了相應(yīng)原型系統(tǒng)。有關(guān)實(shí)驗(yàn)結(jié)果表明本文方法能實(shí)現(xiàn)大尺寸遙感影像的高效配準(zhǔn)以及航空影像的在線拼接。文中圖26個(gè),表3個(gè),參考文獻(xiàn)48篇。
[Abstract]:As a visual form of the energy or state of the objective world, image provides abundant information for people to recognize and transform the objective world. As one of the basic problems of image processing, image registration and image mosaic are widely used in the fields of virtual reality, geological survey, medical image, meteorological prediction, emergency response and so on. The graphics processor GPU uses more transistors as executive units, much more computing power than traditional central processor CPUs. At present, GPU technology has been widely used in the fields of data mining, mathematical statistics, image speech recognition, genetic engineering, accurate prediction of global climate and so on. At the same time, it provides a new solution for the rapid processing of remote sensing images. Aiming at the existing problems in image registration and stitching calculation, this paper analyzes the current image registration and stitching technology, and combines CUDA parallel computing technology. The registration method of large scale and high resolution remote sensing image and the online real-time mosaic method of aerial image are mainly studied in this paper. The main contents of this paper are as follows: (1) large scale remote sensing image registration. Aiming at the problem that the traditional remote sensing image registration method is difficult to adapt to the large scale and high resolution remote sensing image registration, this paper proposes a registration control point matching method from coarse to fine. The small panel differential correction method is used to realize the high accuracy correction of large scale image, and an adaptive scan line filling algorithm is used to calculate the triangle in which each corrected pixel is located. In this paper, by analyzing the bottleneck problem in each step of registration, the CUDA parallel computing technique is used to accelerate the control point matching and image correction. The experiments on IKONOS panchromatic and multispectral images show that the proposed method can achieve high registration accuracy and obtain a high speedup ratio by using GPU acceleration algorithm. (2) Real-time online stitching of aerial images. Aiming at the bottleneck of online real-time mosaic of aerial images, this paper presents an online stitching method based on CPU and GPU. Based on the POS data, the monoclinic transformation between the original image and the corrected image is calculated at the CPU end, and then the correction process of the image is realized by using the GPU parallel computation. In this paper, an adaptive stitching method is proposed, that is, by calculating the overlap degree of subsequent images to determine whether the current images need to be spliced, the redundant computation is greatly reduced, and at the same time, two computers are used to carry out simulation experiments. One of them takes the image as an analog camera and transmits the image to the processing machine. The experimental results show that the method can basically realize the online real-time mosaic of aerial images, and the results of the stitching can provide the decision information for the real-time rescue of disasters and emergencies. In this paper, the proposed registration and splicing methods are implemented with Visual C, and the corresponding prototype system is developed. The experimental results show that the proposed method can achieve the efficient registration of large scale remote sensing images and the online stitching of aerial images. There are 26 figures, 3 tables and 48 references.
【學(xué)位授予單位】：中南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TP751
本文編號(hào)：2154430
[Abstract]:As a visual form of the energy or state of the objective world, image provides abundant information for people to recognize and transform the objective world. As one of the basic problems of image processing, image registration and image mosaic are widely used in the fields of virtual reality, geological survey, medical image, meteorological prediction, emergency response and so on. The graphics processor GPU uses more transistors as executive units, much more computing power than traditional central processor CPUs. At present, GPU technology has been widely used in the fields of data mining, mathematical statistics, image speech recognition, genetic engineering, accurate prediction of global climate and so on. At the same time, it provides a new solution for the rapid processing of remote sensing images. Aiming at the existing problems in image registration and stitching calculation, this paper analyzes the current image registration and stitching technology, and combines CUDA parallel computing technology. The registration method of large scale and high resolution remote sensing image and the online real-time mosaic method of aerial image are mainly studied in this paper. The main contents of this paper are as follows: (1) large scale remote sensing image registration. Aiming at the problem that the traditional remote sensing image registration method is difficult to adapt to the large scale and high resolution remote sensing image registration, this paper proposes a registration control point matching method from coarse to fine. The small panel differential correction method is used to realize the high accuracy correction of large scale image, and an adaptive scan line filling algorithm is used to calculate the triangle in which each corrected pixel is located. In this paper, by analyzing the bottleneck problem in each step of registration, the CUDA parallel computing technique is used to accelerate the control point matching and image correction. The experiments on IKONOS panchromatic and multispectral images show that the proposed method can achieve high registration accuracy and obtain a high speedup ratio by using GPU acceleration algorithm. (2) Real-time online stitching of aerial images. Aiming at the bottleneck of online real-time mosaic of aerial images, this paper presents an online stitching method based on CPU and GPU. Based on the POS data, the monoclinic transformation between the original image and the corrected image is calculated at the CPU end, and then the correction process of the image is realized by using the GPU parallel computation. In this paper, an adaptive stitching method is proposed, that is, by calculating the overlap degree of subsequent images to determine whether the current images need to be spliced, the redundant computation is greatly reduced, and at the same time, two computers are used to carry out simulation experiments. One of them takes the image as an analog camera and transmits the image to the processing machine. The experimental results show that the method can basically realize the online real-time mosaic of aerial images, and the results of the stitching can provide the decision information for the real-time rescue of disasters and emergencies. In this paper, the proposed registration and splicing methods are implemented with Visual C, and the corresponding prototype system is developed. The experimental results show that the proposed method can achieve the efficient registration of large scale remote sensing images and the online stitching of aerial images. There are 26 figures, 3 tables and 48 references.
【學(xué)位授予單位】：中南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TP751
【參考文獻(xiàn)】
相關(guān)期刊論文 前2條
1 方留楊;王密;李德仁;;CPU和GPU協(xié)同處理的光學(xué)衛(wèi)星遙感影像正射校正方法[J];測(cè)繪學(xué)報(bào);2013年05期
2 ;2008年我國(guó)自然災(zāi)害的主要特點(diǎn)[J];中國(guó)減災(zāi);2009年01期
相關(guān)博士學(xué)位論文 前4條
1 張樺;場(chǎng)景圖像拼接關(guān)鍵技術(shù)研究[D];天津大學(xué);2008年
2 馮宇平;圖像快速配準(zhǔn)與自動(dòng)拼接技術(shù)研究[D];中國(guó)科學(xué)院研究生院（長(zhǎng)春光學(xué)精密機(jī)械與物理研究所）;2010年
3 宋智禮;圖像配準(zhǔn)技術(shù)及其應(yīng)用的研究[D];復(fù)旦大學(xué);2010年
4 羅耀華;高性能計(jì)算在高光譜遙感數(shù)據(jù)處理中的應(yīng)用研究[D];成都理工大學(xué);2013年
，本文編號(hào)：2154430
本文鏈接：http://www.wukwdryxk.cn/guanlilunwen/gongchengguanli/2154430.html
最近更新
教材專著
