本文關(guān)鍵詞： 高光譜 特征提取 CUDA 多平臺(tái) GPU 并行計(jì)算　出處：《重慶郵電大學(xué)》2016年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著高光譜圖像開(kāi)始呈現(xiàn)出高時(shí)間分辨率、高空間分辨率、高光譜分辨率的“三高”特性,數(shù)據(jù)量也隨之出現(xiàn)海量增長(zhǎng)的趨勢(shì),使其處理起來(lái)越為復(fù)雜、耗時(shí),對(duì)其進(jìn)行處理的硬件平臺(tái)環(huán)境的要求也大大提高。高光譜圖像的近實(shí)時(shí)、實(shí)時(shí)處理成為一個(gè)難題。目前,國(guó)內(nèi)外的很多學(xué)者提出了從一種平臺(tái)縱向的研究探討高光譜圖像的處理,但幾乎都是從單一平臺(tái)或單方面入手探討對(duì)高光譜圖像的處理。而對(duì)高光譜圖像處理的語(yǔ)言、平臺(tái)繁多,缺乏將多種語(yǔ)言、平臺(tái)資源整合、分析比較,以及歸類整理。本文正是基于這條思路,從橫向和縱向方面針對(duì)高光譜圖像數(shù)據(jù)的特點(diǎn),結(jié)合對(duì)其處理的語(yǔ)言、平臺(tái)環(huán)境入手,以處理時(shí)間、加速比、圖像處理效果、能耗、復(fù)雜度等方面作為評(píng)價(jià)指標(biāo),探尋一種真正適宜于高光譜圖像處理的語(yǔ)言、平臺(tái)環(huán)境。從橫向和縱向方面,對(duì)高光譜圖像處理平臺(tái)適宜性進(jìn)行研究提供了一種新思路。本文的主要研究?jī)?nèi)容如下:首先,對(duì)CUDA并行架構(gòu)的編程模型、存儲(chǔ)器模型、處理機(jī)制進(jìn)行了分析,并探討了基于CUDA程序性能優(yōu)化問(wèn)題。其次,從典型的高光譜圖像特征提取算法入手,探討其缺陷和不足,然后在基于CUDA架構(gòu)的GPU并行環(huán)境下對(duì)最小噪聲分離和主成分分析兩種算法,從數(shù)據(jù)通信、數(shù)據(jù)分塊、存儲(chǔ)器訪問(wèn)、協(xié)方差計(jì)算等方面進(jìn)行優(yōu)化和改進(jìn),達(dá)到對(duì)高光譜圖像特征提取加速處理的目的。通過(guò)仿真實(shí)驗(yàn)發(fā)現(xiàn),優(yōu)化改進(jìn)后兩種算法最高的加速比達(dá)到了122倍,優(yōu)化了處理時(shí)間,提升了加速比。最后,針對(duì)目前處理高光譜圖像的語(yǔ)言、平臺(tái)環(huán)境繁多各異的情形,提出一種多平臺(tái)機(jī)制(ENVI、Matlab、串、并行環(huán)境平臺(tái))對(duì)高光譜圖像進(jìn)行特征提取仿真實(shí)驗(yàn)探究,對(duì)實(shí)驗(yàn)結(jié)果從橫向方面研究比較,評(píng)價(jià)各種處理平臺(tái)的綜合優(yōu)劣,為高光譜圖像處理平臺(tái)的適應(yīng)性提供一種研究新思路。進(jìn)而為高光譜圖像的分類、目標(biāo)探測(cè)、混合像元分解等后續(xù)研究工作的開(kāi)展打下了基礎(chǔ),為高光譜圖像的快速和高效處理帶來(lái)了可能。
[Abstract]:As hyperspectral images begin to exhibit the "three high" characteristics of high time resolution, high spatial resolution, and high spectral resolution, the amount of data is increasing rapidly, which makes the processing more complex and time-consuming. The requirement of hardware platform for processing is also greatly improved. The near real time and real time processing of hyperspectral image has become a difficult problem. Many scholars at home and abroad have proposed to discuss the processing of hyperspectral images from the longitudinal study of a platform, but almost all of them discuss the processing of hyperspectral images from a single platform or unilaterally, and the language of hyperspectral image processing. There are many platforms, lack of integration of multiple languages, platform resources integration, analysis and comparison, and classification. Based on this idea, this paper aims at the characteristics of hyperspectral image data from the horizontal and vertical aspects, combined with the processing language of hyperspectral image data. The platform environment starts with processing time, speedup ratio, image processing effect, energy consumption, complexity and so on as the evaluation index, explores a kind of language which is really suitable for hyperspectral image processing, platform environment. The main contents of this paper are as follows: firstly, the programming model, memory model and processing mechanism of CUDA parallel architecture are analyzed. The performance optimization problem based on CUDA program is discussed. Secondly, starting with the typical feature extraction algorithm of hyperspectral image, the defects and shortcomings of the algorithm are discussed. Then in the GPU parallel environment based on CUDA architecture, the two algorithms of minimum noise separation and principal component analysis are optimized and improved from the aspects of data communication, data block, memory access, covariance calculation and so on. The simulation results show that the maximum speedup ratio of the two improved algorithms is 122 times, the processing time is optimized, and the speedup ratio is increased. In view of the situation that hyperspectral images are processed in different languages and different platforms at present, a multi-platform mechanism named ENVI Matlab, string and parallel environment is proposed to simulate the feature extraction of hyperspectral images. The experimental results are compared horizontally, and the comprehensive advantages and disadvantages of various processing platforms are evaluated, which provides a new way of research for the adaptability of hyperspectral image processing platform, and then for the classification of hyperspectral images and target detection. The following research work, such as mixed pixel decomposition, has laid a foundation for the rapid and efficient processing of hyperspectral images.
【學(xué)位授予單位】：重慶郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP751

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 湯媛媛;周海芳;方民權(quán);申小龍;;基于CPU/GPU異構(gòu)模式的高光譜遙感影像數(shù)據(jù)處理研究與實(shí)現(xiàn)[J];計(jì)算機(jī)科學(xué);2016年02期

2 裴頌文;寧?kù)o;張俊格;;CPU-GPU異構(gòu)多核系統(tǒng)的動(dòng)態(tài)任務(wù)調(diào)度算法[J];計(jì)算機(jī)應(yīng)用研究;2016年11期

3 劉洪波;劉曉敏;鄭永永;趙云偉;;MATLAB在圖形邊界特征提取中的應(yīng)用[J];機(jī)械研究與應(yīng)用;2015年06期

4 梁亮;朱超;楊捷;吳素萍;;高光譜圖像預(yù)處理的Matlab并行化研究[J];計(jì)算機(jī)工程與設(shè)計(jì);2015年08期

5 白t，

本文編號(hào)：1529464