跳到主要內容

臺灣博碩士論文加值系統

(44.192.22.242) 您好!臺灣時間:2021/08/05 11:36
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:王旭
研究生(外文):Hsu Wang
論文名稱:基於GPU之粒子群優法應用於高光譜影像波段選取
論文名稱(外文):Particle Swarm Optimization for Hyperspectral Band Selection Using GPU
指導教授:張陽郎張陽郎引用關係方志鵬方志鵬引用關係
指導教授(外文):Yang-Lang ChangJyh-Perng Fang
口試委員:蔡昌隆王元凱
口試委員(外文):Chang-Lung TsaiYuan-Kai Wang
口試日期:2012-07-03
學位類別:碩士
校院名稱:國立臺北科技大學
系所名稱:電機工程系所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2012
畢業學年度:100
語文別:中文
論文頁數:51
中文關鍵詞:粒子群優法波段選取相關係數矩陣降維CUDA
外文關鍵詞:Particle Swarm Optimization (PSO)band selectioncorrelation coefficient matrixdimension reductionCompute Unified Device Architecture (CUDA)
相關次數:
  • 被引用被引用:16
  • 點閱點閱:137
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
隨著衛星遙測技術近年來廣泛的發展,高光譜影像的波段數與資料量也越趨龐大,使計算複雜度大幅提高,波段中也可能包含雜訊或是錯誤的資訊導致分類正確率降低。因此,在高光譜影像處理中,進行波段選取降低資料複雜度並萃取具代表性的波段,是不可或缺的一個步驟。
過去已有學者以粒子群優法(Particle Swarm Optimization, PSO)進行高光譜影像的波段選取。利用PSO演算法將高光譜影像的相關係數矩陣(Correlation Coefficient Matrix)聚合成一組群聚模組特徵空間,挑選出具代表性的波段,達到降維(Reduction Dimension, RD)的效果。然而處理波段數較多的高光譜影像時,依然需耗費大量時間。因此本論文應用CUDA(Compute Unified Device Architecture)技術實現平行架構的粒子群優法,利用圖形處理器(Graphics Processing Units, GPU)進行加速,更進一步提升高光譜影像波段選取的整體運算速度。
本文採用鰲鼓溼地的 MASTER 遙測影像以及 Northwest Tippecanoe County 的AVIRIS 遙測影像為實驗圖資。最後由實驗結果可以得知,本文所提出的平行粒子群優法波段選取能夠迅速、有效地挑選出有價值的波段,並透過分類器得到良好的分類效果。


In recent years, the satellite image technologies have greatly advanced remote sensing community, resulting in the increased number of bands acquired by hyperspectral sensors. The band selection of hyperspectral imagery can reduce the dimensions which can avoid the Hughes phenomena. Therefore, the band selection of hyperspectral imagery has become very important. A band selection algorithm based on particle swarm optimization (PSO) is proposed in this paper. By using the PSO algorithm, the highly correlated bands of hyperspectral imagery can be grouped into the same modules which can extract the most useful information of hyperspectral bands in each module and can further reduce the dimensionality. However the PSO band selection is a time-consuming procedure when the number of hyperspectral bands is huge. Consequently this paper proposes a parallel PSO (PPSO) band selection based on modern graphics processing unit (GPU) architecture using NVIDIA compute unified device architecture (CUDA) technology. It can improve the computational speed of PSO band selection processes. The natural parallelism of proposed PPSO is in the face that each particle can be regarded as an independent agent. Parallel computation benefits the algorithm by providing each agent with one of the parallel processors. The intrinsic parallel characteristics embedded in PPSO can be therefore suitable for a parallel implementation. The effectiveness of the proposed PPSO is evaluated by AVIRIS hyperspectral images. The experimental results demonstrated that the proposed PPSO band selection not only can improve the computational speed but also can offer a satisfactory classification performance.

摘要 i
ABSTRACT ii
誌謝 iii
目錄 iv
表目錄 vi
圖目錄 vii
第一章 緒論 1
1.1研究背景介紹 1
1.2 研究動機與目的 3
1.3 論文內容大綱 4
第二章 相關文獻回顧 5
2.1 高光譜影像介紹 5
2.2相關係數矩陣 8
2.3貪婪模組特徵空間 8
2.4 粒子群優演算法 10
2.4.1 理論基礎簡介 10
2.4.2演算法流程 10
2.4.3粒子移動公式及參數 12
2.4.4適應值函數及空間轉換 13
2.5非純度維度優先權 16
2.6平行架構CUDA 17
2.6.1 GPGPU 17
2.6.2CUDA 19
第三章 研究方法 22
3.1高光譜影像分類方法 22
3.2 PSO平行架構 23
3.3適應值計算平行架構 24
3.3.1 粒子群平行架構 24
3.3.2 適應值計算平行架構 24
3.4粒子更新階段平行架構 26
3.4.1.更新自身最佳位置 26
3.4.2更新全域最佳位置 27
3.5粒子移動階段平行架構 28
3.6.1平行工作運行流程 29
3.6.2執行緒分配方法 30
3.6.3共享記憶體優化 31
第四章 實驗結果 33
4.1 使用圖資介紹 33
4.1.1鰲鼓溼地 33
4.1.2 Northwest Tippecanoe County 35
4.2 實驗目的與實驗環境 37
4.3實驗結果 38
4.3.1 粒子群優法參數設定 38
4.3.2粒子群優法波段選取在不同門檻值表現 38
4.3.3世代數對平行粒子群優法波段選取的影響 41
4.3.4粒子數對平行粒子群優法波段選取的影響 44
4.3.5粒子群優法各階段運算時間分析 47
第五章 結論與未來展望 48
5.1 結論 48
5.2未來展望 48
參考文獻 49


書籍
[1]David B. Kirk, Wen-meiW.Hwu, “Programming Massively Parallel Processors: A Hands-on Approach”, Elsevier Science Ltd.
期刊論文
[2]G. F. Hughes, "On the mean accuracy of statistical pattern recognizers, " IEEE Transactions on Information Theory, vol. 14 no. 1, 1968, pp. 55–63.
[3]Yang-Lang Chang, Chin-Chuan Han, Kuo-Chin Fan, K.S. Chen, Chia-Tang Chen and Jeng-Horng Chang, "Greedy Modular Eigenspaces and Positive Boolean Function for Supervised Hyperspectral Image Classification," Optical Engineering, Vol. 42, no. 9, 2003, pp. 2576-2587.
[4]Yang-LangChang, Chin-Chuan Han, Bormin Huang, Wen-Yen Chang, Benediktsson, J.A.,Lena Chang, "A Parallel Simulated Annealing Approach to Band Selection for High-Dimensional Remote Sensing Images," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 4, 2011, pp. 579 - 590 .
[5]James Kennedy and Russell Eberhart, "Particle Swarm Optimization" Prooceedings of IEEE Interantional Conference on Neural Networks, Piscataway, 1995 , pp. 1942–1948.
[6]WeiPang, Kang-Ping Wang, Chun-Guang Zhou, "Modified Particle Swarm Optimization Based On Space Transformation For Solving Traveling Salesman Problem" 2004, pp. 2342-2346.
[7]K. Pearson , "On Lines and Planes of Closest Fit to Systems of Points in Space" Philosophical Magazine, vol. 2, no. 6, 1901,pp. 559–572.
[8]Wei Wei, Qian Du, Nicolas H. Younan, "Unsupervised Hyper spectral Band Selection Using Graphics Processing Units,"SPIE JournalofAppliedRemote Sensing, vol. 6, 2012.
會議論文
[9]Yang-Lang Chang,Bin-FengShu,Tung-Ju Hsieh, Chih-Yuan Chu,Jyh-Perng Fang,"Band selection for hyperspectral images based on impurity function," IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Vancouver, USA, 2011, pp. 2369 - 2372 .
學位論文
[10]Yang-Lang Chang, A Novel Approach to Hyperspectral Image Classification, Ph.D. Thesis, National Central University, Taiwan, 2003.
[11]Kenneth A De Jong, An analysis of the behavior of a class of genetic adaptive systems, Ph.D. Thesis, University of Michigan, Michigan, 1975.
[12]王宏原,平行粒子群優法應用於高維度影像特徵抽取,碩士論文,國立臺北科技大學電機工程研究所,台北,2010。
[13]徐斌峰,一個維度優先權方法應用於粒子群優法在高維度影像特徵抽取,碩士論文,國立臺北科技大學電機工程研究所,台北,2011。
其他
[14]NVIDIA(2008). “CUDA Programming Guide v2.0”, NVIDIA.
[15]Jason Sanders, Edward Kandrot, “CUDA by Example: An Introduction to General-Purpose GPU Programming”
[16]科技小百科,http://www.narl.org.tw/upload/tw/publication/file/46/44。存取日期:2012-06-29。
[17]k-nearest neighbor algorithm ,Wikipedia , http://en.wikipedia.org/wiki/K-nearest_neighbor_algorithm.Accessed 29 June 2012.
[18]Remote Sensing ,The Image Analysis Processing&Protectiongroup , http://iapp.det.unifi.it/index.php/english/research_en/remotesensing_en.
Accessed 29 June 2012.


電子全文 電子全文(本篇電子全文限研究生所屬學校校內系統及IP範圍內開放)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊