GPU Programlamada CUDA Platformu Kullanılan Paralel Görüntü İşleme Çalışmalarının İncelenmesi

Öz

Görüntü işleme pek çok alanda kullanılmaktadır. Görüntü işleme teknikleri gün geçtikçe görüntülerin çözünürlüklerinin artmasıyla daha fazla işlemci gücüne ihtiyaç duymaktadır. Görüntü işleme sürecini hızlandırmak için paralel görüntü işleme teknikleri kullanılmaktadır. GPU programlama günümüzde çok kullanılan ve tercih edilen paralel görüntü işleme tekniklerinden biridir. CUDA ise GPU programlamada en çok kullanılan platformdur. Bu çalışmanın temel amacı araştırmacılara ve konuya yeni başlayanlara görüntü işleme uygulamalarında GPU ve CUDA gibi donanım ve yazılım teknolojilerinin kullanımı konusunda bir başvuru kaynağı sağlamaktır. Bu amaç kapsamında çalışmada GPU ve CUDA kullanılarak yapılan görüntü işleme çalışmaları incelenmiş ve değerlendirilmiştir. GPU ve CUDA kullanan görüntü işleme çalışmaları, görüntü geriçatma, görüntü iyileştirme, görüntü bölütleme, görüntü çakıştırma ve görüntü sınıflandırma olmak üzere beş bölümde incelenmiş ve değerlendirilmiştir. Elde edilen sonuçlar doğrultusunda, GPU ve CUDA kullanımının avantajları ve bu teknolojilerin kullanıldığı görüntü işleme uygulamalarında dikkat edilmesi gereken hususlar belirlenmiştir.  

Anahtar Kelimeler

• Görüntü işleme,paralel programlama,GPU,CUDA

A Survey on Parallel Image Processing Studies Using CUDA Platform in GPU Programming

Öz

Image processing is used in a variety of fields. Image processing techniques need high processor performance due to increased image resolution day by day. Parallel processing techniques are used to satisfy the requirements related to high performance in real time image processing applications. Recently, GPU programming is one of the most commonly used and preferred methods in parallel processing. CUDA is the most popular platform in GPU programming. In this survey the studies where CUDA platform was used for image processing are presented and evaluated.  The major purpose of this survey is to provide a comprehensive reference source for the starters or researchers involved in use of CUDA platform in GPU programming for image processing techniques. Studies using CUDA platform in GPU programming have been classified under 5 areas; image reconstruction, image enhancement, image segmentation, image registration and image classification. Advantages of using CUDA in GPU programming for image processing and issues to pay attention in applications have also been underlined.

Anahtar Kelimeler

• Image processing,parallel computing,GPU,CUDA

Kaynakça

1. Brodtkorb A.R., Hagen T.R., Saetra M.L., “Graphics processing unit (GPU) programming strategies and trends in GPU computing”, Journal of Parallel and Distributed Computing, 73: 4-13, (2013)
2. Cook S., CUDA Programming A developer’s Guide to Parallel Computing with GPUs, Elsevier, USA, (2013).
3. Schneider M., Fey D., Kapusi D., Macheidt T., “Performance comparison of designated preprocessing white light interferometry algorithms on emerging multi- and many-core architectures”, Procedia Computer Science, 4:2037–2046, (2011).
4. Ruetsch G., Oster B., Getting Started with CUDA, NVISION08: The World of Visual Computing, California USA, August 25-27, (2008).
5. NVIDIA CUDA Compute Unified Device Architecture Programming Guide, Version 1.0, (2007)
6. NVIDIA Accelerated Computing. GPU-Accelerated Libraries. Yayın tarihi 2017. Erişim tarihi Nisan 10, (2017).
7. Es A., İşler V., “Accelerated regular grid traversals using extended anisotropic chessboard distance fields on a parallel stream processor”, Journal of Parallel and Distributed Computing, 67: 1201-1217, (2007)
8. Göddeke D., Strzodka R., Mohd-Yusof J., McCormick P., Buijssen S.H.M., Grajewski M., Turek S., “Exploring weak scalability for FEM calculations on a GPU-enhanced cluster”, Parallel Computing, 33: 685-699, (2007)

9. Zwart S.F.P., Belleman R.G., Geldof P.M., “High-performance direct gravitational N-body simulations on graphics processing units”, New Astronomy, 12: 641-650, (2007)
10. Anderson A.G., Goddard III W.A., Schröder P., “Quantum Monte Carlo on graphical processing units”, Computer Physics Communications, 177: 298–306, (2007)
11. Martinsen P., Blaschke J., Künnemeyer R., Jordan R., “Accelerating Monte Carlo simulations with an NVIDIA graphics processor”, Computer Physics Communications, 180: 1983–1989, (2009).
12. Che S., Boyer M., Meng J., Tarjan D., Sheaffer J.W., SKadron K., “A performance study of general-purpose applications on graphics processors using CUDA”, Journal of Parallel and Distributed Computing, 68: 1370–1380, (2008)
13. Diez D.C., Mueller H., Frangakis A.S., “Implementation and performance evaluation of reconstruction algorithms on graphics processors”, Journal of Structural Biology, 157: 288-295, (2007)
14. Stone S.S., Haldar J.P., Tsao S.C., Hwu W.W., Sutton B.P., Liang Z.P., “Accelerating advanced MRI reconstructions on GPUs”, Journal of Parallel and Distributed Computing, 68: 1307-1318, (2008)
15. Schenk O., Christen M., Burkhart H., “Algorithmic performance studies on graphics processing units”, Journal of Parallel and Distributed Computing, 68: 1360-1369, (2008)
16. Diez D.C., Moser D., Schoenegger A., Pruggnaller S., Frangakis A.S., “Performance evaluation of image processing algorithms on the GPU”, Journal of Structural Biology, 164: 153–160, (2008)
17. Belleman R.G., Bedorf J., Zwart S.F.P., “High performance direct gravitational N-body simulations on graphics processing units II: An implementation in CUDA”, New Astronomy, 13: 103-112, (2008)
18. Ram M.P.M., Kurfess T.R., Tucker T.M., “Least-squares fitting of analytic primitives on a GPU”, Journal of Manufacturing Systems, 27: 130-135, (2008)Ryoo S., Rodrigues C.I., Stone S.S., Stratton J.A., Ueng S., “Program optimization carving for GPU computing”, Journal of Parallel and Distributed Computing, 68: 1389-1401, (2008)
19. Walsh S.D.C., Saar M.O., Bailey P., Lilja D.J., “Accelerating geoscience and engineering system simulations on graphics hardware”, Computers & Geosciences, 35: 2353–2364, (2009)
20. Komatitsch D., Michea D., Erlebacker G., “Porting a high-order finite-element earthquake modeling application to NVIDIA graphics cards using CUDA”, Journal of Parallel and Distributed Computing, 69: 451-460, (2009)
21. Vazquez F., Garzon E.M., Femandez J.J., “A matrix approach to tomographic reconstruction and its implementation on GPUs”, Journal of Structural Biology, 170: 146–151, (2010)
22. Noel P.B., Walczak A.M., Xu J., Corso J.J. Hoffmann K.R., Schafer S., “GPU-based cone beam computed tomography”, Computer Methods and Programs in Biomedicine, 98: 271–277, (2010)
23. Zheng S.Q., Branlund E., Kesthelyi B., Braunfeld M.B., Cheng Y., Sedat J.W., Agard D.A., “A distributed multi-GPU system for high speed electron microscopic tomographic reconstruction”, Ultramicroscopy, 111: 1137–1143, (2011)
24. Herraiz J.J., Espana S., Cal-Gonzalez J., Vaquero J.J., Desco M., Udias J.M., “Fully 3D GPU PET reconstruction”, Nuclear Instruments and Methods in Physics Research A, 648: S169–S171, (2011)
25. Xu W., Xu F., Jones M., Keszthelyi B., Sedat J., Agard D., Mueller K., “High-performance iterative electron tomography reconstruction with long-object compensation using graphics processing units (GPUs)”, Journal of Structural Biology, 171: 142–153, (2010)
26. Xu F., Xu W., Jones M., Keszthelyi B., Sedat J., Agard D., Mueller K., “On the efficiency of iterative ordered subset reconstruction algorithms for acceleration on GPUs”, Computer Methods and Programs in Biomedicine, 98: 261–270, (2010).
27. Palenstjn W.J., Batenburg K.J., Sijbers J., “Performance improvements for iterative electron tomography reconstruction using graphics processing units (GPUs)”, Journal of Structural Biology, 176: 250–253, (2011)
28. Blas J.G. Abella M., Isaila F., Carretero J., Desco M., “Surfing the optimization space of a multiple-GPU parallelimplementation of a X-ray tomography reconstruction algorithm”, The Journal of Systems and Software, 95: 166–175, (2014)
29. Flores L. A., Vidal V., Mayo P., Rodenas F., Verdu G., “CT Image Reconstruction Based on GPUs”, Procedia Computer Science, 18: 1412–1420, (2013)
30. Flores L.A., Vidal V., Mayo P., Rodenas F., Verdu G., “Parallel CT image reconstruction based on GPUs”, Radiation Physics and Chemistry, 95: 247–250, (2014)
31. Feng C., Yang J., Zhao D., Liu J., “CUDA accelerated method for motion correction in MR PROPELLER imaging”, Magnetic Resonance Imaging, 31: 1390–1398, (2013)
32. Li X., Grigorieff N., Cheng Y., “GPU-enabled FREALIGN: Accelerating single particle 3D reconstruction and refinement in Fourier space on graphics processors”, Journal of Structural Biology, 172: 407–412, (2010)
33. Okitsu Y., Ino F., Hagihara K., “High-performance cone beam reconstruction using CUDA compatible GPUs”, Parallel Computing, 36: 129–141, (2010)
34. Scherl H., Kowarschik M., Hofmann H.G., Keck B., Hornegger J., “Evaluation of state-of-the-art hardware architectures for fast cone-beam CT reconstruction”, Parallel Computing, 38: 111–124, (2012)
35. Chang J.Y., Park H., Park I.K., Lee K.M., Lee S.U., “GPU-friendly multi-view stereo reconstruction using surfel representation and graph cuts”, Computer Vision and Image Understanding, 115: 620–634, (2011)
36. Jiang N., Yang W., Duan L., Xu X., Huang C. Liu Q., “Acceleration of CT reconstruction for wheat tiller inspection based on adaptive minimum enclosing rectangle”, Computers and Electronics in Agriculture, 85: 123–133, (2012)
37. Agulleiro J.I., Vazquez F., Garzon E.M., Femandez J.J., “Hybrid computing:CPU+GPU co-processing and its application to tomographic reconstruction”, Ultramicroscopy, 115: 109–114, (2012)
38. Yang J., Feng C., Zhao D., “A CUDA-based reverse gridding algorithm for MR reconstruction”, Magnetic Resonance Imaging, 31: 313–323, (2013)
39. Piccialli F., Cuomo S., Michele P.D., “A regularized MRI image reconstruction based on Hessian penalty term on CPU/GPU systems”, Procedia Computer Science, 18: 2643–2646, (2013)
40. Monte C.F.P., Pccoli F., Luciano C., Rizzi S., Bianchini G., Scutari P.C., “Estimation of Volume Rendering Efficiency with GPU in a Parallel Distributed Environment”, Procedia Computer Science, 18: 1402–1411, (2013)
41. Gai J., Obeid N., Holtrop J.L., Wu X.W., Lam F., Fu M., Haldar J.P., Hwu W.W. Liang Z. Sutton B.P., “More IMPATIENT: A gridding-accelerated Toeplitz-based strategy for non-Cartesian high-resolution 3D MRI on GPUs”, Journal of Parallel and Distributed Computing, 73: 686–697, (2013)
42. Birk M., Balzer M., Ruiter N.V., Becker J.,” Evaluation of performance and architectural efficiency of FPGAs and GPUs in the 40 and 28 nm generations for algorithms in 3D ultrasound computer tomography”, Computers and Electrical Engineering, 40: 1171–1185, (2014)
43. Chang L., El-Araby E., Dang V.Q., Dao L.H., “GPU acceleration of nonlinear diffusion tensor estimation using CUDA and MPI”, Neurocomputing, 135: 328–338, (2014)
44. Orts-Escolano S., Garcia-Rodriguez J., Serra-Perez J.A., Jimeno-Morenilla A., Garcia-Garcia A., Morell V., Cazorla M., “3D model reconstruction using neural gas accelerated on GPU”, Applied Soft Computing, 32: 87–100, (2015)
45. Huhle B., Schairer T., Jenke P., Straber W., “Fusion of range and color images for denoising and resolution enhancement with a non-local filter”, Computer Vision and Image Understanding, 114: 1336–1345, (2010)
46. Sanchez M.G., Vidal V., Bataller J., Arnal J., “A Parallel Method for Impulsive Image Noise Removal on Hybrid CPU/GPU Systems”, Procedia Computer Science, 18: 2504–2507, (2013)
47. Urena R., Morillas C., Pelayo F., “Real-time bio-inspired contrast enhancement on GPU”, Neurocomputing, 121: 40–52, (2013)
48. Ma T., Li L., Ji S., Wang X., Tian Y., Al-Dhelaan A., Al-Rodhaan., “Optimized Laplacian image sharpening algorithm based on graphic processing unit”, Physica A, 416: 400–410, (2014)
49. Saha D., Darji K., Patel N., Thakore D., “Procedia Computer Science, Implementation of Image Enhancement Algorithms and Recursive Ray Tracing using CUDA”, 7th International Conference on Communication, Computing and Virtualization, 79: 516–524, (2016)
50. Demirhan A., Güler İ., “Özörgütlemeli harita ağları ve gri düzey eş oluşum matrisleri ile görüntü bölütleme”, Journal of the Faculty of Engineering and Architecture of Gazi University. Cilt 25, No2: 285-291, (2010)
51. Uslu R., “Elektronik bir hücresel yapay sinir ağı gerçekleşmesi olan ACE16K üzerinde görüntü bölütleme”, İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, Yüksek Lisans Tezi, (2007)
52. Pan L., Gu L., Xu J., “Implementation of Medical Image Segmentation in CUDA”, 5th International Conference on Information Technology and Application in Biomedicine, Shenzhen, China, 30-31, Mayıs, (2008)
53. Zhuge Y., Cao Y., Miller R.W., “GPU Accelerated Fuzzy Connected Image Segmentation by using CUDA”, 31st Annual International Conference of the IEEE EMBS, Minneapolis, Minnesota, USA, 2-6 Eylül, (2009)
54. Abramov A., Kulvicius T., Worgotter F., Dellen B., “Real-Time Image Segmentation on a GPU”, Facing the Multicore-Challenge:Aspects of New Paradigms and Technologies in Parallel Computing, 6310: 131-142, (2010)
55. Sui H., Peng F., Xu C., Sun K., Gong J., “GPU-accelerated MRF segmentation algorithm for SAR images”, Computers & Geosciences, 43: 159–166, (2012)
56. Diaz-Pernil D., Berciano A., Pena-Cantillana F., Gutierrez-Naranjo M.A., “Segmenting images with gradient-based edge detection using Membrane Computing”, Pattern Recognition Letters, 34: 846–855, (2013)
57. Özdemir A., Altılar T., “GPU Based Parallel Image Processing for Plant Growth Analysis”, Third International Conference on Agro-Geoinformatics, 315-320, (2014)
58. Alvarado R., Tapia J.J., Rolon J.C., “Medical image segmentation with deformable models on graphics processing units”, Journal of Supercomputing, 68:, 339-364, (2014)
59. Bergen R.V., Lin H., Alexander M.E., Bidinosti C.P., “4D MR phase and magnitude segmentations with GPU parallel computing”, Magnetic Resonance Imaging, 33: 134–145, (2015)
60. D’Ambra P., Pilippone S., “A parallel generalized relaxation method for high-performance image segmentation on GPUs”, Journal of Computational and Applied Mathematics, 293: 35-44, (2016)
61. Singh B.M., Sharma R., Mittal A., Ghosh D., “Parallel Implementation of Otsu’s Binarization Approach on GPU”, Internatinal Journal of Computer Applications, 32(2): 16-21, (2011)
62. Jin X., Kang D.J., Jeong M., “GPU-Based Real-Time Range Image Segmentation”, Intelligent Computing Methodologies, 8589: 293-297, (2014)
63. Wang J., Chen W., “MRI Image Segmentation Based on a GPU Shortest Path Algorithm”, 2nd International Conference on Intelligent Computing and Cognitive Informatics, 1-4, (2015)
64. Smistad E., Thomas L.F., Bozorgi M., Elster A.C., Lindseth F., “Medical Image Segmentation on GPUs- A Comprensive Reiview”, Medical Image Analysis, 20(1): 1-18, (2015)
65. D’Ambra P., Pilippone S., “A parallel generalized relaxation method for high-performance image segmentation on GPUs”, Journal of Computational and Applied Mathematics, 293: 35-44, (2016)
66. Diaz-Pernil D., Fondon I., Pena-Cantillana F., Gutierrez-Naranjo M.A., “Fully automatized parallel segmentation of the optic disc in retinal fundus images”, Pattern Recognition Letters, 83 (1): 97-107, (2016)
67. Jaros M., Strakos P., Karasek T., Riha L., Vasatova A., Jarosova M., Kozubek T., “Implementation of K-means segmentation algorithm on Intel Xeon Phi and GPU: Application in medical imaging”, Advances in Engineering Software, 103: 21-28, (2017)
68. Muyan-Özçelik P., Samant J.X.S.S., “Fast Deformable Registration on the GPU:A CUDA Implementation of Demons”, International Conference on Computational Sciences and Its Applications, 223-233, (2008)
69. Gu X., Pan H., Liang Y., Castillo R., Yang D., Choi D., Castillo E., Majumdar A., Guerrero T., Jiang S.B., “Implementation and evaluation of various demens deformeble image registration algorithms on a GPU”, Physics in Medicine and Biology, 55: 207:219, (2010)
70. Huang T.Y., Tang Y., Ju S., “Accelerating image registration of MRI by GPU-based parallel computation”, Magnetic Resonance Imaging, 29: 712–716, (2011)
71. Ruijters D., Romeny B.M.H., Suetens P., “GPU-accelerated elastic 3D image registration for intra-surgical applications”, Computer Methods and Programs in Biomedicine, 103: 104–112, (2011)
72. Park S., Choi S., Kim J., Chae J.S., “Real-time 3D registration using GPU”, Machine Vision and Applications, 22: 837-850, (2011)
73. Dorgham O.M., Laycock S.D., Fisher M.H., “GPU Accelerated Generation of Digitally Reconstructed Radiographs for 2D/3D Image Registration”, IEEE Transactions on Biomedical Engineering, 59 (9): 2594-2603, (2012)
74. Sah S., Vanek J., Roh Y., Wasnik R., “GPU Accelerated Real Time Rotation, Scale and Translation Invariant Image Registration Method”, Image Analysis and Recognition, 7324: 224-233, (2012)
75. Lu M., “3D Medical Images Registration Based on GPU Parallel Computing”, Applied Mechanics and Materials, 241-244: 3010-3013, (2012)
76. Marchelli G., Haynor D., Ledoux W., Tsai R., Storti D., “A Flexible toolkit for rapid GPU-based generation of DDRs for 2D-3D registration”, Medical Imaging 2013: Image Processing, 8669, (2013)
77. Zhang Y., Zhou P., Ren Y., Zou Z., “GPU-accelerated large-size VHR images registration via coarse-to-fine matching”, Computers & Geosciences, 66: 54–65, (2014)
78. Li A., Kumar A., Ha Y., Corporaal H., “Correlation ratio based volume image registration on GPUs”, Microprocessors and Microsystems, 39 (8): 998-1011, (2015)
79. GISGeography. Image Classification Techniques in Remote Sensing. Yayın tarihi January 22, 2017. Erişim tarihi Nisan 5, (2017)
80. BEÜ Mühendislik Fakültesi. Görüntü Sınıflandırma. Yayın tarihi 2016. Erişim tarihi Nisan 2, (2017)
81. Gumbau J., Chover M., Remolar I., Rebollo C., “View-dependent pruning for real-time rendering of trees”, Computers & Graphics, 35: 364–374, (2011)
82. Cesnovar R., Risojevic V., Babic Z., Dobravec T., Bulic P., “A GPU implementation of a structural-similarity-based aerial-image classification”, J. Supercomputing, 65: 978-996, (2013)
83. Haythem B., Mohamed H., Marwa C., Fatma S., Mohamed A., “Fast Generalized Fourier Descriptor for object recognition of image using CUDA”, World Symposium on Computer Applications and Research (WSCAR), : Sousse, TUNISIA, JAN 18-20, (2014)
84. Ganianu M., Onchiş D.M., “Face and marker detection using Gabor frames on GPUs”, Signal Processing, 96: 90–93, (2014)
85. Werff H.M.A., Bakker W.H., “Implementation and performance of a general purpose graphicsprocessing unit in hyperspectral image analysis”, International Journal of Applied Earth Observation and Geoinformation, 26: 312–321, (2014)
86. Lopez-Fandino J., Quesada-Barriuso P., Heras D.B., Argüello F., “Efficient ELM-Based Techniques for the Classification of Hyperspectral Remote Sensing Images on Commodity GPUs”, IEEE Journal Of Selected Topıcs In Applıed Earth Observatıons And Remote Sensıng, 8 (6): 2884-2893, (2015)
87. Alçin Ö.F., Şengür A., İnce M.C., “İleri-geri takip algoritması tabanlı seyrek aşırı öğrenme makinesi”, Journal of the Faculty of Engineering and Architecture of Gazi University, 30 (1): 111-117, (2015)
88. Kaya Y., Tekin R., “Epileptik Nöbetlerin Tespiti için Aşırı Öğrenme Makinesi Tabanlı Uzman Bir Sistem”, Bilişim Teknolojileri Dergisi, 5 (2): 33-39, (2012)
89. Shi L., Liu W., Zhang H., Xie Y., Wang D., “A survey of GPU-based medical image computing techniques”, Quant Imaging Med Surg, 2(3): 188-206, (2012)
90. Sugimoto Y., Ino F., Hagihara K., “Improving cache locality for GPU-based volume rendering”, Parallel Computing, 40: 59–69, (2014)
91. Nam W., Han B., Han J.H., “Macrofeature layout selection for pedestrian localization and its acceleration using GPU”, Computer Vision and Image Understanding, 120: 46–58, (2014)
92. Liu B., Telea A.C., Roerdink J.B.T.M., Clapworthy G.J., Williams D., Yang P., Dong F., Codreanu V., Chiarini A., “Parallel centerline extraction on the GPU”, Computers & Graphics, 41: 72–83, (2014)
93. Zhang J., You S., Greunwald L., “Parallel online spatial and temporal aggregations on multi-core CPUs and many-core GPUs”, Information Systems, 44: 134–154, (2014)
94. Galizia A., D’Agostino D., Clematis A., “An MPI–CUDA library for image processing on HPC architectures”, Journal of Computational and Applied Mathematics, 273: 414–427, (2015)
95. Qin L., Sheng B., Lin W., Wu W., Shen R., “GPU-Accelerated Video Background Subtraction Using Gabor Detector”, J. Vis. Commun. Image R., 32: 1-9, (2015)
96. Lastra M., Carabano J., Gutierrez P.D., Benitez J.M., Herrera F., “Fast fingerprint identification using GPUs”, Information Sciences, 301: 195-214, (2015)
97. Cappelli R., Ferrara M., Maltoni D., “Large-scale fingerprint identification on GPU , Information Sciences, 306: 1-20, (2015)
98. Hermandez-Juarez D., Chacon A., Espinosa A., Vazquez D., Moure J.C., Lopez M., “Embedded real-time stereo estimation via Semi-Global Matching on the GPU”, Procedia Computer Science, ICCS 2016. The International Conference on Computational Science, , 80, 143–153, (2016)
99. Wang H., Zhang N., Creput J., Ruichek Y., Moreau J., “Massively parallel GPU computing for fast stereo correspondence algorithms”, Journal of Systems Architecture, 65: 46-58, (2016)