(JBE Vol. 24, No. 1, January 2019) (Regular Paper) 24 1, (JBE Vol. 24, No. 1, January 2019) ISSN 2287

Transcription

1 (Regular Paper) 24 1, (JBE Vol. 24, No. 1, January 2019) ISSN (Online) ISSN (Print) a), a), a), a), b), b), a), a) Characteristic Analysis for Compression of Digital Hologram Jin-Kyum Kim a), Kyung-Jin Kim a), Woo-Suk Kim a), Yoon-Huck Lee a), Kwan-Jung Oh b), Jin-Woong Kim b), Dong-Wook Kim a), and Young-Ho Seo a)., JPEG Pleno...., JPEG, JPEG2000, AVC/H.264, HEVC/H.265. Abstract This paper introduces the analysis and development of digital holographic data codec technology to effectively compress hologram data. First, the generation method and data characteristics of the hologram standard data set provided by JPEG Pleno are introduced. We analyze energy compaction according to hologram generation method using discrete wavelet transform and discrete cosine transform. The quantization efficiency according to the hologram generation method is analyzed by applying uniform quantization and non-uniform quantization. We propose a transformation method quantization method suitable for hologram generation method through transform and quantization experiments. Finally, holograms are compressed using standard compression codecs such as JPEG, JPEG2000, AVC/H.264 and HEVC/H.265 and the results are analyzed. Keyword : Wavelet, Transform, Quantization, Digital Hologram, Signal process a) (Department of Electronic Material Engineering, Kwangwoon University) b) (ETRI) Corresponding Author : (Young-Ho Seo) yhseo@kw.ac.kr Tel: ORCID: This research was supported by Basic Science Research Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education(NRF-2018R1D1A1B ), This work was supported by Giga KOREA project, [GK19D0100, Development of Telecommunications Terminal with Digital Holographic Table-top Display]. Manuscript received December 31, 2018; Revised January 22, 2019; Accepted January 22, Copyright 2016 Korean Institute of Broadcast and Media Engineers. All rights reserved. This is an Open-Access article distributed under the terms of the Creative Commons BY-NC-ND ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited and not altered.

2 7: (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) Gabor [1], 3. 3, [2]. 3 [3].. (digital hologram signal processing) [4].. [5-8] [9-12]. JPEG Pleno [13].,,.,. [14,15], [16,17] [18,19].,,, [20-22] [23,24], [25,26] [27,28].... JPEG Pleno.... JPEG Pleno.. JPEG Pleno. 2 JPEG Pleno,. 3.,,. DCT(discrete cosine transform) DWT(discrete wavelet transform)., 2. 4,,. 5.. JPEG Pleno. JPEG Pleno. 3, mat,,,,. ERC interfere 3D. 2D, 3D. 2D Multi 3D. 1 ERC interfere. 1.

3 ,. B-COM Institute of Research & Technology (IRT) b-com.. Microsoft Research Interactive Visual Media. 2 B-COM 1080p. (c) (d) (e) 1. ERC interfere 2D Dice, 2D Multi, (c) 3D Multi, (d) 3D Venus, (e) 3D Cat Fig. 1. Reconstruction result using ERC interfere data set 2D Dice, 2D Multi, (c) 3D Multi, (d) 3D Venus, (e) 3D Cat 2D Dices, 2D Multi, 3D Multi, 3D Venus,. 2D Dices,. 2D Multi 3D Multi,,. 3D Cat (c) 2. B-COM 1080p, 1080p, (c) 1080p, (d) 1080p Fig. 2. Reconstruction result using B-COM data set Ballet1080p, Breakdancers1080p, (c) Dices1080p, (d) Piano1080p 1080p. RGB,,, / / /. (d) (c) (d) 3. UBI EmergIMG,, (c), (d) Fig. 3. Reconstruction result using UBI EmergIMG data set Astronaut, Car, (c) Chess, (d) Skull

4 김진겸 외 7인: 디지털 홀로그램의 압축을 위한 특성 분석 (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) 16 7 (c) (d) 그림 4. 실험에 사용된 홀로그램의 히스토그램 2D 멀티의 실수 및 허수, 3D 멀티의 실수 및 허수, (c) 주사위1080p RGB성분의 실수, (d) 주사위1080p RGB 성분의 허수 Fig. 4. Histograms of holograms used in the experiment; real and imaginary of 2D Multi, 3D Multi, (c) real of RGB in Dices1080p, (d) imaginary of RGB in Dices1080p 마지막으로 UBI(Universidade da Beira Interior) EmergIMG의 데이터세트에는 홀로그래피 및 포인트 클라우드 데 IMG 데이터 세트를 이용하여 복원한 결과들이다. 앞에서 설명된 데이터베이스의 데이터 분포를 분석하기 이터가 제공되고 있다. 미만의 복원 거리 및 위하여 2D Multi와 3D Multi, 그리고 Dices1080p에 대한 이하의 픽셀크기로 정의되어있으며, 다른 데이터 세트에 데이터 분포를 그 비하여 파라미터 값이 비교적 작다. 그림 3은 UBI Emerg- 림 4에 나타내었다. 2D Multi와 3D Multi에서 Dices1080p에 비하여 조금 더

5 . Dices1080p 2D Multi 3D Multi, 0. III. 1. DCT DWT,.. DCT (1) DCT DCT Haar. Haar Daubechies 1. Daubechies.,. Haar. Daubechies., Bi-Orthogonal Reverse Bi-Orthogonal. Coiflet 0, 0. (support). Morlet. Maxican Hat, 2. Symlets Daubechies. Meyer [30] DCT Fig Block based DCT energy calculation

6 7: (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) (c) 6. Bi-Orthogonal, Bi-Orthogonal, (c) Reverse Bi-Orthogonal, (d) Reverse Bi-Orthogonal Fig. 6. Filter coefficient Bi-Orthogonal Low pass filter, Bi-Orthogonal high pass filter, (c) Reverse Bi-Orthogonal Low pass filter, (d) Reverse Bi-Orthogonal high pass filter (d) DWT. 10%~90%,., 70., Bi-orthogonal (3,3)., Reverse Bi-orthogonal (3,3) DWT Quad-Tree,. (2) DWT DWT. 7 2 DWT DWT DWT Fig Level DWT energy Calculation A-Law μ-law

7 . (3) () (). Q (dynamic range) [30]. 8-1~ (4) A-Law (compressor).,. A A-Law. A A 1 A-Law [30]. ln ln ln for for (5) A-Law (expander)., [30]. 8. Fig. 8. uniform scalar quantization example..,, ln for ln ln. 9-1~1 3 6 A-Law. 9. A-Law, Fig. 9. A-law non-uniform scalar quantization example compressor expander

8 7: (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) (6) μ-law.,. μ μ-law. μ 255. μ 0 μ-law [31] ln ln 1. Table 1. Parameter according to standard codec (7) μ-law., [3]. Codec Input Compression ratio control JPEG RGB Quality factor JPEG2000 RGB Compression Ratio AVC HEVC YUV YUV Quantization Parameter CU Size 10-1~1 3 6 μ-law. 10. μ-law Fig. 10. μ-law non-uniform scalar quantization example compressor expander 3. JPEG, JPEG2000, AVC Intra HEVC Intra [8]. JPEG JPEG2000, AVC HEVC. AVC HEVC Intra JPEG Quality factor 0~ JPEG2000. AVC HEVC CU Size QP (Quantization Parameter) 0~ bps(bit per sample) PSNR. PSNR 3. 2D Multi 3D Multi JPEG JPEG2000 R AVC HEVC Y. Dices1080p RGB YUV 3. 11

9 .

10 7: (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) 11. 2D Multi, 3D Multi, Dices1080p Dices1080p Fig. 11. Hologram encoding and decoding flowchart 2D Multi, 3D Multi, Dices1080p single channel Dices1080p multi channel IV. 3. DCT DWT. A-Law μ-law Bi-Orthogonal (1, 1). 4~6 RGB. 2D Multi 90% 3.83%, 3D Multi 13.01%. Dices1080p 87.90%.,.. 13 Reverse Bi-Orthogonal (3, 3) 12. Bi-Orthogonal (1, 1) Fig. 12. Bi-Orthogonal (1, 1) filter energy compaction 13. Reverse Bi-Orthogonal (3, 3) Fig. 13. Reverse Bi-Orthogonal (3, 3) filter energy compaction

11 . 2D Multi 3D Multi 90% 20%. Dices1080p % Bi-Orthogonal (1, 1). Reverse Bi-Orthogonal (3, 3). 14 DCT (c) 15. Bi-Orthogonal(1, 1) 2D Multi, 3D Multi, (c) Dices1080p Fig. 15. Bi-Orthogonal (1, 1) results 2D Multi, 3D Multi, (c) Dices 1080p (c) 16. Reverse Bi-Orthogonal(3, 3) 2D Multi, 3D Multi, (c) Dices1080p Fig. 16. Reverse Bi-Orthogonal (3, 3) results 2D Multi, 3D Multi, (c) Dices 1080p 14. DCT Fig. 14. DCT energy compaction 2D Multi 90% 9% 3D Multi 10%. Reverse Bi-Orthogonal (3, 3) Bi-Orthothogonal(1, 1). Dices1080p Bi-Orthogonal(1, 1).Bi-Orthogonal(1,1) DCT, Reverse Bi-Orthogonal(3, 3)., DWT Bi-Orthogonal (1, 1). 2D Multi 3D Multi Dices1080p R. 16 Reverse-Bi- Orthogonal(3, 3), DCT. (c) DCT 2D Multi, 3D Multi, (c) Dices1080p Fig block based DCT results 2D Multi, 3D Multi, (c) Dices 1080p D Multi Dices1080p D Multi 2D Multi.

12 7: (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) (c) (d) (e) (d) 18. Multi 2D 1, 8, (c) 1 A-Law, (d) 8 A-Law, (e) 1 μ-law, (f) 8 μ-law Fig. 18. Reconstructed quantization Multi 2D amplitude image 1bit uniform quantization image, 8bit uniform quantization image, (c) 1bit A-law non-uniform quantization image, (d) 8bit A-law non-uniform quantization image, (e) 1bit μ-law non-uniform quantization image, (f) 8-bit μ-law non-uniform quantization image (c) (d) (e) (f) 19. Dices 1080p 1, 8, (c) 1 A-Law, (d) 8 A-Law, (e) 1 μ-law, (f) 8 μ-law Fig. 19. Reconstructed quantization Dices 1080p amplitude image 1bit uniform quantization image, 8bit uniform quantization image, (c) 1bit A-law non-uniform quantization image, (d) 8bit A-law non-uniform quantization image, (e) 1bit μ-law non-uniform quantization image, (f) 8-bit μ-law non-uniform quantization image PSNR. 20 PSNR , Multi 2D Multi 3D 10dB PSNR. Dices 1080p 10dB 20dB

13 . A-Law μ-law 5dB.. Dice 1080p. PSNR PSNR. A μ... Multi 2D Multi 3D A=86.7, μ=255 PSNR Dices 1080p PSNR... Multi 2D, Multi 3D -0.35~0.35 Dices 1080p -3~3 10. Dice 1080p, ±0.5~ Dices1080p QP 0 PSNR bps QP 50 bps PSNR QP 2D Multi 3D Multi. 21., AVC, HEVC. 22. HEVC AVC Y, JPEG JPEG2000 R. 0~255, HEVC AVC. 20. PSNR Fig. 20. Average PSNR comparison of quantization results of real and imaginary images hologram results reconstructed hologram results

14 7: (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) (c) (d) 21. () 2D Multi, 3D Multi, (c) 4:4:4 Dices1080p (RGB ), (d) Dices1080p(RGB ) Fig. 21. Results for original Hologram and reconstructed Hologram (Floating points) 2D Multi, 3D Multi, (c) 4:4:4 Dices1080p (RGB average), (d) Channel separation Dices1080p (RGB average) (c) (d) 22. () 2D Multi, 3D Multi, (c) Dices1080p(RGB ), (d) Dices1080p(RGB ) Fig. 22. Results for original Hologram and reconstructed Hologram (Normalizaiton) 2D Multi, 3D Multi, (c) 4:4:4 Dices1080p (RGB average), (d) Channel separation Dices1080p (RGB average)

15 (c) (d) 23. () 2D Multi, 3D Multi, (c) Dices1080p(RGB ), (d) Dices1080p(RGB ) Fig. 23. Results for original Hologram and reconstruction decoded Hologram (Normalizaiton) 2D Multi, 3D Multi, (c) 4:4:4 Dices1080p (RGB average), (d) Channel separation Dices1080p (RGB average) 23, Reconstruction Amplitude PSNR SSIM. Reconstruction, PSNR SSIM 21 HEVC. Dices1080p HEVC YUV 4:4:4. 24 HEVC Intra 2D Multi. 24 QP 0 QP 30 (c) QP 50. 2D Multi QP 0 PSNR QP 50 bps 0.03 PSNR HEVC Intra 3D Multi. 25 QP 0 QP 30 (c) QP 50. 3D Multi QP 0 PSNR D Multi. QP 50 bps 0.04 PSNR D Multi. 26 HEVC Intra Dices1080p (c) 24. 2D Multi QP 0, QP 30, (c) QP 50 Fig. 24. Results for compression 2D Multi QP 0, QP 30, (c) QP 50

16 7: (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) (d) (e) (f) 25. 3D Multi QP 0, QP 30, (c) QP 50 Figure 25. Results for compression 3D Multi QP 0, QP 30, (c) QP 50 (g) (h) (i) 26. Dices1080p QP 0, QP 30, (c) QP 50 Fig. 26. Results for compression Dices1080p QP 0, QP 30, (c) QP QP 0 QP 30 (c) QP 50. V... 2D Multi 3D Multi DCT DWT Dices1080p DWT Reverse Bi-Orthogonal (3, 3)..,. HEVC Intra, AVC Intra.. (References) [1] Dennis Gabor, A new microscopic principle, Nature, 161, pp , [2] P. Hariharan, "Basics of Holography", Cambridge University Press, May [3] W. Osten, A. Faridian, P. Gao, K. Körner, D. Naik, G. Pedrini, Al. Kumar Singh, M. Takeda, and M. Wilke, "Recent advances in digital holography [Invited]," Appl. Opt. 53, G44-G63, [4] H. Yoshikawa, "Digital holographic signal processing," Proc. TAO First International Symposium on Three Dimensional Image Communication Technologies, pp. S-4-2, Dec [5] Y.H. Seo, Hyun-Jun Choi, and Dong-Wook Kim, "Lossy Coding Technique for Digital Holographic Signal", SPIE Optical Engineering, Vol. 45, No. 6, pp ~ , Jun [6] Y.H. Seo, H. J. Choi, J. S. Yoo, G. S. Lee, C. H. Kim, S. H. Lee, S. H. Lee, and D. W. Kim, "Digital hologram compression technique by eliminating spatial correlations based on MCTF." Optics Communications, vol. 283, no. 21, pp , Nov [7] F. Dufaux, Y. Xing, Y. B. P. Popescu, and P. Schelkens, "Compression of digital holographic data: an overview." In Applications of Digital Image Processing XXXVIII. International Society for Optics and Photonics. vol. 9599, no I, pp. 1-11, Sep [8] J. P. Peixeiro, C. Brites, J. Ascenso, and F. Pereira, "Holographic data coding: Benchmarking and extending hevc with adapted transforms." IEEE Transactions on Multimedia, vol. 20, no. 2, pp , Feb.2018 [9] Y. H. Seo, H. J. Choi and D. W. Kim, "3D scanning-based compression technique for digital hologram video", Signal Processing: Image

17 Communication, vol. 22, no. 2, pp , Nov [10] E. Darakis and T. J. Naughton, "Compression of digital hologram sequences using MPEG-4", SPIE Proc, vol. 7358, pp , May [11] P. Tsang, K. W. K. Cheung, T. C. Poon, and C. Zhou, "Demonstration of compression ratio of over 4000 times for each digital hologram in a sequence of 25 frames in a holographic video." Journal of Optics, vol. 14, no. 12, pp. 1-7, Dec [12] Y.H. Seo, Y. H. Lee, J. S. Yoo, and D. W. Kim, "Scalable hologram video coding for adaptive transmitting service." Applied optics, vol. 52, no. 1, pp. A254-A268, Jan [13] JPEG Pleno [14] H. Gu Kim and Y. M. Ro, "Ultrafast layer based computer-generated hologram calculation with sparse template holographic fringe pattern for 3-D object," Opt. Express 25, , [15] H. Zhang, L. Cao, and G. Jin, "Computer-generated hologram with occlusion effect using layer-based processing," Appl. Opt. 56, F138-F143, [16] P. Su, W. Cao, J. Ma, Bi. Cheng, X. Liang, L. Cao, and G. Jin, "Fast Computer-Generated Hologram Generation Method for Three-Dimensional Point Cloud Model," J. Display Technol. 12, , [17] P. W. M. Tsang, T.-C. Poon, and Y. M. Wu, "Review of fast methods for point-based computer-generated holography [Invited]," Photon. Res. 6, , [18] H. Yeom, Y. Ji, S. Kim, S. Ko, S. Kim, H. Zhang, B. Li, K. Shin, M. Askari, and J. Park, "Hologram synthesis with correct reflectance distribution in fully analytic mesh-based method," in Imaging and Applied Optics 2016, OSA Technical Digest (online), Optical Society of America, [19] Y. Ju and J. Park, "Fast Generation of Mesh Based CGH in Head-Mounted Displays using Foveated Rendering Technique," in Imaging and Applied Optics 2018 (3D, AO, AIO, COSI, DH, IS, LACSEA, LS&C, MATH, pcaop), OSA Technical Digest, Optical Society of America, [20] Logan A. Williams, Georges Nehmetallah, Rola Aylo, and Partha P. Banerjee, "Application of up-sampling and resolution scaling to Fresnel reconstruction of digital holograms," Appl. Opt. 54, [21] S. Lee, H. Chang, H. Wey, and D. Nam, "Sampling and error analysis of radial symmetric interpolation for fast hologram generation," Appl. Opt. 55, A104-A110, [22] Chor Shen Tay, Ken Tanizawa, and Akira Hirose, "High-quality frame interpolation in computer generated holographic movies using coherent neural networks with a hybrid learning method," Appl. Opt. 47, , [23] P. W. M. Tsang, Y. T. Chow, and T.-C. Poon, "Enhancement on the generation of sampled phase-only holograms," Chin. Opt. Lett. 13, , [24] Yuan Hong, Tielin Shi, Yichun Zhang, and Guanglan Liao, "Fringe contrast enhancement of digital off-axis hologram via sparse representation," Chin. Opt. Lett. 14, , [25] Lingfeng Yu, Yingfei An, and Lilong Cai, "Numerical reconstruction of digital holograms with variable viewing angles," Opt. Express 10, , [26] Xin Li, Juan Liu, Tao Zhao, and Yongtian Wang, "Color dynamic holographic display with wide viewing angle by improved complex amplitude modulation," Opt. Express 26, , [27] P. Memmolo, V. Bianco, M. Paturzo, B. Javidi, P. A. Netti, and P. Ferraro, "Encoding multiple holograms for speckle-noise reduction in optical display," Opt. Express 22, , [28] D. Hincapie, J. H.-Ramírez, and J. G.-Sucerquia, "Single-shot speckle reduction in numerical reconstruction of digitally recorded holograms," Opt. Lett. 40, , [29] M.-Misiti, Y.-Misiti, G.- Oppenheim and J-M.- Poggi, Wavelet Toolobox Reference, Sep Online Only [30] Wavelet browser, [31] Quantization (signal processing), _(signal_processing) [32] A-law algorithm, [33] μ-law algorithm, ~ : - ORCID : - :, 2D/3D

18 7: (Jin-Kyum Kim et al.: Characteristic Analysis for Compression of Digital Hologram) ~ - ORCID : - :,, 2D/3D : () ~ : () - ORCID : - :,, : () : () : () - ORCID : - :, SoC : : : ~ : ~ : - : - :, 2D/3D, : : Texas A&M ~ : Columbia ~ : / - :, 3DTV, UHDTV, Audio Video,,

19 : () : : Georgia () ~ : - ORCID : - : : 3D,, VLSI Testability, VLSI CAD, DSP, Wireless Communication : () : () : () ~ : ~ : - ORCID : - :, 2D/3D,