-
摘要:
高光谱图像中存储了丰富的光谱信息，具有极大的应用价值，但现有大部分高光谱图像压缩方法难以同时兼顾图像中的空间冗余与谱间冗余，导致压缩性能受到局限。针对该问题，提出了一种基于三维修正偏置的子空间(Saab)变换的高光谱图像压缩方法。采用三维Saab变换对高光谱图像的分块进行空间光谱信息融合的降维操作，同时去除谱间冗余和局部空间冗余；利用高效率视频编码(HEVC)中的帧内编码模块进一步去除空间冗余和统计冗余；实现低失真、高比率的高光谱图像压缩。在多个高光谱图像数据集上的实验结果表明，所提方法在同码率下重建图像的信噪比(SNR)比采用主成分分析(PCA)降维的方法至少提高0.62 dB，在高码率的情况下性能优于张量分解的压缩方法。同时，验证了不同降维方法对分类任务的性能影响，结果表明，所提方法更好地保留了图像中的重要特征，在低码率的情况下仍可以保持较高的分类精度。
-
关键词:
- 修正偏置的子空间(Saab)变换 /
- 空间光谱信息融合 /
- 高效率视频编码(HEVC) /
- 高光谱图像 /
- 图像压缩
Abstract:Hyperspectral images contain rich and valuable spectral information, which brings great challenges to storage and transmission. However, most current hyperspectral image compression methods cannot consider spatial and spectral redundancy simultaneously, resulting in limited compression performance. We present a hyperspectral image compression method based on 3D subspace approximation with adjusted bias (Saab) transform. 3D Saab transform is firstly applied to hyperspectral image blocks, which performs spatial-spectral fusion and dimensionality reduction on blocks to remove spectral redundancy and local spatial redundancy simultaneously. Then, we use intra mode of high efficiency video coding (HEVC) to further remove spatial and statistical redundancy. Experimental results demonstrate that the proposed method can improve the signal-to-noise ratio (SNR) by at least 0.62 dB as compared with principle component analysis (PCA) based algorithm. At a high bit rate, the proposed method outperforms the state-of-art tensor decomposition compression method. We also evaluate the impact of different dimensionality reduction methods on classification, which demonstrates that the proposed method can better retain important features, with improved classification accuracy at a low bit rate.
-
图 1 基于三维Saab变换的高光谱图像压缩方法流程
Figure 1. Flowchart of hyperspectral image compression method based on 3D Saab transform
图 6 不同方法在Salinas数据集上的压缩性能
Figure 6. Compression performance of different methods on Salinas dataset
图 7 本文方法和PCA在不同数据集上的压缩性能
Figure 7. Compression performance of the proposed method and PCA on different datasets
-
[1] 马晨光, 曹汛, 季向阳, 等. 高分辨率光谱视频采集研究[J]. 电子学报, 2015, 43(4): 783-790. doi: 10.3969/j.issn.0372-2112.2015.04.022MA C G, CAO X, JI X Y, et al. Research on high resolution hyperspectral capture technique[J]. Acta Electronica Sinica, 2015, 43(4): 783-790(in Chinese). doi: 10.3969/j.issn.0372-2112.2015.04.022 [2] LEITNER R, BIASIO M D, ARNOLD T, et al. Multi-spectral video endoscopy system for the detection of cancerous tissue[J]. Pattern Recognition Letters, 2013, 34(1): 85-93. doi: 10.1016/j.patrec.2012.07.020 [3] CHO W, JANG J, KOSCHAN A, et al. Hyperspectral face recognition using improved inter-channel alignment based on qualitative prediction models[J]. Optics Express, 2016, 24(24): 27637-27662. doi: 10.1364/OE.24.027637 [4] SANTARA A, MANI K, HATWAR P, et al. BASS Net: Band-adaptive spectral-spatial feature learning neural network for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(9): 5293-5301. doi: 10.1109/TGRS.2017.2705073 [5] LANDGREBE D. Hyperspectral image data analysis[J]. IEEE Signal Processing Magazine, 2002, 19(1): 17-28. doi: 10.1109/79.974718 [6] FANG L Y, HE N J, LIN H. CP tensor-based compression of hyperspectral images[J]. Journal of the Opitical Society of America A: Optics, Image Science, and Vision, 2017, 34(2): 252-258. doi: 10.1364/JOSAA.34.000252 [7] 王成. 高光谱图像压缩的方法研究[D]. 南京: 南京理工大学, 2014: 5-6.WANG C. Researches of hyperspectral image compression methods[D]. Nanjing: Nanjing University of Science and Technology, 2014: 5-6(in Chinese). [8] ZHANG J, LIU G Z. A novel lossless compression for hyperspectral image by context-based adaptive classified arithmetic coding in wavelet domain[J]. IEEE Geoscience and Remote Sensing Letters, 2007, 4(3): 461-465. doi: 10.1109/LGRS.2007.897924 [9] DU Q, FOWLER J E. Hyperspectral image compression using JPEG2000 and principal component analysis[J]. IEEE Geoscience and Remote Sensing Letters, 2007, 4(2): 201-205. doi: 10.1109/LGRS.2006.888109 [10] WANG X H, TAO J Z, SHEN Y T, et al. Distributed source coding of hyperspectral images based on three-dimensional wavelet[J]. Journal of the Indian Society of Remote Sensing, 2018, 46(4): 667-673. doi: 10.1007/s12524-017-0735-1 [11] SHINDE T S, TIWARI A K, LIN W Y. Low-complexity adaptive switched prediction-based lossless compression of time-lapse hyperspectral image data[C]//2019 IEEE Global Conference on Signal and Information Processing (GlobalSIP). Piscataway: IEEE Press, 2019: 1-5. [12] CANG S, WANG A. Research on hyperspectral image reconstruction based on GISMT compressed sensing and interspectral prediction[J]. International Journal of Optics, 2020, 2020(12): 1-11. [13] RYAN M J, ARNOLD J F. The lossless compression of AVIRIS images by vector quantization[J]. IEEE Transations on Geoscience and Remote Sensing, 1997, 35(3): 546-550. doi: 10.1109/36.581964 [14] RYAN M J, PICKERING M R. An improved M-NVQ algorithm for the compression of hyperspectral data[C]//IEEE 2000 International Geoscience and Remote Sensing Symposium. Piscataway: IEEE Press, 2000, 2: 600-602. [15] MOTTA G, RIZZO F, STORER J A. Partitioned vector quantization: Application to lossless compression of hyperspectral images[C]//2003 International Conference on Multimedia and Expo. Piscataway: IEEE Press, 2003: 111-553. [16] 宋娟. 基于分布式信源编码的多光谱图像/视频压缩技术研究[D]. 西安: 西安电子科技大学, 2012.SONG J. Researches on comression of multispectral images/video based on distributed source coding[D]. Xi'an: Xidian University, 2012(in Chinese). [17] LI R, PAN Z B, WANG Y, et al. The correlation-based tucker decomposition for hyperspectral image compression[J]. Neurocomputing, 2021, 419: 357-370. doi: 10.1016/j.neucom.2020.08.073 [18] ZHANG L F, ZHANG L P, TAO D C, et al. Compression of hyperspectral remote sensing images by tensor approach[J]. Neurocomputing, 2015, 147: 358-363. doi: 10.1016/j.neucom.2014.06.052 [19] FU W, LI S T, FANG L Y, et al. Adaptive spectral-spatial compression of hyperspectral image with sparse representation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(2): 671-682. [20] DUA Y M, SINGH R S, PARWANI K, et al. Convolution neural network based lossy compression of hyperspectral images[J]. Signal Processing: Image Communication, 2021, 95: 116-255. [21] DUA Y M, KUMAR V, SINGH R S. Comprehensive review of hyperspectral image compression algorithms[J]. Optical Engineering, 2020, 59(9): 090902. [22] DU Q, LY N, FOWLER J E. An operational approach to PCA+JPEG2000 compression of hyperspectral imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2013, 7(6): 2237-2245. [23] KUO C C J, ZHANG M, LI S Y, et al. Interpretable convolutional neural networks via feedforward design[J]. Journal of Visual Communication and Image Representation, 2019, 60: 346-359. [24] LI N, ZHANG Y F, ZHANG Y, et al. On energy compaction of 2D Saab image transforms[C]//2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC). Piscataway: IEEE Press, 2019: 466-475. [25] LI N, ZHANG Y, KUO C C J. Explainable machine learning based transform coding for high efficiency intra prediction[EB/OL]. (2020-11-21)[2021-09-01]. https://arxiv.org/abs/2012.11152. [26] SZEV S, BUDAGAVI M, SULLIVAN G J. High efficiency video coding (HEVC): Algorithms and architectures[M]. Berlin: Springer, 2014: 91-112. [27] LAINEMA J, BOSSEN F, HAN W J, et al. Intra coding of the HEVC standard[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2012, 22(12): 1792-1801. [28] GRANA M, VEGANZONS M A, AYERDI B. Hyperspectral remote sensing scenes[EB/OL]. [2021-09-01]. http://www.ehu.eus/ccwintco/index.php/Hyperspectral_Remote_Sensing_Scenes. [29] CHEN Y S, JIANG H L, LI C Y, et al. Deep feature extraction and classification of hyperspectral images based on convolutional neural networks[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(10): 6232-6251. -
下载:
点击查看大图
计量
- 文章访问数: 305
- HTML全文浏览量: 135
- PDF下载量: 34
- 被引次数: 0
