《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (4): 1191-1197.DOI: 10.11772/j.issn.1001-9081.2021050746
所属专题: 网络空间安全
收稿日期:2021-05-13
									
				
											修回日期:2023-02-21
									
				
											接受日期:2023-02-24
									
				
											发布日期:2021-09-24
									
				
											出版日期:2023-04-10
									
				
			通讯作者:
					李智
							作者简介:高健(1993—),男,山西大同人,硕士,主要研究方向:图形可视化、信息隐藏、医学影像分析;基金资助:
        
                                                                                                                            Jian GAO1, Zhi LI1( ), Bin FAN1, Chuanxian JIANG2
), Bin FAN1, Chuanxian JIANG2
			  
			
			
			
                
        
    
Received:2021-05-13
									
				
											Revised:2023-02-21
									
				
											Accepted:2023-02-24
									
				
											Online:2021-09-24
									
				
											Published:2023-04-10
									
			Contact:
					Zhi LI   
							About author:GAO Jian, born in 1993, M. S. His research interests include graph visualization, information hiding, medical image analysis.Supported by:摘要:
针对三维医学影像的版权保护问题,以及随着待保护图像数量增加而导致的水印存储容量同步增加的问题,提出了一种基于光线投射采样和极复指数矩的三维医学影像鲁棒性零水印算法。首先,提出了一种基于光线投射的采样算法,对多张二维医学图像序列构成的三维医学影像进行特征采样并在二维图像空间中进行描述;其次,提出了一种针对三维医学影像的鲁棒性零水印算法,所提算法通过光线投射采样获取三维医学影像的冠状面、矢状面和横切面的3个二维特征图像,将3个二维特征图像进行极复指数变换获得四元数正交矩;最后,使用四元数正交矩和Logistic混沌加密构建零水印信息。仿真实验结果表明,在各种常见的图像处理攻击和几何攻击下所提算法能够保持零水印提取的比特正确率在0.920 0以上,水印存储容量随着三维医学影像的数据增多而提升,与其他优秀的二维医学图像零水印算法相比,所提算法的存储容量至少提升了93.75%。
中图分类号:
高健, 李智, 樊缤, 姜传贤. 基于光线投射采样和四元数正交矩的高效三维医学影像鲁棒零水印算法[J]. 计算机应用, 2023, 43(4): 1191-1197.
Jian GAO, Zhi LI, Bin FAN, Chuanxian JIANG. Efficient robust zero-watermarking algorithm for 3D medical images based on ray-casting sampling and quaternion orthogonal moment[J]. Journal of Computer Applications, 2023, 43(4): 1191-1197.
| 攻击类别 | 参数值 | BCR | PSNR/dB | 
|---|---|---|---|
| 左上裁剪 | 1/8 | 1.000 0 | 17.43 | 
| 1/4 | 1.000 0 | 11.89 | |
| 1/2 | 0.939 5 | 6.72 | 
表1 裁剪攻击实验结果
Tab. 1 Experiment results of cropping attacks
| 攻击类别 | 参数值 | BCR | PSNR/dB | 
|---|---|---|---|
| 左上裁剪 | 1/8 | 1.000 0 | 17.43 | 
| 1/4 | 1.000 0 | 11.89 | |
| 1/2 | 0.939 5 | 6.72 | 
| 攻击类别 | 缩放比例/% | BCR | PSNR/dB | 
|---|---|---|---|
| 缩小 | 25 | 0.924 4 | 25.75 | 
| 50 | 1.000 0 | 30.50 | |
| 75 | 1.000 0 | 34.44 | |
| 放大 | 125 | 1.000 0 | 38.35 | 
| 150 | 1.000 0 | 41.16 | |
| 200 | 1.000 0 | 40.10 | 
表2 缩放攻击实验结果
Tab. 2 Experiment results of scaling attacks
| 攻击类别 | 缩放比例/% | BCR | PSNR/dB | 
|---|---|---|---|
| 缩小 | 25 | 0.924 4 | 25.75 | 
| 50 | 1.000 0 | 30.50 | |
| 75 | 1.000 0 | 34.44 | |
| 放大 | 125 | 1.000 0 | 38.35 | 
| 150 | 1.000 0 | 41.16 | |
| 200 | 1.000 0 | 40.10 | 
| 攻击类别 | 参数值 | BCR | PSNR/dB | 
|---|---|---|---|
| 中值滤波 | 3×3 | 1.000 0 | 21.87 | 
| 5×5 | 0.988 3 | 16.47 | |
| 7×7 | 0.984 4 | 15.28 | |
| 高斯低通滤波 | 1.0 | 1.000 0 | 20.70 | 
| 1.5 | 1.000 0 | 15.77 | |
| 2.0 | 0.986 3 | 15.00 | 
表3 滤波攻击实验结果
Tab. 3 Experiment results of filtering attacks
| 攻击类别 | 参数值 | BCR | PSNR/dB | 
|---|---|---|---|
| 中值滤波 | 3×3 | 1.000 0 | 21.87 | 
| 5×5 | 0.988 3 | 16.47 | |
| 7×7 | 0.984 4 | 15.28 | |
| 高斯低通滤波 | 1.0 | 1.000 0 | 20.70 | 
| 1.5 | 1.000 0 | 15.77 | |
| 2.0 | 0.986 3 | 15.00 | 
| 攻击类别 | 旋转角度/(°) | BCR | PSNR/dB | 
|---|---|---|---|
| 逆时针旋转 | 5 | 1.000 0 | 23.98 | 
| 15 | 1.000 0 | 22.23 | |
| 30 | 0.972 7 | 21.17 | |
| 45 | 0.963 9 | 20.69 | 
表4 旋转攻击实验结果
Tab. 4 Experiment results of rotating attacks
| 攻击类别 | 旋转角度/(°) | BCR | PSNR/dB | 
|---|---|---|---|
| 逆时针旋转 | 5 | 1.000 0 | 23.98 | 
| 15 | 1.000 0 | 22.23 | |
| 30 | 0.972 7 | 21.17 | |
| 45 | 0.963 9 | 20.69 | 
| 攻击类别 | 方差 | BCR | PSNR/dB | 
|---|---|---|---|
| 高斯噪声 | 0.01 | 1.000 0 | 22.14 | 
| 0.02 | 1.000 0 | 20.17 | |
| 0.03 | 0.975 6 | 18.77 | |
| 椒盐噪声 | 0.01 | 1.000 0 | 23.84 | 
| 0.02 | 0.987 3 | 21.91 | |
| 0.03 | 0.987 3 | 20.75 | 
表5 噪声攻击实验结果
Tab. 5 Experiment results of noise attacks
| 攻击类别 | 方差 | BCR | PSNR/dB | 
|---|---|---|---|
| 高斯噪声 | 0.01 | 1.000 0 | 22.14 | 
| 0.02 | 1.000 0 | 20.17 | |
| 0.03 | 0.975 6 | 18.77 | |
| 椒盐噪声 | 0.01 | 1.000 0 | 23.84 | 
| 0.02 | 0.987 3 | 21.91 | |
| 0.03 | 0.987 3 | 20.75 | 
| 攻击类别 | 质量因子 | BCR | PSNR/dB | 
|---|---|---|---|
| JPEG压缩 | 10 | 0.937 5 | 29.42 | 
| 30 | 0.984 4 | 33.67 | |
| 70 | 0.988 3 | 35.48 | |
| 90 | 1.000 0 | 42.19 | 
表6 JPEG压缩实验结果
Tab. 6 Experiment results of JPEG compression
| 攻击类别 | 质量因子 | BCR | PSNR/dB | 
|---|---|---|---|
| JPEG压缩 | 10 | 0.937 5 | 29.42 | 
| 30 | 0.984 4 | 33.67 | |
| 70 | 0.988 3 | 35.48 | |
| 90 | 1.000 0 | 42.19 | 
| 攻击1 | 攻击2 | 攻击3 | BCR | 
|---|---|---|---|
| JPEG压缩(30) | 中值滤波(3×3) | 旋转(30°) | 0.953 2 | 
| 左上裁剪(1/4) | 中值滤波(3×3) | 缩放(50%) | 0.997 9 | 
| 左上裁剪(1/8) | 缩放(75%) | 旋转(30°) | 0.958 3 | 
| 椒盐噪声(0.01) | 高斯噪声(0.01) | — | 1.000 0 | 
| 旋转(30°) | 高斯噪声(0.03) | — | 0.988 3 | 
| 旋转(15°) | 高斯低通(1.0) | 缩放(150%) | 1.000 0 | 
表7 多重攻击实验结果
Tab. 7 Experiment results of multiple attacks
| 攻击1 | 攻击2 | 攻击3 | BCR | 
|---|---|---|---|
| JPEG压缩(30) | 中值滤波(3×3) | 旋转(30°) | 0.953 2 | 
| 左上裁剪(1/4) | 中值滤波(3×3) | 缩放(50%) | 0.997 9 | 
| 左上裁剪(1/8) | 缩放(75%) | 旋转(30°) | 0.958 3 | 
| 椒盐噪声(0.01) | 高斯噪声(0.01) | — | 1.000 0 | 
| 旋转(30°) | 高斯噪声(0.03) | — | 0.988 3 | 
| 旋转(15°) | 高斯低通(1.0) | 缩放(150%) | 1.000 0 | 
| 算法 | 不同攻击(参数值)下的BCR | |||||||
|---|---|---|---|---|---|---|---|---|
| 高斯噪声(0.01) | 椒盐噪声(0.03) | JPEG压缩(30) | 中值滤波(3×3) | 缩放(50%) | 旋转(45°) | 左上裁剪(1/8) | 左上裁剪(1/2) | |
| 本文算法 | 1.000 0 | 0.987 3 | 0.984 4 | 1.000 0 | 1.000 0 | 0.963 9 | 1.000 0 | 0.939 5 | 
| 文献[ | 0.975 6 | 0.969 7 | 0.997 1 | 0.985 2 | 0.994 1 | 0.906 3 | 1.000 0 | 0.631 8 | 
| 文献[ | 0.971 5 | 0.930 2 | 0.991 6 | 0.982 2 | 0.956 2 | 0.684 5 | 0.774 4 | — | 
| 文献[ | 0.990 1 | 0.941 2 | 0.996 1 | 0.983 6 | 0.992 0 | — | — | — | 
| 文献[ | 0.999 7 | 0.999 7 | 0.937 1 | 0.996 1 | — | — | — | — | 
表8 本文算法与其他算法的单项攻击鲁棒性对比
Tab. 8 Robustness comparison against single attack between proposed algorithm and other algorithms
| 算法 | 不同攻击(参数值)下的BCR | |||||||
|---|---|---|---|---|---|---|---|---|
| 高斯噪声(0.01) | 椒盐噪声(0.03) | JPEG压缩(30) | 中值滤波(3×3) | 缩放(50%) | 旋转(45°) | 左上裁剪(1/8) | 左上裁剪(1/2) | |
| 本文算法 | 1.000 0 | 0.987 3 | 0.984 4 | 1.000 0 | 1.000 0 | 0.963 9 | 1.000 0 | 0.939 5 | 
| 文献[ | 0.975 6 | 0.969 7 | 0.997 1 | 0.985 2 | 0.994 1 | 0.906 3 | 1.000 0 | 0.631 8 | 
| 文献[ | 0.971 5 | 0.930 2 | 0.991 6 | 0.982 2 | 0.956 2 | 0.684 5 | 0.774 4 | — | 
| 文献[ | 0.990 1 | 0.941 2 | 0.996 1 | 0.983 6 | 0.992 0 | — | — | — | 
| 文献[ | 0.999 7 | 0.999 7 | 0.937 1 | 0.996 1 | — | — | — | — | 
| 攻击1 | 攻击2 | BCR | ||
|---|---|---|---|---|
| 本文算法 | 文献[ | 文献[ | ||
| 椒盐噪声(0.03) | 缩放(125%) | 0.987 3 | 0.960 9 | — | 
| JPEG(30) | 高斯噪声(0.03) | 0.970 5 | 0.963 9 | — | 
| 旋转(2°) | JPEG(10) | 0.937 5 | — | 0.909 6 | 
| 中值滤波(5×5) | 高斯噪声(0.03) | 0.970 4 | — | 0.905 6 | 
表9 本文算法与其他算法的多重攻击鲁棒性对比
Tab. 9 Robustness comparison against multiple attacks between proposed algorithm and other algorithms
| 攻击1 | 攻击2 | BCR | ||
|---|---|---|---|---|
| 本文算法 | 文献[ | 文献[ | ||
| 椒盐噪声(0.03) | 缩放(125%) | 0.987 3 | 0.960 9 | — | 
| JPEG(30) | 高斯噪声(0.03) | 0.970 5 | 0.963 9 | — | 
| 旋转(2°) | JPEG(10) | 0.937 5 | — | 0.909 6 | 
| 中值滤波(5×5) | 高斯噪声(0.03) | 0.970 4 | — | 0.905 6 | 
| 数据集 | 三维图像尺寸 | 存储容量/bit | ||
|---|---|---|---|---|
| 本文算法 | 文献[ | 文献[ | ||
| ABIDE | 2 048 | 348 160 | 116 054 | |
| ADNI | 2 048 | 98 304 | 32 768 | |
| ICBM | 2 048 | 352 256 | 117 419 | |
表10 零水印存储容量
Tab. 10 Zero-watermarking storage capacity
| 数据集 | 三维图像尺寸 | 存储容量/bit | ||
|---|---|---|---|---|
| 本文算法 | 文献[ | 文献[ | ||
| ABIDE | 2 048 | 348 160 | 116 054 | |
| ADNI | 2 048 | 98 304 | 32 768 | |
| ICBM | 2 048 | 352 256 | 117 419 | |
| 1 | GIAKOUMAKI A, PAVLOPOULOS S, KOUTOURIS D. A medical image watermarking scheme based on wavelet transform[C]// Proceedings of the 25th Annual International Conference of the IEEE on Engineering in Medicine and Biology Society. Piscataway: IEEE, 2003: 856-859. | 
| 2 | 邓小鸿,陈志刚,毛伊敏.基于无损水印的医学图像篡改检测和高质量恢复[J].中国图象图形学报,2014,19(4):583-591. 10.11834/jig.20140413 | 
| DENG X H, CHEN Z G, MAO Y M. Lossless watermarking algorithm for medical image’s tamper detection and recovery with high quality[J]. Journal of Image and Graphics, 2014, 19(4):583-591. 10.11834/jig.20140413 | |
| 3 | 陈怡,李智,张健,等.弥散加权图像的鲁棒水印算法研究[J].中国图象图形学报,2019,24(9):1434-1449. 10.11834/jig.180672 | 
| CHEN Y, LI Z. ZHANG J,et al. Robust watermarking algorithm for diffusion weighted images[J]. Journal of Image and Graphics, 2019, 24(9):1434-1449. 10.11834/jig.180672 | |
| 4 | HUA Z Y, YI S, ZHOU Y C. Medical image encryption using high-speed scrambling and pixel adaptive diffusion[J]. Signal Processing, 2019, 144: 134-144. 10.1016/j.sigpro.2017.10.004 | 
| 5 | HUA Z Y, ZHOU Y C, HUANG H J. Cosine-transform-based chaotic system for image encryption[J]. Information Sciences, 2019, 480: 403-419. 10.1016/j.ins.2018.12.048 | 
| 6 | MOUSAVI S M, NAGHSH A, MANAF A A, et al. A robust medical image watermarking against salt and pepper noise for brain MRI images[J]. Multimedia Tools and Applications, 2017, 76(7): 10313-10342. 10.1007/s11042-016-3622-9 | 
| 7 | MA B, SHI Y Q. A reversible data hiding scheme based on code division multiplexing[J]. IEEE Transactions on Information Forensics and Security, 2016, 11(9): 1914-1927. 10.1109/tifs.2016.2566261 | 
| 8 | 高海波,邓小鸿,陈志刚.基于可逆可见水印的医学图像隐私保护算法[J].计算机应用,2014,34(1):119-123,157. 10.11772/j.issn.1001-9081.2014.01.0119 | 
| GAO H B, DENG X H, CHEN Z G. Medical image privacy protection scheme based on reversible visible watermarking[J]. Journal of Computer Applications, 2014, 34(1):119-123, 157. 10.11772/j.issn.1001-9081.2014.01.0119 | |
| 9 | WEN Q, SUN T F, WANG S X. Based zero-watermark digital watermarking technology[C]// Proceedings of the 2001 in the Third China Information Hiding and Multimedia Security Workshop . Xi’an: Xidian University Press, 2001:102-109. | 
| 10 | 温泉,孙锬锋,王树勋. 零水印的概念与应用[J]. 电子学报, 2003, 31(2):214-216. 10.3321/j.issn:0372-2112.2003.02.015 | 
| WEN Q, SUN T F, WANG S X. Concept and application of zero-watermark[J]. Acta Electronica Sinica, 2003, 31(2):214-216. 10.3321/j.issn:0372-2112.2003.02.015 | |
| 11 | CHEN T H, HORNG G, LEE W B. A publicly verifiable copyright-proving scheme resistant to malicious attacks[J]. IEEE Transactions on Industrial Electronics, 2005, 52(1): 327-334. 10.1109/tie.2004.841083 | 
| 12 | CHANG C C, LIN P Y. Adaptive watermark mechanism for rightful ownership protection[J]. Journal of Systems and Software, 2008, 81(7): 1118-1129. 10.1016/j.jss.2007.07.036 | 
| 13 | 宋伟,侯建军,李赵红,等. 一种基于Logistic混沌系统和奇异值分解的零水印算法[J]. 物理学报, 2009, 58(7):4449-4456. 10.7498/aps.58.4449 | 
| SONG W, HOU J J, LI Z H, et al. A novel zero-bit watermarking algorithm based on Logistic chaotic system and singular value decomposition[J]. Acta Physica Sinica, 2009, 58(7):4449-4456. 10.7498/aps.58.4449 | |
| 14 | TSAI H H, TSENG H C, LAI Y S. Robust lossless image watermarking based on α-trimmed mean algorithm and support vector machine[J]. Journal of Systems and Software, 2010, 83(6): 1015-1028. 10.1016/j.jss.2009.12.026 | 
| 15 | TSAI H H, LAI Y S, LO S C. A zero-watermark scheme with geometrical invariants using SVM and PSO against geometrical attacks for image protection[J]. Journal of Systems and Software, 2013, 86(2): 335-348. 10.1016/j.jss.2012.08.040 | 
| 16 | GAO G Y, JIANG G P. Bessel-Fourier moment-based robust image zero-watermarking[J]. Multimedia Tools and Applications, 2015, 74(3): 841-858. 10.1007/s11042-013-1701-8 | 
| 17 | SUN L, XU J C, ZHANG X X, et al. A novel generalized Arnold transform-based zero-watermarking scheme[J]. Applied Mathematics and Information Sciences, 2015, 9(4): 2023-2035. | 
| 18 | SHAO Z H, SHANG Y Y, ZHANG Y, et al. Robust watermarking using orthogonal Fourier-Mellin moments and chaotic map for double images[J]. Signal Processing, 2016, 120: 522-531. 10.1016/j.sigpro.2015.10.005 | 
| 19 | THANH T M, TANAKA K. An image zero-watermarking algorithm based on the encryption of visual map feature with watermark information[J]. Multimedia Tools and Applications, 2017, 76(11): 13455-13471. 10.1007/s11042-016-3750-2 | 
| 20 | LIU Y Z, YANG F, GAO K, et al. A zero-watermarking scheme with embedding timestamp in vector maps for big data computing[J]. Cluster Computing, 2017, 20(4): 3667-3675. 10.1007/s10586-017-1251-3 | 
| 21 | XIA Z Q, WANG X Y, LI X X, et.al. Efficient copyright protection for three CT images based on quaternion polar harmonic Fourier moments[J]. Signal Processing, 2019, 164: 368-379. 10.1016/j.sigpro.2019.06.025 | 
| 22 | XIA Z Q, WANG X Y, WANG C P, et al. Local quaternion polar harmonic Fourier moments-based multiple zero-watermarking scheme for color medical images[J]. Knowledge-Based Systems, 2021, 216: No.106568. 10.1016/j.knosys.2020.106568 | 
| 23 | WANG C P, WANG X Y, XIA Z Q, et al. Ternary radial harmonic Fourier moments based robust stereo image zero-watermarking algorithm[J]. Information Sciences, 2019, 470: 109-120. 10.1016/j.ins.2018.08.028 | 
| 24 | KANG X B, LIN G F, CHEN Y J, et al. Robust and secure zero-watermarking algorithm for color images based on majority voting pattern and hyper-chaotic encryption[J]. Multimedia Tools and Applications, 2020, 79(1/2): 1169-1202. 10.1007/s11042-019-08191-y | 
| 25 | LEVOY M. Display of surfaces from volume data[J]. IEEE Computer Graphics and Applications, 1988, 8(3): 29-37. 10.1109/38.511 | 
| 26 | 李智,高健,张丽,等. 一种基于ORC采样和QGPCE变换的弥散张量图像鲁棒零水印方法: 202110333207.0[P]. 2021-07-09. | 
| LI Z, GAO J, ZHANG L, et al. A robust zero-watermarking method for diffusion tensor image based on ORC sampling and QGPCE transform: 202110333207.0[P]. 2021-07-09. | |
| 27 | YAP P H, JIANG X D, CHICHUNG KOT A. Two-dimensional polar harmonic transforms for invariant image representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(7): 1259-1270. 10.1109/tpami.2009.119 | 
| 28 | 赵雪峰,殷国富. 基于复合混沌系统的数字图像加密方法研究[J]. 计算机应用, 2006, 26(4):827-829. | 
| ZHAO X F, YIN G F. Research on digital image encryption method based on hybrid chaotic system[J]. Computer Applications, 2006, 26(4):827-829. | |
| 29 | 孙秀花,戴跃伟,王执铨. 混沌序列产生方法及其在图像加密中的应用[J]. 南京师范大学学报(工程技术版), 2004, 4(1):56-59. 10.3969/j.issn.1672-1292.2004.01.016 | 
| SUN X H, DAI Y W, WANG Z Q. The generation of chaotic sequence and its application to image encryption[J]. Journal of Nanjing Normal University (Engineering and Technology), 2004, 4(1):56-59. 10.3969/j.issn.1672-1292.2004.01.016 | |
| 30 | ESSEN VAN D C, UGURBIL K, AUERBACH E, et al. The human connectome project: a data acquisition perspective[J]. NeuroImage, 2012, 62(4): 2222-2231. 10.1016/j.neuroimage.2012.02.018 | 
| 31 | WANG C P, WANG X Y, XIA Z Q, et al. Geometrically resilient color image zero-watermarking algorithm based on quaternion Exponent moments[J]. Journal of Visual Communication and Image Representation, 2016, 41: 247-259. 10.1016/j.jvcir.2016.10.004 | 
| [1] | 陈学斌, 任志强, 张宏扬. 联邦学习中的安全威胁与防御措施综述[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1663-1672. | 
| [2] | 董炜娜, 刘佳, 潘晓中, 陈立峰, 孙文权. 基于编码-解码网络的大容量鲁棒图像隐写方案[J]. 《计算机应用》唯一官方网站, 2024, 44(3): 772-779. | 
| [3] | 黄杰, 武瑞梓, 李均利. 高效的自适应复杂网络鲁棒性优化算法[J]. 《计算机应用》唯一官方网站, 2024, 44(11): 3530-3539. | 
| [4] | 汪韩, 万源, 王东, 丁义明. 宽度学习系统中鲁棒性权值矩阵组合的筛选方法[J]. 《计算机应用》唯一官方网站, 2024, 44(10): 3032-3038. | 
| [5] | 赵徐炎, 崔允贺, 蒋朝惠, 钱清, 申国伟, 郭春, 李显超. CHAIN:基于重合支配的边缘计算节点放置算法[J]. 《计算机应用》唯一官方网站, 2023, 43(9): 2812-2818. | 
| [6] | 葛孟婷, 万鸣华. 基于近邻监督局部不变鲁棒主成分分析的特征提取模型[J]. 《计算机应用》唯一官方网站, 2023, 43(4): 1013-1020. | 
| [7] | 范贤博俊, 陈立家, 李珅, 王晨露, 王敏, 王赞, 刘名果. 鲁棒的视觉机械臂联合建模优化方法[J]. 《计算机应用》唯一官方网站, 2023, 43(3): 962-971. | 
| [8] | 孙梦迪, 孙忠贵, 孔旭, 韩红燕. 针对多模态图像的自适应引导形态学设计[J]. 《计算机应用》唯一官方网站, 2023, 43(2): 560-566. | 
| [9] | 柏财通, 崔翛龙, 郑会吉, 李爱. 基于自监督知识迁移的鲁棒性语音识别技术[J]. 《计算机应用》唯一官方网站, 2022, 42(10): 3217-3223. | 
| [10] | 高工, 杨红雨, 刘洪. 基于深度学习的三维点云人脸识别[J]. 计算机应用, 2021, 41(9): 2736-2740. | 
| [11] | 徐少伟, 秦品乐, 曾建朝, 赵致楷, 高媛. 基于注意力机制的两阶段纵膈淋巴结自动分割算法[J]. 计算机应用, 2021, 41(2): 556-562. | 
| [12] | 龚云鹏, 曾智勇, 叶锋. 基于灰度域特征增强的行人重识别方法[J]. 《计算机应用》唯一官方网站, 2021, 41(12): 3590-3595. | 
| [13] | 李华, 卢桂馥, 余沁茹. 基于干净数据的流形正则化非负矩阵分解[J]. 《计算机应用》唯一官方网站, 2021, 41(12): 3492-3498. | 
| [14] | 裴仪瑶, 郭会明, 张丹普, 陈文博. 基于定位不确定性的鲁棒3D目标检测方法[J]. 计算机应用, 2021, 41(10): 2979-2984. | 
| [15] | 王本杰, 农丽萍, 张文辉, 林基明, 王俊义. 基于Spider卷积的三维点云分类与分割网络[J]. 计算机应用, 2020, 40(6): 1607-1612. | 
| 阅读次数 | ||||||
| 全文 |  | |||||
| 摘要 |  | |||||