用于闪光图像消噪的改进BM3D方法

危才华; 唐志鹏; 景越峰; 管永红; 刘进

doi:10.11884/HPLPB202436.240217

用于闪光图像消噪的改进BM3D方法

doi: 10.11884/HPLPB202436.240217

中国工程物理研究院流体物理研究所，四川绵阳 621900

基金项目: 冲击波物理与爆轰波物理重点实验室基金项目(JCKY202212005)

详细信息

作者简介:
危才华，wch01592@qq.com

中图分类号: TN919
计量
- 文章访问数: 42
- HTML全文浏览量: 23
- PDF下载量: 5
- 被引次数: 0
出版历程
- 收稿日期: 2024-06-30
- 修回日期: 2024-09-15
- 录用日期: 2024-09-15
- 网络出版日期: 2024-09-24
- 刊出日期: 2024-10-15

Improved BM3D method for flash X-ray radiograph denoising

Institute of Fluid Physics, CAEP, Mianyang 621900, China

摘要

摘要: 高斯噪声是闪光图像中的主要噪声，将在密度反演等后续处理中被放大，严重影响密度重建及客体边界提取结果，因此，消高斯噪声是闪光图像消噪研究的重点内容。针对闪光照相图像噪声及照相客体轴旋转对称的特点，研究了基于三维块匹配滤波（Block Matching and 3D Filtering，BM3D）的闪光照相图像消噪算法，针对闪光照相图像中难以获得更高质量相似块的缺陷，在不破坏噪声独立性的情况下，通过对含噪退化图像进行旋转与镜像操作，增加了提供相似块的图像来源。同时，通过引入图像块的灰度变换，降低了原有相似性要求中的灰度值要求，提高了形状相似的要求，增加了获得高质量相似块的能力。图像的消噪结果表明，由于相似块的质量得到保证，用于闪光图像消噪的改进BM3D方法取得了更好的消噪效果。
- 闪光图像 /
- 高斯噪声 /
- BM3D /
- 消噪
Abstract: Gaussian noise is the main noise in flash X-ray radiographs, which will be magnified in the subsequent density inversion and other processing, and seriously affect the results of density reconstruction and object boundary extraction. Therefore, eliminating Gaussian noise is the key content of flash X-ray radiograph denoising research. According to the characteristics of image noise in flash X-ray radiographs and the rotational symmetry of object axes, this paper studies the denoising algorithm of flash X-ray radiographs based on BM3D (Block Matching and 3D Filtering). To overcome the defect that it is difficult to obtain high-quality similar blocks in flash X-ray radiographs, the image sources for providing similar blocks are increased by rotating and mirroring the noisy degraded images without destroying the noise independence. At the same time, by introducing grayscale transformation of image blocks, the grayscale value requirements in the original similarity requirements are reduced, the shape similarity requirements are improved, and the ability to obtain high-quality similar blocks is increased. Image denoising results show that the improved BM3D method in this paper achieves better denoising effect because the quality of similar blocks is guaranteed.
- flash X-ray radiograph /
- Gaussian noise /
- BM3D /
- denoising

HTML全文

图 1 FTO客体图像及其光程投影图

Figure 1. Object image of FTO and the projection image of light distance

下载: 全尺寸图片幻灯片

图 2 本文消噪方法第一步的相似块选取示例

Figure 2. Examples of similar block selection in the first step of denoising method in this paper

下载: 全尺寸图片幻灯片

图 3 本文消噪方法第二步的相似块选取示例

Figure 3. Examples of similar block selection in the second step of denoising method in this paper

下载: 全尺寸图片幻灯片

图 4 本文消噪方法第二步中的相似块与参考块灰度比较

Figure 4. Gray scale comparison between similar block and reference block in the second step of denoising method in this paper

下载: 全尺寸图片幻灯片

图 5 FTO仿真图像消噪结果

Figure 5. Denoising results of FTO simulation image

下载: 全尺寸图片幻灯片

图 6 过中心的水平剖线对比情况

Figure 6. Comparison of horizontal cross-sections over the center

下载: 全尺寸图片幻灯片

图 7 实验图像消噪结果

Figure 7. Denoising results of experimental image

下载: 全尺寸图片幻灯片

图 8 消噪结果局部细节

Figure 8. Local details of denoising results

下载: 全尺寸图片幻灯片

表 1 FTO仿真图像消噪方法评价结果

Table 1. Evaluation results of denoising methods for FTO simulation image

filtering method	MSE/10⁻³	SNR	PSNR
mean filtering	84.86	37.99	39.43
Gaussian filtering	83.74	38.49	39.56
P-M anisotropic diffusion filtering	81.42	39.59	39.75
nonlocal mean filtering	85.98	37.49	39.21
BM3D	63.71	50.60	41.88
method in this paper	55.66	57.91	43.03

下载: 导出CSV

参考文献(15)

[1]	Aufderheide III M B, Martz H E Jr, Slone D M, et al. Concluding report: quantitative tomography simulations and reconstruction algorithms[R]. UCRL-ID-146938, 2002: 1-19.
[2]	Clark D A, Espinoza C J. Proton radiography[R]. Physics Division Progress Report, 1999-2000: 156-168.
[3]	危才华, 景越峰, 张小琳, 等. 基于极大似然模型和期望最大化算法的闪光图像重建[J]. 强激光与粒子束, 2016, 28:054003 doi: 10.11884/HPLPB201628.054003 Wei Caihua, Jing Yuefeng, Zhang Xiaolin, et al. Image reconstruction algorithm based on maximum likelihood-expectation maximum for radiography[J]. High Power Laser and Particle Beams, 2016, 28: 054003 doi: 10.11884/HPLPB201628.054003
[4]	景越峰, 管永红, 刘军. 基于改进开关中值滤波的多孔网栅图像脉冲噪声消除[J]. 强激光与粒子束, 2015, 27:084006 doi: 10.11884/HPLPB201527.084006 Jing Yuefeng, Guan Yonghong, Liu Jun. Removal of impulse noise of anti-scatter grided images with modified switching median filters[J]. High Power Laser and Particle Beams, 2015, 27: 084006 doi: 10.11884/HPLPB201527.084006
[5]	Meng Yizhen, Zhang Jun. A novel gray image denoising method using convolutional neural network[J]. IEEE Access, 2022, 10: 49657-49676. doi: 10.1109/ACCESS.2022.3169131
[6]	Holla K S, Park N, Lee B. EFID: edge-focused image denoising using a convolutional neural network[J]. IEEE Access, 2023, 11: 9613-9626. doi: 10.1109/ACCESS.2023.3239835
[7]	谷学静, 杨宝上, 刘秋月. 基于高斯滤波和AKAZE-LATCH的图像匹配算法[J]. 半导体光电, 2023, 44(4):639-644 Gu Xuejing, Yang Baoshang, Liu Qiuyue. Image matching algorithm based on Gaussian filtering and AKAZE-LATCH[J]. Semiconductor Optoelectronics, 2023, 44(4): 639-644
[8]	李文娟, 陈军, 张永刚, 等. 基于改进加权均值滤波的医学影像图像除噪研究[J]. 辽宁大学学报(自然科学版), 2022, 49(1):30-35 Li Wenjuan, Chen Jun, Zhang Yonggang, et al. Research of medical image denoising based on improved weighted mean filtering[J]. Journal of Liaoning University (Natural Sciences Edition), 2022, 49(1): 30-35
[9]	肖丹, 黄玉清. 改进的各向异性扩散图像去噪算法[J]. 自动化仪表, 2017, 38(7):1-3 Xiao Dan, Huang Yuqing. Improved anisotropic diffusion image denoising algorithm[J]. Process Automation Instrumentation, 2017, 38(7): 1-3
[10]	Buades A, Coll B, Morel J M. A review of image denoising algorithms, with a new one[J]. Multiscale Modeling & Simulation, 2005, 4(2): 490-530.
[11]	Dabov K, Foi A, Katkovnik V, et al. Image denoising by sparse 3-D transform-domain collaborative filtering[J]. IEEE Transactions on Image Processing, 2007, 16(8): 2080-2095. doi: 10.1109/TIP.2007.901238
[12]	俞汪涛, 许鹏, 鲍杰, 等. 基于BM3D算法的快中子图像降噪方法[J]. 核电子学与探测技术, 2023, 43(2):369-375 doi: 10.3969/j.issn.0258-0934.2023.02.025 Yu Wangtao, Xu Peng, Bao Jie, et al. Fast neutron image denoising method based on BM3D algorithm[J]. Nuclear Electronics & Detection Technology, 2023, 43(2): 369-375 doi: 10.3969/j.issn.0258-0934.2023.02.025
[13]	唐艳, 潘伟, 张利, 等. 基于BM3D去噪算法在天文图像中的应用[J]. 智能计算机与应用, 2022, 12(9):193-197 doi: 10.3969/j.issn.2095-2163.2022.09.035 Tang Yan, Pan Wei, Zhang Li, et al. Application of BM3D denoising algorithm in astronomical images[J]. Intelligent Computer and Applications, 2022, 12(9): 193-197 doi: 10.3969/j.issn.2095-2163.2022.09.035
[14]	张小琳, 景越峰, 刘军. 基于Facet模型的闪光图像边缘检测[J]. 强激光与粒子束, 2010, 22(7):1640-1644 doi: 10.3788/HPLPB20102207.1640 Zhang Xiaolin, Jing Yuefeng, Liu Jun. Edge detection method based on Facet model for flash X-ray radiographs[J]. High Power Laser and Particle Beams, 2010, 22(7): 1640-1644 doi: 10.3788/HPLPB20102207.1640
[15]	Huynh-Thu Q, Ghanbari M. Scope of validity of PSNR in image/Video quality assessment[J]. Electronics Letters, 2008, 44(13): 800-801. doi: 10.1049/el:20080522