基于引导滤波和NSST的工业CT图像边缘检测

doi:10.3969/j.issn.0255-8297.2016.04.006

应用科学学报 ›› 2016, Vol. 34 ›› Issue (4): 405-416.doi: 10.3969/j.issn.0255-8297.2016.04.006

基于引导滤波和NSST的工业CT图像边缘检测

孟天亮¹, 吴一全^1,2,3, 吴诗婳¹

1. 南京航空航天大学电子信息工程学院, 南京 211106;
2. 华中科技大学数字制造装备与技术国家重点实验室, 武汉 430074;
3. 南昌航空大学江西省图像处理与模式识别重点实验室, 南昌 330063

收稿日期:2015-11-10 修回日期:2015-12-24 出版日期:2016-07-30 发布日期:2016-07-30
通信作者: 吴一全,教授,博导,研究方向:图像处理与分析、目标检测与识别、视觉检测与图像测量、视频编码与视频处理、生物特征识别与数字水印等,E-mail:nuaaimage@163.com E-mail:nuaaimage@163.com
基金资助:
数字制造装备与技术国家重点实验室开放基金(No.DMETKF2014010)；江西省图像处理与模式识别重点实验室项目基金(南昌航空大学)(2015)；江苏省高校优势学科建设工程项目基金资助

Edge Detection for Industrial CT Image Based on GuidedFiltering and Non-subsampled Shearlet Transform

MENG Tian-liang¹, WU Yi-quan^1,2,3, WU Shi-hua¹

1. College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China;
2. State Key Lab of Digital Manufacturing Equipment & Technology, Huazhong University ofScience and Technology, Wuhan 430074, China;
3. Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition, Nanchang Hangkong University, Nanchang 330063, China

Received:2015-11-10 Revised:2015-12-24 Online:2016-07-30 Published:2016-07-30

摘要/Abstract

摘要：

针对传统边缘检测算法不能准确检测有噪工业CT图像边缘的问题，提出一种鲁棒性好、能有效保持细小边缘的边缘检测算法.用引导滤波取代高斯滤波作为边缘检测的预处理，避免Canny算法对边缘的损坏，得到初步检测结果.在此基础上采用非下采样Shearlet变换分解图像，提取包含图像边缘细节信息的各尺度不同方向的高频系数.对每个方向的系数进行模极大值检测，并结合不同分解程度下边缘像素处的系数关系进一步调整模极大值，低频置零并通过反变换得到高频边缘检测结果.将初步检测结果与高频检测结果进行融合，经数学形态学处理得到最终边缘检测图像.实验对比了Canny算子以及近年来提出的同类边缘检测算法的结果，所提算法表现出更好的边缘保持特性，检测的完整性和准确性更高，品质因数比实验中的其他算法平均高出12%，边缘检测效果优越，为工业CT无损检测系统提供了更好的边缘检测方案.

关键词: 非下采样Shearlet变换, 引导滤波, 边缘检测, 工业CT图像

Abstract:

As existing image edge detection algorithms cannot accurately detect edges from noisy industrial computed tomography (CT) images, a robust edge detection algorithm capable of preserving fine edges is proposed. Instead of Gaussian filtering, guided filtering is used in image pre-processing for edge detection to avoid edge destruction of the Canny algorithm. Having obtained the preliminary detection result, non-subsampled shearlet transform (NSST) is applied for image decomposition. High-frequency coefficients of various scales in different directions containing edges and details are extracted. Modulus maximum detection is performed on the coefficients in each direction, and the maximum modulus values are adjusted depending on the property of coefficients of the edge points under different decomposing conditions. By setting the low-frequency coefficients to zero, inverse NSST is performed to get the high-frequency edge detection result. Finally, the preliminary result and the high-frequency detection result are combined. The final edge map is obtained with mathematical morphology. Experiments are performed and detection results are compared with those of classical Canny algorithm and several recent and similar edge detection algorithms. The proposed algorithm shows better edge preserving property, higher edge integrity and accuracy. An average increase of 12% of the figure of merit (FOM) indicator is achieved. The proposed edge detection algorithm provides a better edge detection scheme for industrial CT nondestructive testing systems.

Key words: non-subsampled Shearlet transform, guided filtering, edge detection, industrial CT image

中图分类号:

孟天亮, 吴一全, 吴诗婳. 基于引导滤波和NSST的工业CT图像边缘检测[J]. 应用科学学报, 2016, 34(4): 405-416.

MENG Tian-liang, WU Yi-quan, WU Shi-hua. Edge Detection for Industrial CT Image Based on GuidedFiltering and Non-subsampled Shearlet Transform[J]. Journal of Applied Sciences, 2016, 34(4): 405-416.

参考文献

[1] 朱敏, 卢洪义, 肖志斌. 固体发动机CT 图像的一种自动分割方法[J]. 固体火箭技术, 2008, 31(2): 201-204. Zhu M, Lu H Y, Xiao Z B. A kind of automatic segmentation method for solid motor CT image[J]. Journal of Solid Rocket Technology, 2008, 31(2): 201-204. (in Chinese)
[2] 邱钊, 朱庆生, 卢霞. 基于边缘信息的工业CT 图像分割法[J]. 计算机工程, 2004, 30(8): 159-161. Qiu Z, Zhu Q S, Lu X. Image segmentation of industrial CT based on edge information [J]. Computer Engineering, 2004, 30(8): 159-161. (in Chinese)
[3] 曾理, 蒲云, 马睿. 基于工业CT 的铁路货车铸件缺陷自动检测[J]. 中国铁道科学, 2009, 30(4): 76-80. Zeng L, Pu Y, Ma R. Automatic detection for the casting defects of railway freight car based on industrial CT [J]. China Railway Science, 2009, 30(4): 76-80. (in Chinese)
[4] Zeng L, An B B, Wan T Y. Crack edge extraction and measure of industrial CT image based on ridgelet transform [J]. Journal of Computational Information Systems, 2009, 5(5): 1393-1401.
[5] Ohtake Y, Suzuki H. Edge detection based multi-material interface extraction on industrial CT volumes [J]. Science China Information Sciences, 2013, 56(9): 092108.
[6] 霍冠英, 王敏, 程晓轩. 用于侧扫声纳图像边缘检测的改进Canny 算子[J]. 应用科学学报, 2011, 29(6): 613-618. Huo G Y, Wang M, Cheng X X. Edge detection for side-scan sonar images based on improved Canny operator [J]. Journal of Applied Sciences, 2011, 29(6): 613-618. (in Chinese)
[7] Ebrahim M J, Pourghassem H. A novel automatic synthetic segmentation algorithm based on mean shift clustering and canny edge detector for aerial and satellite images [J]. International Review on Computers and Software, 2012, 7(3): 1122-1129.
[8] 段振云, 罗晓凤, 杨旭. 基于各向异性扩散滤波的边缘提取算法[J]. 沈阳工业大学学报, 2014, 36(4): 421-425. Duan Z Y, Luo X F, Yang X. Edge extraction algorithm based on anisotropic diffusion filtering[J]. Journal of Shenyang University of Technology, 2014, 36(4): 421-425. (in Chinese)
[9] He K M, Sun J, Tang X O. Guided image filtering [J]. Lecture Notes in Computer Science, 2010, 6311: 1-14.
[10] Easley G, Labate D, Lim W. Sparse directional image representations using the discrete shearlet transform [J]. Applied and Computational Harmonic Analysis, 2008, 25(1): 25-46.
[11] 冯鑫, 王晓明, 党建武. 基于Shearlet 变换的红外与可见光图像融合[J]. 光电子·激光, 2013, 24(2): 384-390. Feng X, Wang X M, Dang J W. Fusion of infrared and visible images based on Shearlet transform [J]. Journal of Optoelectronics·Laser, 2013, 24(2): 384-390. (in Chinese)
[12] Karami A, Heylen R, Scheunders P. Band-specific shearlet-based hyperspectral image noise reduction [J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(9): 5054-5066.
[13] 肖创柏, 赵宏宇, 禹晶. 基于引导滤波的Retinex 快速夜间彩色图像增强技术[J]. 北京工业大学学报, 2013, 39(12): 1868-1873. Xiao C B, Zhao H Y, Yu J. Rapid retinex algorithm for night color image enhancement based on guided filtering [J]. Journal of Beijing University of Technology, 2013, 39(12): 1868-1873. (in Chinese)
[14] 郭德全, 杨红雨, 刘东权. 采用引导滤波的超声纹理补偿图像优化[J]. 计算机辅助设计与图形学学报, 2014, 26(1): 40-46. Guo D Q, Yang H Y, Liu D Q. Optimization of ultrasonic imaging in texture compensation using guided filter[J]. Journal of Computer-Aided Design & Computer Graphics, 2014, 26(1): 40-46. (in Chinese)
[15] 肖易寒, 席志红, 海涛. 基于非下采样contourlet 变换的图像边缘检测新方法[J]. 系统工程与电子技术, 2011, 33(7): 1668-1672. Xiao Y H, Xi Z H, Hai T. Image edge detection based on nonsubsampled contourlet transform[J]. Systems Engineering and Electronics, 2011, 33(7): 1668-1672. (in Chinese)
[16] Sheng Y, Demetrio L, Glenn R E. A shearlet approach to edge analysis and detection [J]. IEEE Transactions on Image Processing, 2009, 18(5): 929-941.
[17] Heath M D, Sarkar S. A robust visual method for assessing the relative performance of edge detection algorithms [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(12): 1338-1935.

基于引导滤波和NSST的工业CT图像边缘检测

Edge Detection for Industrial CT Image Based on GuidedFiltering and Non-subsampled Shearlet Transform

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

编辑推荐

Metrics

本文评价

[1]	刘松, 熊邦书. 基于计算机视觉的钢轨焊缝自动定位方法[J]. 应用科学学报, 2019, 37(6): 844-850.
[2]	吴一全1,2,3，李立1，陶飞翔1. 基于Shearlet域各向异性扩散和稀疏表示的图像去噪[J]. 应用科学学报, 2014, 32(3): 221-228.
[3]	舒鹏1，孙延奎1，田小林2. 眼前节光学相干层析图像中央角膜厚度自动测量[J]. 应用科学学报, 2012, 30(6): 619-623.
[4]	李刚1，万幼川1，李萌2. 应用启发式边缘生长的遥感影像分割[J]. 应用科学学报, 2012, 30(2): 173-180.
[5]	霍冠英1;2，王敏1;2，程晓轩1;2，李庆武1;2. 用于侧扫声纳图像边缘检测的改进Canny算子[J]. 应用科学学报, 2011, 29(6): 613-618.
[6]	朱方时龙兴. 图像块状瑕疵消除及边缘作用的分析[J]. 应用科学学报, 2009, 27(4): 353-358.
[7]	马煜，顾晓东，汪源源. 视觉机制模拟中的局部参数自组织映射方法[J]. 应用科学学报, 2009, 27(3): 247-252.
[8]	徐旦华, 鲍旭东, 舒华忠, 罗立民. 基于简化Mumford-Shah模型的活动轮廓边缘检测模型[J]. 应用科学学报, 2006, 24(4): 363-367.
[9]	丁兴号, 邓善熙, 杨永跃, 赵前程. 基于空间矩和Zernike矩的亚像素边缘检测[J]. 应用科学学报, 2004, 22(2): 191-194.